1. These transformations in the state may generally be understood in terms of a shift from a providing to an enabling role, with responsibility for provision remaining with the purchasing authority, whilst service delivery is contracted out to independent bodies in the private sector, or to quasi-independent units within the organisation. See P Hoggett Modernisation, Political Strategy and the Welfare State (University of Bristol: School for Advanced Urban Studies, 1990); Harrison, A. (ed) From Hierarchy to Contract (Oxford: Policy Journals, 1993).Google Scholar

2. The language of contract is playing an ideological/political role in the radical Conservative assault on the concept of the welfare state as public service provider, and its transformation into a purchasing business run on private sector lines. The present paper focuses on the direct economic functions of contract.

3. The main emphasis will be on local government, drawing on provisional results of ongoing research into contractual and non-contractual relations in the public sector (hereafter CNCRPS) involving three local authorities and other public bodies in the north of England.

4. From the classical legal perspective: ‘Formal rules of law are needed to facilitate the making and enforcement of promises, because the parties cannot themselves make arrangements which adequately guard against the risk of a promise being broken,’ per D Harris and C Veljanovski, in T Daintith and Teubner, G. (eds) Contract and Orgunisation: Legal Analysis in the Light of Economic and Social Theory (Berlin: Walter de Gruyter, 1986) p 110.Google Scholar See also D Harris ‘Incentives to Perform, or Break Contracts’ (1992) 45 Current Legal Problems 30.

5. For an account of the Weberian ideal-type methodology informing the theoretical model, see: P Vincent-Jones ‘Contract and Business Transactions: A Socio-Legal Analysis’ (1989) 16 Jnl of Law and Society 166.

6. The essence of contractual and near-contractual governance is the use of legal or quasi-legal processes in planning and adjusting exchanges. According to this conception, it does not matter whether the obligations of the parties are considered to arise from ‘agreement’, or from reliance or expectation, or from some other source. Contractual relations may therefore encompass forms of liability described in legal doctrine as ‘quasi-contractual’, provided that the criteria identified below are satisfied. It follows that the term ‘non-contractual’ is defined more broadly here (as in the existing empirical literature) than in conventional legal usage.

7. S Macaulay, Non-Contractual Relations in Business: A Preliminary Study' (1963) 27 Am Sociological Rev 55, p 56. In reality, any of these elements is sufficient for a finding of contractual relations (very few business exchanges result in dispute and the use or threat of legal sanctions - this does not render the relationship ‘non-contractual’ if it has been created contractually).

8. J Macneil ‘The Many Futures of Contracts’ (1974) 47 Southern California Law Review 691, p 800 n 310 (quoting the Oxford English Dictionary definition) and p 697 respectively. See also Macneil, ‘Contracts: Adjustment of Long-Term Economic Relations under Classical, Neo-Classical, and Relational Contract Law’ (1978) 72 Northwestern Univ Law Review 854, p. 863.

9. This assumes arbitration of a ‘rights’ dispute, as distinct from an ‘interests’ dispute, in which the purpose of the hearing may be not to apply legal norms, but to till gaps in the original agreement. On this distinction, see Macneil, (1978) op cit n 8, pp 866–867. To the extent that arbitration procedures have evolved in distinct ways within particular industries, and may be said to involve extra-legal norms, customs and conventions, reference to arbitration cannot indicate ‘contractual’ relations. See: J Flood and A Caiger ‘Lawyers and Arbitration: The Juridification of Construction Disputes’ (1993) 56 MLR 412; R Ferguson ‘The Adjudication of Commercial Disputes and the Legal System in Modem England’ (1980) 7 Brit Jnl of Law and Society 141.

10. See J Adams and R Brownsword Understanding Contract Law (London: Fontana, 1994) pp 3–15.

11. See Macaulay, op cit n 7; H Beale and T Dugdale ‘Contracts Between Businessmen: Planning and the use of Contractual Remedies’ (1975) 2 Brit Jnl of Law and Society 45.

12. Macneil, op cit n 8 (1974), p 715. For an extensive review and bibliography relating to extra-legal normative constraints, see Cotterrell, R. The Sociology of Law: An Introduction (London: Buttenvorths, 1992) pp 27–28 and pp 317–321.Google Scholar

13. Customary norms may be said to differ from their conventional counterparts in being sanctioned internally by conscience and self-discipline, rather than externally by informal or formal social disapproval. In practice, the weakest end of the normative-conventional spectrum is likely to shade into the normative-customary, or even into the completely non-normative categories. See Vincent-Jones, op cit n 5.

