The Partnerships Victoria guidance material on risks is detailed and constitutes a professional 'how to' manual for staff. Most of the bulk of the risk allocation and contractual issues documented present detailed guidance and discussion on each of the ten categories of risks previously defined; key contractual issues of particular relevance to Victoria are also included. The material notes that 'to be legally effective, the agreed risk allocation must be reflected in the contractual provisions and mechanisms'. 322 Aligning the payment mechanism with the agreed risk allocation and achievement of government objectives is of particular importance, and the material offers general guidance on this complex exercise, highlighting the need for contractual provisions to be constantly scrutinised.323
The guidelines also present a sophisticated overview of ideas underpinning the PSC and detail its major financial components. Of particular interest is the inclusion of some advanced probability evaluation techniques together with simpler probability based techniques to determine the value of risks.324
Providing examples is also helpful for practitioners and the considerable guidance provided on contract management concepts is useful. The Committee considers that the Partnerships Victoria guidelines appear to be comprehensive and rational from a commercial perspective.
The guidelines, however, do not appear to be broad enough to be regarded as a firm basis on which the government would make a final decision on the use of private investment in public infrastructure, from either a public policy or a governance perspective. Several types of risks, for example, political risk and environmental risk are not adequately covered and essentially lie outside the commercial assumptions of the guidelines. The treatment of some risks is subject to extensive controversy and is not acknowledged and may well be handled inadequately at present. Further, the guidelines do not recognise the inherent optimism bias usually accompanying infrastructure proposals.
Some real world risks exist that are not typically the subject of negotiation between government and private infrastructure sponsors. The commentator Arndt noted that 'the concept of political risk does not sit comfortably with the legal concept of risk allocation', and 'it is not a negotiated risk' when considered directly in concession deeds for projects.325 The critical political risks here are outside the contractual negotiated domain and typically go unrecognised. One academic has argued that despite the concept of risk spreading:326
The government has to stand behind these projects because it cannot allow them to fail. The cynical view is that companies know full well that it will not be their job to bail out important services. Ministers will always ride to the rescue. Under such conditions, the taxpayer meets all the costs and gets none of the benefits.
A further systemic risk not specifically identified in the framework is given by the commentators de Bettignies and Ross, that governments should always provide systems of checks and balances:327
Granting too much authority to a government PPP office to implement projects without adequate review via effective departments (and later by public auditors) risks having the private industry capture the PPP process through its repeated interactions through the government PPP office. Both the private sector players and the PPP office will see benefits in maximising PPP activity whether all projects are in the public interest or not.
Dr Duffield reinforced this point when he informed the Committee that:328
I think when we do PPPs, take government's perspective, there is a commercial responsibility to the state; there is servicing of the community interest; there is regulation long term, and there is an aspect which I would call stewardship, which is long-term management of the state's asset. I have a personal view [that] … to expect one individual or group of individuals to be all things without separating them out is a tough call.
This is particularly so when, as the Committee was informed by one witness:329
The problem is with the way we structure our finance and so on; most people have an incentive to make the deal happen.
A further risk not explicitly covered in the guidelines, but inherent within their document, is that of service purchasing. The underlying assumption that purchasing services through contract over a long period involves less risk than purchasing infrastructure (based on a detailed initial design in the short term) remains just that, - an assumption. The 'black and white' specification of services and signed contracts can differ from experiences throughout the contract period. Because initial expectations of large savings from contracts over the long term operation of the project can differ from experience, large risks go unrecognised. This significant issue is demonstrated by the public transport franchising experience of the Victorian Government, which saw large initial expected savings at the time of contract signing evaporate.
Optimism bias refers to the 'demonstrated, systematic, tendency for project appraisers to be overly optimistic.330 This phenomenon is not acknowledged in the guidelines. Now recognised as a worldwide phenomenon, optimism bias 'affects both the private and public sectors'. It is the difference between the initial expected costs and benefits of a project and the final actual costs or 'out turn'.331 This applies to all infrastructure projects regardless of method of delivery.
Mott MacDonald reviewed 50 projects over 20 years in the United Kingdom, and found high levels of optimism bias. For traditionally procured projects, cost overruns recorded were 47 per cent for capital expenditure (net present value) and 17 per cent for time overruns. Optimism bias involving time overruns was lowest on standard buildings and highest on standard civil engineering projects. In all cases, the effects were significant and overshadowed project evaluations undertaken.
see: www.hm-treasury.gov.uk/media/97705/geenbool03_pres.pdf , slide 19 of 22.
