Thames Water, TDRA and Best Value decision making.
Ofwat specifically required Thames Water to clarify aspects of how best value information was presented 2024. Despite the supposedly improved presentation in the Final WRMP24 documents SOLAR considers that the best value explanation is incredibly convoluted and opaque, and the justification for TDRA is weak and unclear.
The following examples demonstrate that point:
-Best Value model - data inputs - feasible option checking: options not assessed for inclusion in the BV model.
-Best value model – alternative plan dismissal: superficial discussion and dismissal by Thames Water of potential alternatives to TDRA.
-Best value model - Option Cost Comparison: simplistic cost only comparison without taking account of the vastly different processes involved, forecast water requirements and potentially improved wastewater regulations.
Best Value model - data inputs - feasible option checking
The key modelling tool that is used by water companies to develop the water resource management plans is the IVM – the Investment Model.
One way that water companies influence the best value modelling outcomes is by deciding what options are introduced into the IVM in the first place. From the large list of all possible supply options only those that that are screened as feasible are input into the IVM. This means the initial decisions on “what is feasible and what is not” has a significant influence on how best value plans are developed.
For example, in 2018 Thames Water considered the option of a Surbiton intake of 300mld- with a corresponding 300mld treated sewage outfall into the Burnell riverside near Teddington Weir. This option was rejected as not feasible because of the area of land needed for such a big intake and the ecological impact of such a large outfall of treated sewage.
In 2023, Thames Water did a backchecking exercise of all rejected options. One of these was the Surbiton intake. Once again it was rejected, for the exact same reasons. The question is why was the Surbiton option not re-considered at different (smaller) sizes? Between 2018 and 2023 TDRA was reconsidered at various sizes with the result that the smaller sized 75mld scheme was deemed feasible. Why was the potential Surbiton option not reconsidered in the same way?
This means that when the IVM was run for best value outcomes the possibility of a Surbiton Intake option was not even considered.
This example highlights how Thames Water can influence Best Value modelling outcomes before the model is even run. How many other options were rejected without a genuine reassessment?
Best value model – alternative plan dismissal.
Because of the high interest in TDRA Thames Water did 3 different model runs that specifically relate to TDRA:
-Remove TDRA as an option entirely
-Delay 1-in-200 resilience to 2035
-Look at the cost that TDRA would have to be to make other alternatives cost equivalent.
(No modelling was run assessing option selection if there was no 1-in-200 resilience target date set at all, nor if TDRA was delayed to the mid-2040s.)
The results of these alternative models are as follows: -If TDRA is removed as an option entirely it is replaced by a transfer from Affinity Waters new Grand Union Canal (GUC) scheme, a transfer from SES water and some small groundwater schemes and would cost £210 million more than the Least Cost plan. -If 1-in-200 is delayed to 2035 the 2035 date is achieved by the same GUC transfer and SES transfer, TDRA is deferred to 2050 and this would cost £130 million more than the Least Cost plan.
The Thames Water discussion on these alternatives is brief. The “expert judgement” on both is that they are not viable; because they would cost more; because they (Thames Water) cannot rely on the chance that Affinity Water will manage to construct the GUC and because even if Affinity Water did build the GUC they might not achieve their own demand reduction targets and then would keep all their GUC water to themselves.
Here is a critique of those points:
Costs: What is not made clear is that the numbers used in this modelling are from before the £100 million increase in TDRA cost arising from the new tunnelling method. Also, cost is discussed against the Least cost plan but against the Preferred plan (best value plan) the “delay to 2035” option is cheaper. In the context of an overall plan spend of £19 or £21bn these costs are relatively minimal.
Relying on the GUC: Thames Waters preferred BV plan does in fact already include plans to have a transfer from the GUC, so to say this is not something that can be relied on does not make sense.
Failure to meet demand targets: Finally, if Affinity Water was to fail to meet its demand management to the extent that it would not be able to share any GUC outputs then is it not likely that Thames Water would be in a very similar position? In that situation then a larger scale solution than TDRA would be needed for London.
One benefit of the “delay 1-in-200 to 2035” model option is that it defers TDRA to 2050. Such a delay would allow Thames Water to assess how it is meeting demand management (leaks, meter installation etc) and give the water industry time to consider the impact of new regulations on wastewater treatments.