14. Whilst non-contractual norms may also be formally applied in third-party adjudication, for example by private tribunals administering trade association codes of practice or club rules, near-contractual private ordering differs in that it involves obligations deriving from planned exchange between two parties.

15. The importance of this economic approach is well recognised in socio-legal studies of business transacting - see: Britt-Mari Blegvad (1990) ‘Commercial Relations, Contract, and Litigation in Denmark: A Discussion of Macaulay's Theories’ 24 Law and SOC Review 397; T Daintith, ‘The Design and Performance of Long-Term Contracts’ in Daintith, T. and Teubner, G (eds) Contract and Organisation: Legal Analysis in the Light of Social and Economic Theory (Berlin: Walter de Gruyter, 1986)CrossRefGoogle Scholar. For a critical analysis of the basis of the economic approach, see D Campbell and D Harris, ‘Flexibility in Long-Term Contractual Relationships: The Role of Co-operation’ (1993) 20 Jnl of Law and Society 166.

16. The classical approach is based on rational choice by economic actors motivated by utility maximization - see Burrows, P and Veljanovski, C. (eds) The Economic Approach to Law (London: Butterworths, 1981) Introduction.Google Scholar

17. Williamson ‘Contract Analysis: The Transaction Cost Approach’ in Burrows and Veljanovski (eds), op cit n 16, p 39. Williamson states: ‘A wider (than market) conception of economic organization is contemplated by the transaction cost approach. Although market exchange remains important, it is only one of several governance structures’ (p 41). The recognition that ‘the importance of legal rules … varies with the nature of the transaction’ (ibid) does not lead to an adequate analysis of the specifically legal, as opposed to economic, aspects of governance. It will be argued that, for present purposes, this omission is a fundamental weakness, and that the transaction cost literature needs to be incorporated within a broader inter-disciplinary approach, with contractual governance more to the fore.

18. Williamson ‘Transaction Cost Economics: The Governance of Contractual Relations’ (1979) 22 Jnl Law and Economics 233, at p 239. Williamson also refers to governance structures as ‘institutional arrangements’ - see M Aoki, B Gustafsson & 0 Williamson (eds) The Finn as a Nexus of Treaties (London: Sage, 1990) p 9.

19. Opcitn 18 (1979). pp 234–235. A mode of governance intermediate between markets and internal organization is described as obligational contracting, and in a later article as ‘obligational market contracting’, op cit n 17, p 47.

20. Ibid. This aspect of transacting is more usually referred to in terms of limited or asymmetric information.

21. Ibid. Elsewhere Williamson states: ‘Specifically, economic agents are permitted to disclose information in a selective and distorted manner. Calculated efforts to mislead, disguise, obfuscate and confuse are thus admitted. This self-interest seeking attribute is variously described as opportunism, moral hazard and agency’ op cit n 18 (1990) p 12.

22. Ibid.

23. Op cit n 18 (1979) p 246.

24. ‘Items that are unspecialized among users pose few hazards, since buyers in these circumstances can easily turn to alternative sources, and suppliers can sell output intended for one order to other buyers without difficulty. Non-marketability problems arise where the specific identity of the parties has important cost-bearing consequences. Transactions of this kind … (are)… referred to as idiosyncratic’ ibid pp 239–240. For an interesting application of Williamson on this point, see T Palay ‘Comparative Institutional Economics: The Governance of Rail Freight Contracting’ (1984) 13 Jnl of Legal Studies 265. (Idiosyncratic investments may be made in human as well as physical capital, through training and staff development, etc).

25. Op cit n 18 (1979) p 248.

26. Also referred to under the heading ‘Unified Governance: Internal Organization’, ibid p252.

27. Ibid. ‘The advantage of vertical integration is that adaptations can be made in a sequential way without the need to consult, complete or revise interfirm agreements’ at p 253.

28. bid p 250.

29. Also referred to as ‘Bilateral Governance: Obligational Contracting’, ibid p 250.

30. Ibid pp 250–252.

31. It is not clear whether networks and treaties, which have recently received much attention (see, for example, Aoki et al (4s) op cit n 18). are forms of economic governance, or frameworks of normative regulation.

32. See Macneil, op cit n 8 (1978) pp 862–865.

33. Williamson, op cit n 18 (1979) p 237.

34. In practice in the real world, such a transaction might also be expected to have ‘non-contractual’ characteristics - see below.

