I have just got back to base after an interesting and worthwhile Dispute Review Board Foundation International conference in Santiago, Chile. A topic of particular interest was the use of Dispute Boards in PPP projects.

Most readers of this post will know what a PPP project is, but for those who don’t, I have included a rough guide below[1]. Likewise for those who might like a quick guide to dispute boards[2]. But what has this got to do with game theory, and in particular, prisoner’s dilemma?

I was first introduced to prisoner’s dilemma by the revised version of Richard Dawkins The Selfish Gene. It is all to do with iteration, and why, when only the fittest survive, so many organisms, including homo sapiens, will often do selfless things for the benefit of their own kind.

Roughly, it goes like this. A couple of likely lads are engaged in a criminal enterprise, they get caught. They are imprisoned and questioned separately. For each of them, the question is the same: do they rat on their colleague, and thereby get a lighter sentence at the expense of the other, or do they say nothing? It turns out that the answer is to do with how many times the game is played. If only once, it might be a good strategy to behave badly, but if the same scenario is played out over and over again, it turns out that the better strategy is to cooperate with your fellow prisoner. People who like this sort of thing write computer algorithms to test all sorts of strategies, including how to cope with a colleague who has previously ratted on you (it seems that the best strategy is to retaliate once, and then forgive).

The parallel with construction contracts is not hard to see. If you are a construction project owner, is it better to insist on an onerous construction contract laden with time-barring notice provisions, and then to shaft the contractor when, as planned, it proves impossible for the contractor to comply with those provisions? Or is it better to cooperate with the contractor, and to pay the contractor for what the contractor is reasonably and legitimately entitled to? The same question applies the other way round: if you are a contractor, is it better to bid low, claim high and to cut every possible corner in the construction process? It seems that, to at least some extent, the game theory applies: aggressive behaviour might work on a “one-off” basis, but in the long run, it is better for owners and contractors alike to cooperate with each other, and to behave reasonably. For owners, they get their projects delivered more reliably on time and on budget, and for contractors, they are much more likely to be able to earn the profit that they need to stay in business. A key feature, of course, is that these are not zero-sum games. Aggressive behaviour typically leads to long and expensive litigation or arbitration which places huge commercial and financial strain on both parties alike.

It is because of this feature that PPP projects pose a particular challenge. Typically, the concessionaire, or Single Purpose Vehicle as it usually is, is concerned only with the one project that it undertakes. There is no iteration. Successful owners and successful contractors alike may well come to understand that, in the long run, they do better by behaving reasonably, but for an SPV, there is no long run. And for that reason, it may well be thought that the well-documented advantages of a Dispute Board may well be particularly relevant in the context of a PPP. Which variety of Dispute Board is used may not matter much; it may be a traditional dispute review board, or a FIDIC style Dispute Adjudication Board, or an Australian style Dispute Avoidance Board.  There is a massive evidence to demonstrate that the use of these boards hugely reduces the likelihood of litigation or arbitration and that is why, of course, funding authorities like the World Bank and JICA insist on their use.

Almost always, these boards are set up to operate only at the top level of any contractual regime which, in the case of a PPP, is the concession agreement between the owner (usually a government agency) and the SPV. But that may not be the best way of doing things, at any rate where the board is in the more modern form, where it is tasked not merely with deciding disputes, but with assisting the parties in preventing potential issues from hardening into disputes. That is because, sitting below the concession agreement, there is typically a design and construct contract by which the SPV procures the delivery of the project, at any rate before it moves into the operations phase. These arrangements are often back to back such that if, for example, the owner orders a scope change in the project, the provisions for adjustment of price in the concession agreement are typically mirrored more or less exactly in the design and construct contract which sits below it.

So, here is the problem: an issue arises as to how much time or money should attach to an owner’s scope change. In practice, the SPV is likely simply to pass on the D&C contractor’s claim to the owner. A discussion merely between the owner and the SPV is unlikely to resolve any issues as to how such claim is to be treated, because the SPV is not going to agree anything without the D&C contractor’s consent. Indeed, there may well be contractual provisions preventing the SPV from agreeing anything without the D&C contractor’s consent. And so, from a practical point of view, if a Dispute Avoidance Board is to succeed in preventing such issues from involving into full-blown disputes, it is likely to have to engage, not only with the owner and the SPV, but also with the D&C contractor.

The notion of a Dispute Avoidance Board agreement including all three parties, and not just the owner and the SPV, may seem novel. But if these boards are increasingly to be used in the PPP context, it will probably be sensible for the agreements to go that way.

[1] Suppose a government wants a piece of infrastructure. It might be a road, or a bridge, or a tunnel, or a desalination plant, or a hospital. The traditional way for the government to finance the project is to borrow the cost of the project from the market, and eventually to recoup that cost either by taxation, or by some sort of tolls or other charge, or a combination. But governments have become increasingly averse to this approach, because an overall level of borrowing by a government that is too high in relation to its annual tax take is dangerous, and can lead to a lack of confidence, as countries like Greece and Argentina have discovered to their cost. So they want to get the cost of these projects off the balance sheet.

A public-private partnership, or PPP, can achieve this. The government enters into an agreement with an entity which is usually a consortium of financiers and contractors – variously described as a concessionaire, or a Single Purpose Vehicle  – whereby the SPV agrees to construct and then for a lengthy period of time operate the asset without any upfront payment. The SPV then eventually gets its money back either by the government renting the asset from the SPV for a lengthy period of time, or by enjoying the income that the asset produces – such as by road tolls – or by a combination.

The financiers within the SPV will typically look for as much certainty as they can possibly obtain in the arrangements, so as to secure their long-term return. The SPV’s obligation to the owner to design and construct is typically a “turnkey” type of obligation, and the SPV will typically pass that same obligation down the line to a D&C contractor. In practice, the SPV will typically be a consortium, including the contractors who are going to build the project, and sometimes also the people who are going to operate the project once it has been built.

Some PPPs have been very successful, and others much less so. One feature of them is that the documentation tends to be extraordinarily complex. They are sometimes criticised as being an expensive way of procuring major projects, and because they have the effect of passing the cost of projects down to future generations. But they show no signs of passing out of fashion.

[2] A dispute review board is usually a panel of 3 experienced and independent people who are appointed at the beginning of a project to manage the process of any issues or disputes which may arise out of the project. A standing board will regularly meet with the parties on site to review the progress of it. If any disputes arising out of the work cannot be resolved by the parties, it can be referred to the board which will promptly evaluate the issues and give a non-binding recommendation as to how the dispute should be resolved. The mechanics for this process have evolved since the 1990s such that the track record is extremely good. Whereas it is typical for a project without a dispute review board to lead to litigation or arbitration, the vast majority of dispute review board recommendations are accepted by the parties, thereby avoiding that litigation or arbitration.

In the international context, the concept of review boards has been somewhat extended to cover much the same ground as is covered by the adjudication process in the UK, Australia, Singapore and Malaysia: a dispute adjudication board is charged with making decisions which are binding as to cash flow, i.e. the same “pay now, argue later” concept which applies to adjudication. The FIDIC suite of contracts contains provisions for dispute review boards (decisions not binding), dispute adjudication boards (decisions binding on an interim basis)or combined dispute boards (which are a hybrid).

Meanwhile, particularly in Australia, the emphasis has been on the avoidance of disputes, and hence Dispute Avoidance Boards, where the emphasis is on the board working to prevent potential disputes (known as issues) from developing into hardened disputes, with the board’s power to make formal recommendations or determinations being a power of last resort.