See for yourself: The government's business case for PPPs

Education Minister Anne Tolley is pressing on with plans to privatise a significant chunk of New Zealand’s school property programme. It is looking increasingly likely that new schools and school…

schoolcrossing_1Education Minister Anne Tolley is pressing on with plans to privatise a significant chunk of New Zealand’s school property programme. It is looking increasingly likely that new schools and school building projects will be bundled into build-and-maintain public private partnerships. Australian companies have expressed interest in such contracts. EA has obtained the government’s stage one school property PPP business case under the Official Information Act. Copies can be downloaded here. The business case was made by an international consultancy, Castalia Ltd, that specialises in PPPs. Concerns about the case were expressed by two economists, John Carran from Infometrics and Bill Rosenberg from the Council of Trade Unions, including that schools and parents are being excluded from this process, and as a result seriously shortchanged.

Following the demotion of ACT MP Heather Roy, Rodney Hyde has been appointed Associate Education Minister with responsibility for school PPPs. As the New Zealand Herald‘s business columnist Fran O’Sullivan wrote, with reference to the South Canterbury Finance bailout, “the smart money now knows that the Key Government can be˜gamed’.”

The documents might take a little while to download. Following the documents are comments from Bill Rosenberg at the Council of Trade Unions, and John Carran from Infometrics.

Letter to Jane Blaikie from the Ministry of Education
Official Information Act Document 1 pdf
Official Information Act Document 2 pdf
Official Information Act Document 3 pdf
Official Information Act Document 4 pdf
Official Information Act Document 5 pdf
Official Information Act Document 6 pdf
Official Information Act Document 7 pdf

The Proposal for School PPPs

Bill Rosenberg, Policy Director/Economist, New Zealand Council of Trade Unions Te Kauae Kaimahi

The Government contracted Castalia Ltd, a consultancy which, according to its web site “pioneered public private partnerships [PPPs] in infrastructure around the world”, to provide a business case for PPPs “as a method of procuring schools in New Zealand”. The reports of the business case, dated May 2010, were released to the New Zealand Educational Institute (NZEI) under the Official Information Act. They consist of a “Stage One Business Case” and four Working Papers providing more detail.

There have been numerous details deleted from the released papers, notably the numerical details of the comparative costs of the status quo and PPPs. While the reports give the results of these comparisons in general terms, because the numerical details are not provided there is no way to check either the conclusions or how large the differences are between the different scenarios. Because, as is described below, many of the judgements made in the comparison, are inherently political, and Castalia is not neutral as to PPPs compared to other procurement methods, the public is being significantly short-changed by the gaps in these releases. It is difficult to have a reasoned debate when the full facts are not made available.

Unfortunately that is the pattern for PPPs when they have been used overseas. Commercial confidentiality is commonly used to prevent PPP contracts being made available to allow the public, including concerned parties such as parents of school pupils, to judge whether deals made were reasonable and whether the commercial contractors are adhering to the conditions.

The PPPs proposed are contracts for what will typically be “a private consortium of financiers, constructions companies and facilities management firms” (p. 1 ) to design, build, operate and maintain a school’s buildings for a period of around 30 years, transferring the school to the government at the end of the period. The government would pay for the PPP by “constant payments over the life of the contract”. The provision of education would remain with the school’s board of trustees (BOT) and the land would still be owned by the government.

While the documents describe the PPP as “essentially the same concept as any government agency leasing space in a commercial building”, in fact there are significant differences. The PPP contractor has the right to use the premises out of hours and possibly for other commercial purposes such as installing vending machines, which would be unusual for leases. The PPP contractor also supplies and maintains furniture and equipment other than specific items procured by the school, and controls their movement (p.12, 13); such items are usually provided and controlled by tenants of commercial leases.

Currently, schools are typically “procured” by a contract between the Ministry of Education and a construction firm with design, operation and maintenance remaining with the Ministry and the BOT. The method is called “Guaranteed Maximum Price” (GMP) which Castalia concedes “stakeholders generally agree … works well in terms of delivering high quality school facilities, within budget and in a reasonable time frame” (p.9). It is used as a comparator for the PPP analysis. Undoubtedly it could be improved or could be modified to achieve additional objectives, but Castalia considers only one alternative “modified GMP” process, that of increasing the number of bidders.