35. Macneil, op cit n 8 (1978) pp 865–886.

36. Williamson, op cit n 18 (1979) pp 249–250.

37. Macneil, op cit n 8 (1978) p 870.

38. Ibid p 866.

39. Op cit n 18 (1979) p 248.

40. Ibid p 250.

41. Ibid.

42. Macneil, op cit n 8 (1978) at p 898. ‘In on-going contractual relations we find such broad norms as distributive justice, liberty, human dignity, social equality and inequality, and procedural justice, to mention some of the more vital… Changes in such … relations must accord with norms established respecting these matters, just as they do the more traditional contract norms.’ ibid.

43. For an in-depth critique of Macneil's position from a different perspective, see D Campbell ‘The Social Theory of Relational Contract: Macneil as the Modem Proudhon’ (1990) 18 Int Jnl Sociology of Law 75.

44. The dynamic and processual aspect of transaction governance has been stressed by Blegvad, op cit n 15.

45. For example: ‘More commonly it is the promise which is the gap-tiller while the great sea of custom … forms the main structure of contract’, op cit n 8 (1974) p 731. On the present analysis, custom is a non-contractual form of order, analytically quite distinct from contract.

46. Macneil, op cit n 8 (1978) p 857. See also bid (1974) pp 720–721.

47. Even on the New Jersey Turnpike - see Macneil op cit n 8 (1978) p 857 n 10; cf Williamson op cit n 18 (1979) p 247.

48. Whereas Macneil locates discrete transactions and classical contract law together at one end of his continuum, as illustrated in the petrol purchase example, (1978) pp 862–865. From the present perspective, the mere fact of discreteness does not lead to the conclusion that this exchange must have been contractual in the sense required by the classical model; indeed it could be described as non-contractual - an immediate exchange in which uncertainties and risks are minimal, in which planning is not necessary, and in which disputes are highly unlikely to result. The motorist would most likely be very surprised to learn that she had been engaged in ‘classical’ contracting. The classical model is a legal construction, superimposed on socio-economic relations. Unless its rules are called upon by one of the parties in an exchange, it may be of no practical significance whatsoever. ‘Discreteness’, in this important sense, is produced by the classical models of lawyers and economists - it is not empirically derived from socio-economic relations.

49. Op cit n 7.

50. See notes 11 and 15, above.

51. This is so despite some limited evidence suggesting an increase in contract disputing and litigation in recent years - see M Galanter ‘The Life and Times of the Big Six: or, The Federal Courts since the Good Old Days’ (1988) 6 Wisconsin Law Review 921; P Vincent-Jones ‘Contract Litigation in England and Wales 1975–1991: A Transformation in Business Disputing?’ (1993) 12 Civil Justice Quarterly 337.

52. The economic variables (asset specificity, uncertainty, bounded rationality, and opportunism) have been condensed into a single opposition ‘high-risk - low risk’ for presentational purposes (see Figure 2).

53. Turpin, C. Government Procurement and Contracts (Harlow: Longman, 1989) p 102 Google Scholar; ‘The belief is held that the best settlements are achieved by discussion between those most familiar with the contract work, often of an extremely technical nature, and with the conventions and understandings forming the background to the contract’ (at p 222).

54. Ibid p 225. Requirement for such reference was originally included within standard form contracts. Although arbitration clauses are now commonly included instead, in practice these do not preclude the more informal alternative.

55. Ibid p 190. Performance bonds have drawbacks however. They may indirectly increase the cost of the tender, and can be a source of conflict with third-party guarantors. The retention of final payment of the contract price, now commonly 25% in defence contracts, against satisfactory completion is a more reliable safeguard - see p 205.

56. See K Ascher The Politics of Privatisation: Contracting Out of Public Services (Macmillan, 1987) p 146.

57. Local Government Management Board (LGMB) CCT Information Service survey report, (No 6: November 1992). The survey covered 3 887 contracts in England and Wales. The LGMB national database is composed from answers to questionnaires sent to all authorities in England and Wales in January and July each year.

58. Turpin. op cit n 53, p 222.

59. Circular HC (83) 18.

60. See C Whitbread and N Hooper, in Harrison (ed), op cit n 1.

61. Sports and leisure management were included by Order, under LGA 1988 s 2(3), in November 1989. Following proposals in the White Paper ‘Competing for Quality’ in 1991 (Cm 1730). the government has announced a timetable for the compulsory tendering of white-collar services, including law, architecture, engineering, property management, information technology, housing management, finance and personnel.