Overall, the expert judgement dismissing these two options appears to be more about rubberstamping TDRA rather than a genuine consideration of alternative plans.
Best value model - Option Cost Comparison
To demonstrate that TDRA is a best value option Thames Water ran an exercise comparing the relative costs of TDRA and Beckton Reuse 100mld. In the model they assess that TDRA and Beckton Reuse are equivalent in all best value metrics except Cost and Carbon. They then estimate that TDRA would have to end up costing cost 3 times as much as it does before Beckton would become a viable alternative.
Thames Water make 3 main points in their discussion:
1.Thames Water says “As a direct river abstraction, it performs well against alternatives such as Beckton and Mogden Recycling because it requires a less intensive treatment process, reducing capital and operating costs and carbon emissions. … “
So, straight from the horse’s mouth “because it is a Direct River Abstraction”; There is no doubt that TDRA is “performs well” at a simple cost level than Beckton Reuse. This is because Beckton Reuse is an advanced wastewater reuse scheme; it is a wastewater for drinking water reuse scheme which inherently involves more processing. It could produce 100mld of drinking water, scaling up to 300mld over time if needed. Beckton reuse would take just a little longer than TDRA to construct. It is possible there could be industrial non-potable benefits as well. A direct comparison based only on cost is not appropriate for such vastly dissimilar options.
2.Thames Water say that there are some things that could lead to cost increases for TDRA: “… substantial improvement in environmental performance of the wastewater recycling alternatives, a step change in the forecast demand for water or supply capability, or a change in drought resilience policy.” Related to the issue of TDRA cost increases, Ofwat has queried the space constraints at Mogden and how they might impact the cost of development of TDRA including its future viability.
These things are important:
-As of Jan 2026, specific to the point on environmental performance the Government Water White Paper has described issues with wastewater treatment, the Government is having policy discussions about PFAs in the environment and even Emma Hardy (water minister) stated that “we are not going to be the dirty man of Europe under my watch”. These points highlight the need to consider the future flexibility of sewage infrastructure so it can match new requirements.
-On the possibility of a step change in forecast demand, the push to increase housebuilding and new AI infrastructure amongst other economic drivers seem to be just the step changes in forecast demand that would require a braver approach to supply development.
-On the issue of drought resilience SOLAR has pointed out that the date to achieve 1-in-200 is not a fixed date yet Thames Water do not assess the full range of alternative scenarios.
-On the impact of the space constraints at Mogden Thames Water have not provided any information.
So, on every one of these points there seems in fact to be a verry likely “upward” pressure on costs.
3.Thames Water also has this to say; “We do not agree that scheme flexibility is linked to the chances of an option becoming a stranded asset. With a complex supply demand problem, programme-level flexibility is more important than option-level flexibility and it is equally possible to provide programme flexibility by using fixed output schemes as it is from developing modular assets. Both types of option may be selected by the Investment Model.”
SOLAR has pointed out that the combination of lack of scalability, significant “down time”, utilisation of tertiary as opposed to advanced treatment and the impact of climate change on river flow, means that over time TDRA risks becoming a stranded asset. In other words, it is not sufficiently adaptable to future requirements.
What Thames Water are saying is that basically they don’t agree with that idea! They think individual options do not have to be flexible to be useful in the overall plan (programme). This may be true when you are talking about options already in existence; but it is not true when you are planning to develop new options. New supply options are being developed specifically because of climate change, because of population (and housing) pressures and because of the need to protect the environment. They should therefore be adaptable to those things.
Basically, if you have an existing house with single glazing, you might continue to use it, but if you are building a new house, you don't install single glazing, you install triple glazing because of the benefits now and in the future.
TDRA is the single glazing short term fix. If more water is needed, then development of Beckton Reuse not TDRA first makes more sense. It is the 'triple glazing' water supply option; climate change ready, can be upscaled if needed and uses advanced wastewater treatment.
Despite these points the “expert judgement” seems to focus only on the simplistic relative cost difference between TDRA and Beckton reuse. And that is the end of the discussion. Thames Water, WRSE, and DEFRA should develop the topics mentioned above in their discussion of any reuse schemes. TDRA should be reassessed and the public consulted on what the best value future is in the context of regulatory improvements and option costs for long term future benefits.