The water company has attributed its missed leakage targets as stemming from a decision in 2015 to develop an innovative delivery vehicle with its infrastructure partners to accelerate improvements which failed to deliver.

According to the consultation paper, the failure resulted from how Thames designed, implemented and managed the contracting arrangements for the infrastructure alliance model it chose to adopt for the outsourcing of its leakage reduction and management activities for AMP6 .

The paper says that under the model, the utility devolved operational decisions to its Alliance partners and did not hold the partners to minimum standards of performance but instead relied primarily on the contract’s financial incentives to drive performance. The regulator pointed out that alliance models have been “successful elsewhere in delivering efficient and innovative approaches to leakage management.”

Thames Water only allowed a three-month transition period for testing the innovative arrangement with its Alliance partners, before moving to full implementation in April 2015. At this point it stepped back from the detail of overseeing and managing the operational delivery of its leakage performance commitments.

Ofwat commented:

“We have looked into other company supplier arrangements, some of whom refer to their suppliers as framework partners and some operate as alliances.

“We understand from other companies that it has taken several years to establish alignment of understanding and incentives between purchaser and contractor. This level of understanding is necessary effectively to operate partnerships and alliances similar to the one Thames Water sought to establish in three months.”

“Alliance underperformed badly, both operationally and financially, from outset of AMP6”

The regulator said that Thames Water had explained that “financial incentives drove perverse results and the Alliance underperformed badly, both operationally and financially, from the outset of AMP6.”

Thames Water’s poor performance was further affected by:

• Poor management of the transition from three main leakage contractors to two, which resulted in the loss of an important and experienced leakage detection team and its replacement with less experienced work gangs;
• Mis-targeted work (partly through deterioration in expertise, partly driven by the financial incentives) leading to work gangs digging a high proportion of ‘dry holes’ i.e. an attempt to repair a leak where no leak is found at the point highlighted for excavation:
• Poor design and implementation of Thames Water’s “OneDesk” initiative which had been intended to introduce a more efficient means of scheduling works and relied on key staff being prepared to move job locations from London to Reading, which did not materialise. Workforce performance declined and Onedesk had to be reversed;
• Reducing work gangs available for leakage detection and repair at a time when Thames Water’s leakage reduction performance was falling further behind target; and
• Slowing down of mains replacement activities.
• Thames didn’t fully consider underlying causes of declining performance which continued to deteriorate throughout 2016-17

By the middle of 2015, Thames Water was sufficiently concerned about its deteriorating leakage reduction performance to call a “leakage summit” which led to a number of one-off actions that enabled it to meet its leakage and SoSI performance commitments for 2015-16.

However, “Thames Water did not fully consider the underlying causes of its declining performance, or make material changes to the Alliance and leakage reduction performance continued to deteriorate throughout 2016-17.” the paper says.

By September 2016 Thames’ operational staff had acknowledged that there was “no realistic prospect of meeting” the leakage performance commitment for 2016-17. However, the Group Management Reports for September and October did not convey the operational staffs’ assessment.

It was only in November 2016 that the Group Management Report warned Thames Water’s Board that meeting its target was “very challenging”, and not until January 2017 that it was told that the performance commitment would not be met.

"Thames has spent customers and shareholders' money inefficiently"

Thames Water initially informed Ofwat in April 2017 that it had missed its performance commitment on leakage and was seeking to address the root causes of its poor leakage performance. The regulator then subsequently learned that it had taken the following steps:

• From December 2016 new appointments to a number of key roles, including a new Managing Director of Wholesale Water and a Head of Water Networks to investigate and address the structural issues in the Alliance, oversee leakage operations, and improve performance.
• More than 700 outsourced employees brought back in-house and increased the number of dedicated leakage work gangs
• Negotiated new terms with Alliance partners for remaining two years of AMP6
• gave itself a much more active role as “Purchaser” than before to manage and monitor the Alliance
• Reinstated the most experienced of its leakage contractors from AMP5 as a main contractor for North London
• Now using acoustic loggers to detect leaks (more effective than listening sticks used previously which are also still used). The utility has already deployed 18,000 and expects to deploy further 8,000.

Summing up its findings, Ofwat commented:

“We find that Thames Water has failed to maintain an efficient and economical system of water supply within its area in relation to its management of leakage during AMP6. ……the Alliance’s structure and contractual arrangements incentivised inefficient leakage management which Thames Water did not materially address until two years of poor performance had elapsed.”

“In doing so, it has spent customers’ (and its shareholders’) money inefficiently and added additional, unnecessary stress to its ability to continue to secure water supply to its customers.”

The paper says that Thames Water failed to ensure it had adequate management resources and systems of planning and control to manage network leakage from 2015-16 and that the reasons for its underperformance were within its control. These stemmed from issues in the design, implementation, management and oversight of its Alliance arrangement, alongside “the problems associated with its abortive efficiency rationalisations.”

"Thames Water's actions are more negligent rather than deliberate"

Pointing out that as it was originally conceived, the Alliance structure did not provide Thames Water with adequate systems of control over leakage management, Ofwat criticised the utility for relying on incentives in the contractual framework to drive performance and relinquishing “too much direct control of its operations.”

“It handed responsibility for operational performance management of a key component of its operational resilience of critical importance to its customers to its contractors without adequate safeguards.”

Ofwat also commented:

“While there is no evidence of Thames Water deliberately making efforts to conceal the failure, it was, until recently, less open than we would expect a public utility to be.”

“We are particularly concerned about the insufficiency of Thames Water’s Board’s oversight of leakage performance, particularly its apparent lack of challenge on operational aspects of its leakage management within the Alliance before performance fell irretrievably beyond the ODI collar for 2016-17.”

“On the evidence, however, Thames Water’s actions are more negligent rather than deliberate.”