62. Private sector penetration in service provision under the LGA 1988 was approximately 14% in the first tendering round - see Audit Commission ‘Realising the Benefits of Competition: The Client Role for Contracted Services’ (London: HMSO, 1993). para 16.

63. This brief review will emphasise local government because available data is more complete in this area.

64. Embodied in the Rayner reforms (1973, and the ‘value for money’ and ‘financial management’ initiatives of the late 1970s and early 1980s.

65. For a brief account and further references, see A Barron and C Scott ‘The Citizens Charter Programme’ (1992) 55 MLR 538, at p 539; Harrison, op cit n 1.

66. See for example: R Laughlin ‘Accounting Control and Controlling Accounting: The Battle for the Public Sector’ (1991) Sheffield University Management School Discussion Paper No 92–29. The following brief description is based on Laughlin's account.

67. According to R Laughlin and J Broadbent, the FMI involved ‘central tenets of devolution of financial responsibility with cash-limited budgets to smaller front-line service units and the increasing concentration on measurable performance in an accountability framework,’ in ‘Accounting and Law: Partners in the Juridification of the Public Sector in the UK?’. Second European Management Control Symposium, HEC School of Mumgernent (July 1992) at p 16.

68. bid p 12.

69. See Ascher, op cit n 56, p 144. See also Walsh, K. Compulsory Competitive Tendering for Local Authority Services: Initial Experiences (London: HMSO, 199l)Google Scholar, at para 5.7.

70. Ibid ch 7 and App B. Many authorities in this study had no explicit process of inspection before CCT, whereas monitoring now involves assessment by reference to written standards, with user panels, surveys, and consultative committees, with some authorities employing specialist inspectors - see paras 7.13 and 7. 14. According to the LGMB survey (op cit n 57), daily or weekly supervision of contractors and random spot checks are common across all areas, but especially in refuse collection, ground maintenance, and all types of cleaning contract. The use of computers in monitoring is developing slowly, with about 14% of authorities using information technology in the first round of tendering in this way.

71. Systems for payment deduction for unsatisfactory service are rigorously and exactly specified - see Walsh, op cit n 69, para 7.17. The LGMB survey (op cit n 57) contains extensive data on default and penalty clauses written into contracts, and on actual defaults and penalties - see sections F1 and F2 respectively throughout.

72. See Macneil, op cit n 8 (1974): ‘Services are inherently relational, and inherently less subject to exact prior planning’ at p 763 n 209.

73. Ibid p 729.

74. For example in local government, the lack of adequate information regarding the nature and extent of tasks to be performed, and the absence of proper information systems - see Walsh, op cit n 69, at para 4.21.

75. Ibid para 5.16.

76. See Walsh, op cit n 69: ‘Staff skills were lacking in many areas, particularly on the client side for specification, and to a lesser extent contract preparation,’ at para 4.22.

77. On the adequacy of time for preparation under the LGA 1988, see ibid para 4.15.

78. See Whitbread and Hooper, op cit n 60.

79. LGMB survey, op cit n 57, p 10. The survey revealed 66 terminations (29 of which involved contracts held by DSOs - see below) in the period studied, including 16 in grounds maintenance, 21 in building cleaning, 6 in catering, 8 in vehicle maintenance, and 7 in refuse collection. Difficulties or problems were reported in refuse collection (28% of contracts), general cleaning (21%), ground maintenance (16%) and building cleaning (17%). Defaults leading to financial penalties (liquidated damages or the withholding of payments) were reported for 10.4% of private sector contracts (4% being terminated) and 5.4% of DSO contracts (1 % being terminated). There is little evidence to suggest that public sector disputes have required resort to the ultimate guarantee - binding adjudication or arbitration. Most local authority standard forms now contain arbitration clauses, but where this form of external involvement does occur, it will not necessarily be ‘rights’ arbitration.

80. K Walsh, in J Stewart and G Stoker (eds) Competition and Service in Local Government (Macmillan 1989). Although made in the context of local government, the points are valid for the public sector generally. For an application of the Williamson variables in the private sector, see T Daintith, op cit n 15.

81. Ibid p 48. The suggested scores are medium for opportunism, and high for bounded rationality and asset specificity.

82. Ibid p 47.

83. For example. security guarding, aircraft servicing, air traffic control, and cleaning, laundering and catering - see K Hartley, in Harrison (ed), op cit n 1, p 58. See also C Turpin, op cit n 53.