Educational needs

Castalia is at pains to emphasise that “the operation of the school itself (that is, the provision of education) would remain the responsibility of the BOT”. However there is the risk that the way in which the buildings are operated by the PPP contractor interferes with the provision of education. Castalia implicitly assumes that the contract for the PPP will be complete and ensure that such interference will not occur. That is an heroic assumption. It is very difficult, costly, and makes for complex and possibly inflexible contracts. Castalia in fact gives two significant examples where problems could occur: the roll numbers are greater or less than anticipated in the contract; and that educational needs change over the 30 years of the contract and require changes to the buildings. We deal with the financial implications of such changes below, but there are other possibilities that could also interfere with educational provision such as delays in maintenance, disputes as to whether work on established buildings is maintenance or building change, conflicts between the school’s or the community’s after-hours use of the buildings and commercial use bringing income to the PPP contractor, and conflicts in design objectives. Many of these have been cited as concerns with PPPs elsewhere.

After-hours use of buildings is important for school sports and cultural activities related to educational provision. It may also be important for community education or as a facility (and sometimes the only one available) for use by the community. However PPP contractors will aim to make money from hire of the facilities, and may in fact “design the school with after hours use in mind” (p.28). Castalia suggests that community and school use can be catered for from a “bank” of free community hours built into the PPP contract. That allows for little flexibility such as may be required by changing government funding of community education, and changing school or community needs. The government will not see savings in the contract from the income received by the PPP contractor: “Experience in the U.K. and Australia suggests that, in practice, PPP contractors are reluctant to take risk around after-hours revenue, and tend not to factor this into their bid price”. The analysis assumes that the best measure of the value of after-hours use to the community is their willingness to pay. This implies that its availability may be limited for low income groups.

Castalia emphasises that one of the savings brought by PPPs is reduced maintenance costs over the lifetime of the building. “A PPP contractor would approach the design and construction of a new school differently from a design-build contractor – Rather than seeking to maximise the quality of the school within the Ministry’s budget, a PPP contractor will try to design and construct the schools in a way that minimises the total lifetime costs for the schools subject to meeting specified quality standards.” (p.27) Minimising lifetime costs may mean compromises in what would be useful for teaching. For example, it may be more expensive to maintain a classroom whose walls are designed to make it easy to display pupils’ work, posters or other material, than one with easily cleaned hard surfaces.

Effect on school employees

Castalia anticipates that there will be an increase in subcontracting and that wages and conditions will be forced down in order to save costs. One of the means of cutting costs is specified as follows: “We assume a PPP contractor will improve the efficiency of caretaking and cleaning by 20 percent including through contracting out and stronger labour bargaining” (p.25, Working Paper No. 2, Cost Benefit Analysis).

Cost analysis

The documents provide two cost analyses. One is a cost-benefit analysis which claims to look at the full economic effects, taking into account broader costs and benefits to the economy as a whole. The other is a “public sector comparator” which looks at the financial cost to the government. Both have biases against public ownership on a number of counts. Even then they conclude that PPP procurement will provide only “a small improvement in value for money” over current practice (p.24). While the size of the “small improvement” has been suppressed in the released documents, it is probable that without the biases, the “small improvement” would become a deterioration. These biases include:

1. Financing

Despite the lower costs of finance that the government faces compared to the private sector, Castalia follows the Treasury’s guidelines for PPPs and uses commercial rates of finance for its estimates (p.32ff, Working Paper No. 2, Cost Benefit Analysis ). It acknowledges that this is “controversial” and indeed it is, because it has a significant effect on the comparison. These rates are the basis for “discount rates” used to compare the costs of the alternatives over time using net present values. As Treasury states, “small changes in the discount rate can have a significant impact on the total value” of each comparator . Castalia argues that using a public sector cost of borrowing would result – in at least two perverse outcomes:

  1. The government would have an advantage over private investors for all investments, and in theory would replace all commercial sources of finance in order to improve economic outcomes.
  2. Investment decisions made by the government would be biased towards sectors with higher risk and higher returns” eventually unbalancing the risk profile of the entire investment portfolio.”

These are politically loaded arguments which take the possibilities to an absurd extreme. Of course the government has an advantage over private investors due to its lower risk. That is a reality, which should be used in the public interest, and not “neutralised” away. It is absurd to suggest that therefore the government would therefore take over financing of the entire economy. Private investors have other advantages: financing costs are not the only consideration, as is repeatedly pointed out in Castalia’s own analysis. Regarding risk, the government may well have a role in taking on higher risk projects which are socially or economically beneficial and which the private sector is unwilling to take on sufficiently (education may be considered one example) but Castalia’s argument ignores the fact that government investment is not in general primarily for the purpose of obtaining financial returns. It is also likely to have other controls on financial risk.