Optimism bias suggests greater net benefits from major infrastructure projects compared with reality. And it affects capital investment decisions by government when policies require trade-offs between infrastructure investment decisions and investment decisions into other areas of public need. Such optimism bias is also crucial when, for projects such as toll roads, government is effectively committing future revenue flows from the public, with the public, rather than the government, paying the fees.
The Committee received advice that the general principle of allocating risk to the party best able to bear it was sound but the detailed treatment of risk was less satisfactory.332 And while the government's preferred positions on the treatment of risk were supported by some witnesses, such as the Property Council of Australia,333 Professor Quiggin334 indicated that shortfalls occurred in the treatment of risk for construction, operation, services, demand, regulatory risk and network risk, and that optimal allocation was likely to be a mix of several standard public procurement options.
A further weakness of the assessment framework was the inability to accurately determine the degree to which dealing with only one supplier provided best value for money. Concerns about this issue are particularly critical in view of the known existence of optimism bias. In the words of one witness:335
The principle of optimal risk allocation requires the availability of a range of contracting arrangements. A single contractor model will be appropriate only in a minority of cases. For most infrastructure projects, standard public procurement procedures, with subsequent public ownership of the asset will be preferable.
The use of a risk adjusted PSC (adopting advanced probability valuation techniques) in project evaluation, as detailed in the PSC Supplementary Technical Note (July 2003, p.30), carries with it particular risks that should be explicitly acknowledged.
Whilst the application of probability based methods is conceptually sound, their proper use requires databases of empirical experience and reliable information as to their foundation. In the absence of this information and in consequent use of various assumed distributions, there is a substantial risk that such assessment methods can mask inherent uncertainties in the assessment, and result in evaluation methods and results being inaccessible to all but the official undertaking the evaluation. Moreover, the use of such methods may be subject to significant 'tuning' or manipulation to provide plausible or desirable results rather than assessment of strength and veracity. The use of sophisticated assessment techniques to massage broad assumptions and information of doubtful accuracy leads to results no more accurate than the initial assumptions.
Past these broad principles, the central issue here is not whether one party or other bears particular risks; the key issue is the valuation put on these risks and the price paid to parties for bearing these risks. The valuation of risks is an area fraught with uncertainty and as noted in the previous discussion of the PSC, an area in which there appears to be only a weak empirical basis of experience. The Committee was informed that different parties could easily value the same risk environment quite differently:336
Risk is a bit like beauty, it is in the eye of the beholder, so even for exactly the same set of circumstances two different parties - be they public or private sector - could come up with a different analysis on what that risk will be.
In respect to paying private parties to bear these risks, the Committee noted that the guidelines stated:337
There is in fact a profitable market in risks, with private parties keen to assume risks for which government pays a high price relative to the likelihood or consequence of these risks materialising.
In concept, the PSC aims to ensure that government is not charged an excessive risk premium. This comparator documents the value of risks on the basis of previous estimates for likelihood and consequence and enables relative value for money assessments to be established.
__________________________________________________________________________________________________
322 Department of Treasury and Finance, Partnerships Victoria Guidance Material, Risk Allocation and Contractual Issues, June 2001, p.121
323 ibid.
324 Department of Treasury and Finance, Partnerships Victoria Guidance Material, Public Sector Comparator, Technical Note, June 2001, pp.32-33
325 Dr R Arndt, transcript of evidence, p.47
326 D Harding, 'Running the Risk of Failing Public Services', Public Finance, April 5-11 2002, London, p.12
327 J E de Bettignies and T W Ross, The Economics of Public-Private Partnerships, draft report, Vancouver: Sauder School of Business, University of British Columbia, 2003 (?)
328 Dr C Duffield, Senior Lecturer, Faculty of Engineering, transcript of evidence, p.49
329 Mr N Morris, Chief Executive, Tasman Economics, transcript of evidence, p.43
330 HM Treasury (UK), The Green Book Appraisal and Evaluation in Central Government, 2003, p.29
331 The Green Book: Evaluation and Appraisal in Central Government (2002), HM Treasury (UK), presentation on launch, 22 July 2002,
332 Professor J Quiggin, submission no.25, p.27
333 Property Council of Australia, submission no.33, p.2
334 Professor J Quiggin, submission no.25, p.32
335 ibid., p.33
336 Mr J Miller, Executive Director, Macquarie Bank, transcript of evidence, p.144
337 Department of Treasury and Finance, Partnerships Victoria Guidance Material, Risk Allocation and Contractual Issues, June 2001, p.21