84. See M Starks Nor for Profit, Not For Sale (Policy Journals, 1991) pp 115–134.

85. Ascher, op cit n 56, p 43.

86. Ibid p 23.

87. Ibid p 35.

88. 34 hospitals were involved in such contracts by 1965- see ibid p 24.

89. In some construction projects, for example hospital building, legal proceedings for defective work are fairly common - see Turpin, op cit n 53, p 222. Even though ‘the cases commonly end in an agreed settlement rather than being carried through to judgment’ (ibid), the result may have been achieved by reference to legal rights and threat of legal sanctions.

90. Ibid p 223. Some local authorities have expressed an interest in ADR, two having joined the Centre for Dispute Resolution (CEDR). What is interesting is not the relative lack of interest, but the fact that these authorities should anticipate problems with commercial contracts that cannot be resolved bilaterally. See J Wills ‘The Mission to Keep People out of Court’Local Government Chronicle 29 January 1993. Where external mediation does occur, it is likely to involve the application of market rationality and noncontractual norms, with some role for neo-classical contract law.

91. Competitiveness cannot be assumed in CCT transactions because of the strong possibility of cartels and monopolies, especially in the longer term when weaker competitors may have been driven out of the market.

92. Where there is competitive bidding, the winning bidder may be a victim of a phenomenon described by economists as the ‘winner's curse’ - see E Capen, R Clapp and W Campbell ‘Competitive Bidding in High Risk Situations’ (1971) 23 Journal of Petroleum Technology 641. This hypothesis states that in any auction-type setting, the party who outbids other competitors and wins the contest is likely to have over-estimated the value of the object. In the context of CCT, transaction problems may be expected to result if the winning contractor is ‘cursed’ by having bid too cheaply, at a price for which the service cannot realistically be provided. The contractor's incentive to opportunism is likely thereby to have been increased.

93. See R Crofts ‘Trimming the Extra Pounds Off Contracts,’Public Finance and Accountancy, June 1992.

94. bid at p 17. ‘The main concern of one company was the preservation of the company image as providers of community leisure.’ An important factor in this resolution was the economic balance of power. This lay clearly with the council, because of the possibility of future contract awards, redressing its apparently weak legal position.

95. For example, where there are likely to be economic gains from changing to a new supplier, local authorities may be expected to take this option. In the rapidly expanding market for facilities management, for example, one local authority is considering exercising an opt-out clause (itself an example of contractual planning) and scrapping an existing contract after only six months of a three-year term, despite satisfaction with the contract, because a cheaper supplier has entered the market - see D Jellinek ‘Six Months is a Long Time in Facilities Management’Local Government Chronicle 11 September 1992.

96. Quasi-markets are characterised by three related elements: the fragmentation and separation of purchaser and provider interests; the devolution of decision-making responsibility and accounting autonomy to semi-independent cost centres; and real competition between internal providers, or between internal providers and potential external private sector bidders. In relation to local government, see P Vincent-Jones ‘The Limits of Near-contractual Governance: Local Authority Internal Trading under CCT’ (1994) 21 Jnl of Law and Society 214.

97. Harrison, op cit n 1, p 8.

98. This section will concentrate on experience within local authorities. In the LGW survey (op cit n 57, at p 7) DSO success in winning CCT contracts varied between 54% in building cleaning, and 91% in catering (education and welfare). DSOs had a 100% success record in 171 authorities, including 20 Metropolitan Districts and 9 London Boroughs.

99. Most local authorities are using and adapting core contract documentation supplied by local authority associations, or are using existing JCT or ICE standard terms, even for internal exchanges - see Walsh, op cit n 69, para 5. 7. Provisional evidence from the CNCRPS study indicates closer attention to terms within standard forms than suggested in the existing private sector research.

100. LGMB survey generally, op cit n 57. Overall, the survey suggests that DSOs are treated little differently from external contractors as regards contract management.

101. See Walsh, in Harrison (ed) op cit n 1, p 114.

102. Ibid p 115.

103. See K Pinder ‘Service Level Agreements and the Internal Market in Local Government,’ in J Chandler and J Darwin (eds) The Waves of Change: Strategic Management in the Public Services (Sheftield Business School, 1993).

