We recommend that data from the procurement and construction phases of projects is collected centrally. Standard questionnaires, applicable to projects from any sector should be completed by procuring authorities at two fixed points in the process: (i) at financial close; and (ii) at the start of operations. At both of these points data should be readily available - and the authority bid team should be in a position to provide it at relatively low cost17. The majority of the data collected would be factual, although some more subjective questions, for example about the quality of design or the relationship with the private sector, could also usefully be included.
A potentially important area of added value going forward in seeking to answer the Scottish Executive's key question would be to collect the same data for non-PPP procurements. Many of the data fields would be applicable to design and build contracts. We therefore recommend the same questionnaire should be used to collect data from non-PPP projects (over a fixed capital value), enabling comparisons to be drawn.
The data fields used should be consistent with the data collected in stage 1 of this study (which was designed to be consistent with previous NAO studies), and with Partnership UK's database. Figure 19 and Figure 20 below show the main categories of data that should be collected and the research areas that the data could be used to inform.
Figure 19: Procurement phase data to be collected at financial close
| Data Category | PPP / Non-PPP | Research Area |
| Project name / description / sector / parties / contacts etc. | Both | For reference and to enable segmented analysis by e.g. sector, major contractor. |
| Timing from OJEU to financial close | Both | Procurement timescales (i) PPP vs non-PPP; and (ii) over time. |
| Number of bidders | Both | Level of competition (i) PPP vs non-PPP; and (ii) over time. |
| OBC and FBC price assumptions | Both | Optimism bias |
| Price data and underlying costs | Both | Costs of PPP vs non-PPP; information for future business case / affordability projections. |
| Price movements (value and reasons) | Both | Effectiveness of competition (i) PPP vs non-PPP; and (ii) over time. |
| Accounting treatment | PPP only | Changes in accounting treatment over time. |
| Both (Design /Construction risks) PPP only (Operational risks) | Comparison of risk transfer between PPP and non-PPP projects. Changes in risk transfer arrangements over time. | |
| Procurement costs | Both | Procurement costs (i) PPP vs non-PPP; and (ii) over time. |
| Views on design / innovation / relationship | Both | Comparison of PPP with non-PPP projects. Comparison with responses to the same questions at later stages of the same project. |
| Financing details | PPP only | Comparison of financing terms with PPP projects in rest of UK (using PUK data) |
Figure 20: Construction phase data to be collected at operations start
| Data Category | PPP / Non-PPP | Research Area |
| Timing of delivery and reasons for delay | Both | Construction timescales (i) PPP vs non-PPP; and (ii) over time. |
| Price changes and reasons | Both | Delivery within budget (i) PPP vs non-PPP; and (ii) over time. Flexibility (i) PPP vs non-PPP; and (ii) over time. |
| Views on design / innovation / relationship | Both | Comparison of PPP with non-PPP projects. Comparison with responses to the same questions at earlier and later stages of the same project. |
Most of this data is now available for the projects that responded to the survey. However, it is likely to be a reasonable amount of work to collect data for all projects, both those that did not reply to our survey and those that were not operational at the time of the survey. After that, we anticipate that it should be relatively easy to track and collect the data for new projects, although a single person within Scottish Executive would need to take responsibility for ensuring that returns are completed to the required to standard.
Our suggestion would be that the data is collected using Excel-based questionnaires and is held on a database.
It is of course important to recognise that whilst the development of this database would highlight differences between PPP and non-PPP projects, or changes in the performance of PPP / non-PPP projects over time, it would not give the underlying reasons for those differences. It is likely that this data would highlight areas where further, more in depth analysis would be required.
______________________________________________________________
17 One of the issues noted in carrying out this survey has been the time costs imposed on respondents required to answer detailed procurement and financial close questions on projects that closed many years ago.