Not only does Castalia discard the lower costs of public finance but it argues for a lower cost of finance than Treasury (4.5 percent compared to Treasury’s 6.1 percent), thus lowering further the estimate of PPP costs. Their reasoning is interesting: that PPP contractors’ risks are lower than the commercial building providers Treasury uses for its comparisons. The risks are lower because the government will shoulder the risk of lower than anticipated student numbers compared to (say) AMP NZ Office Trust which takes the risk it cannot fully lease out its buildings. The government would “make availability payments to the contactor regardless of student numbers”, says Castalia (p.34, Working Paper No. 2, Cost Benefit Analysis). This reflects U.K. experience where for example “Balmoral High School in Belfast closed six years after it was built, when pupil numbers halved. However, the Northern Ireland Department of Education owes the contractor $370,000 a year for the next 18 years” .

2. Risk

Castalia treats risk in inconsistent ways. For example it lays out three possible “regulatory” risks for a PPP contractor: those of the government changing policies during a 30 year contract; of the uncertainty of a new procurement process; and failure of the government to fund payments as they fall due (this last one being regarded as low for New Zealand). Yet it does not increase the costs of the PPP contract for these on the grounds that it is difficult “and could be contentious”, despite evidence from the U.K. that contractors demanded a premium for some of these risks (p.35-36, Working Paper No. 2, Cost Benefit Analysis). There are other examples where the PPP contractor is asserted to be able to spread risks over a number of contracts, disregarding the likely greater ability of the government to do the same over many more schools and contracts.

Risk is a central issue. One of the major advantages of PPPs is said to be that it transfers certain risks from government to the PPP contractor. It would be expected that if that were happening the PPP contractor would increase charges to reflect the risk. Yet the Castalia comparison does not appear to fully include those costs and exaggerates the risks that are in theory transferred.

Some of its claims for private sector superiority in reducing risks simply lack credibility. For example, they state that “the private contractor is more likely to manage and monitor construction of the facilities more carefully to ensure that problems such as leaky buildings do not arise in the future” (p.34). There is no shortage of examples of private sector built and maintained buildings which suffered from the leaky building problems.

In other places, Castalia appears to be suggesting that risks to the introduction of PPPs should be mitigated by distorting the process of calling for bids or Ministry priorities. For example, it appears to suggest that to ensure that there is a successful bidder, a tender greater than the public sector alternative should be allowed (p.9, Working Paper No. 4, Risk and Implementation). It identifies a risk that “school(s) selected for PPP are needed sooner or later than [the] specified delivery date under PPP” due to changes in demographics. Their mitigation strategy is to “allow more suitable schools to be substituted should demographic drivers change” (ibid., p.5). Reactions of BOTs, unions and other stakeholders are to be met by “stakeholder management” by communications teams (ibid, p.3, 7). It is acknowledged that a single successful PPP is unlikely to be viable because of its transaction and administration costs to the Ministry; therefore additional budget should be allocated (ibid, p.10) and a “visible pipeline” of school PPP opportunities should be provided (p.26, 46).

3. Inadequate existing funding

Castalia acknowledges that “in principle a procurement method may be cost benefit justified, but still require a greater budget outlay than the current method” (p.24). In other words, PPPs may be more expensive than current procurement methods” but that is justified by better facilities. Yet they also claim as an advantage for PPPs that the contractor will “incur higher maintenance costs [increased by 25 percent] in order to extend asset lives and save on capital replacement costs” to the school’s advantage, and that “this is in part because¦ the PPP contractor will not be subject to budget constraints in the way schools are under GMP” (p.27; see also p.34). In other words, the claim is of a PPP benefit due to school underfunding. The straightforward resolution of that problem is to increase school funding for building maintenance rather than increase funding for the PPP, but this is not discussed. They simply report that “Our analysis shows that New Zealand secondary schools currently spend around 1.2 percent [of capital costs, per year] which is significantly more than the property maintenance grant funded by the Ministry”. A PPP contractor would spend 1.5 percent (p.27). The effect of the PPP is to force the government to spend more on maintenance in preference to other school needs, but to do it in a way that benefits a PPP contractor.