104. Walsh, in Hanison (ed), op cit n 1, p 116. See ibid, op cit n 69: ‘The purpose is that support service charges become a controllable element within the budget of the DSO,’ at para 10. 16.

105. See LGMB survey generally, op cit n 57.

106. Some proportion needs to be maintained, however. According to Walsh, op cit n 69: ‘Differences could be significant, but few contractual difficulties have in practice been taken beyond the stage of debate in the various meetings between client and contractor,’ at para 9. 5. 10′7. Grievance management procedures have always existed within local authorities in some form. From the 1980s, however, there have been more internal disputes, the dispute-resolution process has become more formal, and the basis for settlement more related to ‘contractual’ documents.

108. As within one authority in CNCRPS research, op cit n 3, discussed below.

109. See P Vincent-Jones and A Harries ‘Near-Contractual Relations in Internal Trading under the Local Government Act 1988: A Comparative Study of Refuse Collection and Buildings Cleaning’, paper presented to SLSA Annual Conference, Nottingharn University, March 1994. For a more developed theoretical analysis in relation to local government, see Vincent-Jones, op cit n 96.

110. See D Hughes ‘The Reorganisation of the NHS: The Rhetoric and the Reality of the Internal Market’ (1991) 54 MLR 88; Harrison, in Harrison (ed) op cit n 1.

111. Under s 4 (1) of the National Health Service and Community Care Act 1990, an NHS contract is defined as: ‘an arrangement under which one health service body (the ‘acquirer’) arranges for the provision to it by another health service body (the ‘provider’) of goods and services which it reasonably requires for the purposes of its functions.'

112. Ibid s 4(3).

113. The three types of contractual arrangement proposed by the government. See Starks, op cit n 84, p 130; I Harden The Contracting State (Buckingham: Open UP, 1992) p 16.

114. ‘The contractual process should cover all the aspects of the services to be provided and the expectation of both parties. A fundamental merit of the contract is that it will provide a vehicle to identify much more of what health service providers are expected to deliver.’ See Contracts for Health Services: Operational Principles (HMSO, 1989), quoted in Starks, op cit n 84, p 131.

115. S4(4).

116. S4(5).

117. Hughes, op cit n 110, p 92.

118. ‘The presumption in determining a dispute is likely to be that the outcome will give effect to the agreement which was originally reached, rather than a new agreement which the parties should have reached’, NHS Management Executive Guidance, quoted in Harden, op cit n 113, p 43.

119. Hughes, op cit n 110, at p 90.

120. Ibid.

121. ‘Improving Management in Government: The Next Steps’ (London: HMSO, 1988).

122. Next Steps Briefing Note, June 1993. There is considerable variety in size and function, from the Social Security Benefits Agency (64,215 staff) to Wilton Park Conference Centre (25 staff).

123. See D Goldsworthy Setting Up Next Steps (London: HMSO, 1991).

124. Ibid p 7.

125. In the Annual Review for 1992, for example, around three-quarters of targets were met, the same as for the previous year, but there is no explicit consideration of the consequences of under-performance - see ‘The Next Steps Agencies: Review 1992’ Cm 2111 (London: HMSO, 1992).

126. Disputes that do arise are likely to require examination of enabling legislation, rather than the Framework Document -see E Mellon in Harrison (ed) op cit n 1, p 20.

127. Central Office of Information, Chessington Computer Centre, Companies House, Defence Research Agency, Fire Service College, Hh4S0, HM Land Registry, Medicines Control Agency, Patents Office, Royal Mint, Buying Agency, and the Vehicle Inspectorate - see Next Steps Briefing Note, op cit n 122, para 23.

128. Ibid. Goldsworthy (op cit n 123) states: ‘A trading fund … provides a financial framework covering operating costs and receipts, capital expenditure, borrowing and net cash flow. It is required to break even, taking one year with another, and to make a return on capital, and has powers to borrow to meet capital expenditure and working capital requirements, and to establish reserves out of surpluses’, at p 30.

129. It may be argued that SLAs do not exist in a competitive context either, and that the relationship here is also hierarchical. On the other hand, evidence of arbitration and adjudication over SLAs, with the agreement forming the basis of the settlement, does suggest near-contractual governance - these elements are not present in framework agreements.

130. Briefing Note, op cit n 122, para 38.

131. Although the markets for some goods and services may have become more risky and volatile, with more new entrants and high attrition amongst established players, resulting in a decline in the effectiveness of non-contractual ordering, the Macaulay thesis remains fundamentally valid.