7.3.1 Authorities should consider the following principles in structuring their regime:
• where the mechanism measures both availability and Service performance, the distinction between points to be included as part of the definition of availability and those to be part of a performance measurement system is important.120 Occasionally there is an overlap, for example, in a hospital cleanliness is an important element of availability, especially for operating theatres, but lack of cleaning can also constitute a performance failure. In this example the distinction is ultimately one of degree; it may be appropriate for such items to feature in both the availability definition and the performance measurement system so long as the arrangement is clear and double- counting is avoided.121
• the mechanism might involve (a) direct cash deductions for failure, whether in terms of unavailability, poor performance, or other terms appropriate to the mechanism for a specific project, or (b) a two-stage system whereby failure leads firstly to points which in turn lead to cash implications (e.g. once a certain number of points are reached). If the definition of availability omits key requirements - these being instead covered by the performance measurement system - the latter could include both immediate payment deductions and a system of performance points;
• the payment for a given period, e.g. a month, might be structured as (a) starting from zero, with payment increased in response to availability and performance, or (b) increasing for availability but subject to deductions for poor performance, or (c) a base case figure for the Unitary Charge for that period which is subject to deductions for both unavailability and poor performance;
• a number of performance regimes amongst early projects have been over-elaborate and, as a result, ineffective, and some have not been designed with enough consideration of the practicability of day-to-day procedures. As a rule, simple is best. The payment mechanism should seek to "measure less but measure well";
• unnecessary complexity can also arise where the movement from specification of inputs to specification of outputs has not been endorsed by the Authority. The mechanism should concentrate on measuring "outputs" not measuring "inputs"; and
• Authorities should first build their model from the bottom up, ensuring that for any particular area, or service failure, there is an appropriate potential deduction. The model should then be reviewed from the top down, considering the cumulative effect of all types of deduction. If it is "over-weighted" in terms of the scale of likely deductions, it will encourage excessive risk pricing and cash reserving by bidders. If it is "under-weighted", bidders will not be sufficiently incentivised to rectify Service failures.
7.3.2 The key objectives are value for money and effective incentivisation; the Contract should contain sufficient incentives for the Contractor to rectify any default.
7.3.3 A variety of payment mechanism structures have been used across various sectors and project types. Models A to C below illustrate alternative features of a payment mechanism. They are not intended to be prescriptive:
• Model A - the Unitary Charge is built up from the number of available places or units, so only places or units that are available are paid for. The definition of available places or units incorporates the provision of ongoing services that are core to the requirements of the Authority to carry out its functions. On top of this, a performance regime is imposed to address levels of performance that do not impact on the availability of places and/or Services that are outside the definition of the core Service. Sub-standard performance leads initially to "performance points" accruing and, only indirectly, to deductions from the Unitary Charge once a certain level of points has accrued. A ratchet increases the deduction for repeated performance failings.
• Model B - the Unitary Charge (before any deductions) is based on a full provision of the overall requirement and the payment mechanism determines deductions both for unavailability and sub-standard performance, i.e. both are represented in the calculations as deductions from the 100% level. Availability is defined in terms of being usable and accessible and different deductions are made depending on which area is unavailable. There is a system of weightings whereby each section of accommodation is divided into units and given a weighting depending on its importance; for each failure to provide an available unit there is initially a payment deduction equal to the Unitary Charge multiplied by the relevant weighting. The deduction can be based on an escalating tariff so that subsequent days of unavailability of the same space may lead to progressively higher deductions.
• Model C - the Unitary Charge is based both on availability and usage, for example in the context of a training service with payment per training place. There is a minimum take-or-pay level (i.e. if usage by the Authority is below that level it must nonetheless pay for usage at that level) in order to assist with the financeability of the Project and to strike a balance between flexibility and the cost implications of greater risk transfer. There is still a requirement for availability to be defined, as the Authority will not pay above its actual usage if the units of Service are not available. There can in addition be scope for payment deductions for poor performance of Services.
_________________________________________________________________
120 For example, if prison cells or classrooms are in practical terms un-usable, it may be most appropriate for them to be regarded as unavailable, with direct consequences in terms of the Unitary Charge. Alternatively, some problems with prison cells or classrooms might not make the room un-usable and instead the shortcoming might be better described as a performance failure. The bar in terms of 'usability' need not be set so high as "not able to be used" - the concept of "unavailable-but-used" can provide a compromise position (see Section 7.5.12).
121 The split between 'availability' and 'Service performance' is not itself a requirement of this guidance, however, in most sectors this distinction has been found to be helpful. The underlying principles set out here apply even where this split is not observed.