4. Insurance

Insurance costs are estimated using commercial insurance rates despite the Ministry’s ability to partly self-insure due to the size of its property holdings (pp.26, 28 Working Paper No. 2, Cost Benefit Analysis).

5. Disadvantages of PPPs

There is now a considerable literature from official, academic and media sources of the problems that have occurred with PPP and similar schemes particularly in the U.K., U.S.A. and Australia. These are scantily reported and dismissed by Castalia (e.g. p.21). The only one explicitly acknowledged is increased transaction costs (e.g. p.26). It is apparently assumed that all the problems have been resolved and/or can be addressed in PPP contracts. Even if that is true, it would add considerably to the cost of designing and negotiating those contracts, and the risks for the government of mistakes in negotiation. In fact as already noted, it is impracticable to anticipate and address all such risks.

One of the significant disadvantages of PPPs is if changes are required to buildings or facilities during the period of the contract. For a Ministry-owned building, changes are simply a matter of design and build on a competitive basis at the time of need. In contrast, the typical 30 years of a PPP contract is a very long time, and it is most unlikely that all possible changes can be anticipated at the time of designing the school and signing the PPP contract. Unanticipated changes in the school population or changes in teaching methods such as to or from team teaching or increased use of technology are likely over such a period and can force major changes in building requirements.

As Castalia itself puts it: “Any changes or future investment (such as modernisation) need to be agreed with PPP contractor, which has a powerful bargaining position in negotiations for contract variation” (p.35). The PPP contractor has in effect a monopoly position, enabling it to push up costs. In the U.K., “a 2008 National Audit Office report found that £180m a year is paid out for contractual amendments [not only for schools]. And it highlighted extortionate charges for routine maintenance” such as £302 for an electric socket to be fitted, £47 for a key, and almost £500 to fit a lock” .

Castalia deny the significance of this “powerful bargaining position” because “the contract can also specify a process to agree fair compensation for changes in capital or operating and maintenance requirements. For example, competitive pressure can be put on contract variations through the use of independent experts or references to the market cost of comparable investments” (p.35). Yet elsewhere they acknowledge that accurate comparators can be difficult in the small New Zealand market, and criticise similar Ministry procurement processes for insufficient use of competitive bidders to create price competition: “Strong competition is a fundamental driver of value for money. The benchmarking and reasonableness tests used to generate the Ministry’s budgeted cost of new schools are no substitute for a competitive bidding process. This is particularly true in New Zealand where the pool of comparators is small, and where regional factors such as transport distances or local labour shortages can introduce inaccuracies into benchmarks.” (p.16, Working Paper No. 2, Cost Benefit Analysis).

Castalia do acknowledge the additional transaction costs of PPPs. Reasons include

  • PPP transaction is more complex, requiring more time inputs from professional advisers, charging higher fee rates
  • PPP transaction requires a significant amount of work to be completed to prepare bids. At least two bidders submit a proposal that will include an initial design of the school” (p.29); and
  • PPP contract will introduce more complex relationships between the Ministry, the PPP contractor and the board of trustees” (p. 5).

There will be direct costs to the Ministry and to the BOT, but further transaction costs will be incorporated in the price of the PPP contract reflecting costs to the contracting consortium. The first PPP contract will be even more expensive” Castalia estimates 50 percent higher than the long run transaction costs” but that is not included in their cost-benefit comparisons. The full transaction cost estimate for New Zealand has been suppressed, but for the first school, the public sector cost “will be as high as $6 million” (p.47), and for Australia they suggest it is a long run cost of $8.7 million per school, $4.7 million of which is public sector costs (p.31). These are very significant overheads when the estimated cost of even a large (2,200 pupil) school is only $55 million (p.17, Working Paper No. 2, Cost Benefit Analysis).

The methodology does not expose the additional costs to the government due to the need for the PPP contractor to take profits over the entire contract period, nor the economic losses if, as Castalia acknowledges is likely (p.46), significant parts of the consortium are overseas owned and profits are remitted overseas.

PPP advantages

It is acknowledged by Castalia that there is uncertainty even in the “small improvement” in value for money they find for PPPs, and they therefore rely on intangible benefits such as “improved educational outcomes through better property maintenance and certainty of costs under PPP through better risk management” to conclude that “on balance it seems likely enough that PPP will bring value for money” (p.24). However many of the intangible and quantifiable advantages claimed for PPPs could be incorporated into the existing procurement process if they were indeed considered attractive. This would eliminate the high transaction costs of PPPs and retain control with the school or Ministry. Examples include

  • Multiple bidders to increase price competition (though there may be a pay-off with quality and it may increase transaction costs)
  • Involvement of bidders in school design (though this may increase transaction costs)
  • Designing for reduced energy costs
  • Designing for reduced maintenance costs and/or total lifetime costs of the school (though this may be a compromise with other quality requirements)
  • Designing for increased use by the community of school facilities
  • Increasing charges for community use of school facilities.


Castalia’s analysis is unconvincing due to its lack of neutrality. The suppression of crucial information from the documents makes informed discussion even more difficult.

However there is considerable evidence that Castalia’s recommendation that PPP procurement of schools should proceed is unsupportable on either financial or educational grounds.

John Carran, Senior Economist, Infometrics

I have drafted some initial thoughts on the Ministry of Education’s proposal to consider using PPP arrangements to procure the building and operation of some new school properties. My comments are based on my reading of the documents you supplied me which you obtained under the OIA. I have not looked at wider evidence or analysis on PPPs apart from reading a couple of Treasury papers on PPPs. So I have limited my comments at this stage to highlighting what I think are probably the main costs, benefits and risks of PPPs for schools and where further investigation might be necessary to reach a more definitive position on whether they offer value for money.

  • PPPs are a procurement arrangement whereby a private sector supplier is contracted by the Government to design, build and maintain a facility for a specific use. Typically the contracts are very long term (20+ years). The contract can be to provide all or part of the service associated with the use of a particular property. The MoE proposal is for PPP contracts to relate to the design and construction of schools, ongoing maintenance, including cleaning and security, and the provision of certain fixtures. The contracts would not relate to the provision of education services, which the Government would continue to provide.
  • PPPs are a reasonably common Government procurement approach in a number of countries including Australia, the UK, Canada, US, and Scandinavian countries. They are commonly used for procuring and operating large infrastructure projects. The Castalia papers released under the OIA highlight the use of PPPs for school property procurement in Australia and the UK. I have not investigated at this stage whether PPP approaches to school property procurement are more widely used internationally.
  • The main advantages put forward for PPPs are:
  • They utilise private sector expertise in the management of assets, which can lead to the more efficient use
  • The risks associated with the construction and ongoing maintenance of facilities can be allocated to parties that have the best incentives to manage them effectively
  • They provide more certainty to the Government in relation to the costs of construction and ongoing maintenance of assets
  • They can provide incentives for contractors to maximise benefits to users
  • In relation to PPPs for schools, they would free up school boards from focusing on property maintenance issues. It is possible under PPP contractual arrangements to give boards responsibility for certain aspects of maintenance or fixtures and fittings if having direct control over them is important to the delivery of education services.
  • The main costs put forward for PPPs are:
  • Contracting and tendering costs are higher than conventional procurement
  • Costs of contract variations can be high
  • There can be significant costs associated with specifying, monitoring and enforcing service expectations
  • The aspect of PPPs that is of primary importance in determining their costs and benefits is the quality of the contract management, including the specifying, monitoring and enforcement of contracts.
  • Factors that may be of concern to schools in relation to PPPs are:
  • The degree of control they have over their properties and the environment that is created by the private sector provider
  • The consequences if a private provider fails.
  • With regard to 1, in principle it is possible to specify important quality dimensions of the property and its ongoing maintenance in the PPP contractual arrangements. As suggested above, allowance can be made for school control of some aspects of the property if they are particularly important to the quality of education provided. However, as mentioned above, the contracting and the ongoing management of the contract will be important to this.
  • With regard to point 2, the failure of a contractor could provide some uncertainty to the continuation of a school in a particular location. The most likely outcome in this situation would be that the contract for the continuing operation of the school property would be sold to a new contractor. This is because there would be more value to be extracted from selling the rights for providing services to the school than from just selling the property, which has limited alternative uses.
  • There are likely to be three key factors that will determine whether PPPs are a practical option for the procurement of school properties in NZ and deliver net financial benefits:
    • Whether the scale of PPP contracts is sufficient to attract interest from quality private sector contractors and suppliers
    • Whether there is a sufficient number of private sector contractors in NZ with the capability and capacity to deliver the services required under PPP contracts
    • Whether there is likely to be sufficient ongoing capability and capacity in MoE or other Government departments to manage the PPP contracts effectively.