Ofwat published its final determinations (FDs) for the England and Wales water sector at 07.00 on 16 December 2019. There are many documents and a lot of detail that the industry will need to pore over in order to understand the implications. Here we explore the three main contentious issues in the sector: finance, cost efficiency and outcome delivery incentives (ODIs). We take an initial look at what has changed since the draft determinations (DDs), published earlier in the year. It will now be for the water companies to assess the FD proposals and ascertain whether they can live with the settlement or whether they will choose to refer the package to the Competition and Markets Authority for further scrutiny. Companies have two months to decide.
Ofwat published its final determinations (FDs) for the England and Wales water sector at 07.00 on 16 December 2019.1 There are many documents and a lot of detail that the industry will need to pore over in order to understand the implications.
This briefing note examines the three main contentious issues in the sector: finance, cost efficiency and outcome delivery incentives (ODIs). We take an initial look at what has changed since the draft determinations (DDs), published earlier in the year.2
Cost of capital
The cost of capital used in the FDs is 5.02% in nominal terms (2.96%, CPI-real; 1.96% RPI-real). This represents a 23bp reduction from the DDs.
The cost of equity estimate has reduced by 29bp, and the main driver of the reduction is the decrease in the risk-free rate. Ofwat has retained the same estimate of asset beta from the DDs (0.36). The beta estimation methodology now also looks at betas estimated over a five-year period, compared with the focus on two-year betas in the DDs.
The cost of debt estimate has reduced by 20bp, which is almost entirely due to a decrease in the yields on benchmark corporate bond indices. Ofwat now assumes companies will outperform the benchmark indices by 15bp when issuing new debt instead of the 25bp that was assumed in the DDs.
Portsmouth Water and South Staffordshire Water have both passed the three tests set by Ofwat to qualify for a company-specific adjustment to the cost of capital. Ofwat has increased the cost of debt allowance for these companies by 33bp.
The FDs make extensive use of higher pay-as-you-go (PAYG) and regulatory capital value (RCV) run-off ratios to bring forward to AMP7 the cash flow that would have materialised in future price control periods. Ofwat reports that the FDs advance revenue for 12 companies.
On a notional basis, Ofwat calculates that after these adjustments all companies will have an Adjusted Interest Cover Ratio of 1.5x or higher.
Ofwat also assumes that companies with RCV growth higher than 10% will reduce the dividend yield below the 3% modelling assumption so that credit metrics remain consistent with a BBB+/Baa1 credit rating.
Sharing gearing outperformance
The FDs introduce a glide path for the threshold level of gearing that triggers additional payments to customers. Starting in 2020–21, companies with gearing higher than 74% will make additional payments to customers. The threshold reduces by one percentage point each year, reaching 70% in 2024–25.
Overall, Ofwat’s efficiency challenge on companies’ costs (as submitted in their business plans) has reduced from 11.3% on TOTEX at the DDs to 5% at the FDs. However, much of this reduction has been driven by companies reducing their planned cost levels by 4.8% since the DDs, while Ofwat has only increased its allowance by 2% for TOTEX (0.9% for BOTEX and 8.4% for enhancement expenditure). As enhancement expenditure is examined on a more bottom-up basis, and companies have provided more evidence in this area, it is not unsurprising that the enhancement expenditure allowance has increased more than for BOTEX.
Key changes since the DDs
While Ofwat has not changed its model specifications since the DDs, the change in the efficiency challenge has been driven by a number of factors, some of which offset each other. The key changes include the following.
- Ofwat has now modelled costs using 2018–19 data—a relatively high cost year compared with the historical period over which the modelling has been undertaken. This increases the predicted cost level by around 6% in water and 1% in wastewater.3
- Ofwat has made its catch-up efficiency assumption more challenging, using the fourth-placed company in wholesale water and the third-placed company in wholesale wastewater as the efficiency benchmark, rather than the upper quartile company. This reduces the predicted cost level by around 0.8% for water and wastewater services.
- Ofwat has reduced the frontier shift challenge from 1.5% p.a. to 1.1% p.a., which is more aligned with precedent. This brings the net frontier-shift challenge to around 0.6% p.a.;4 although Ofwat has extended the frontier-shift challenge to all wholesale base costs5 and to the environmental programme and new meter installations. In terms of impact on the cost allowance, this completely offsets the reduction in the frontier-shift assumption.
- In its DDs, Ofwat moved some enhancement expenditure into its BOTEX models, without changing the model specifications. A number of companies, notably those in regions with high forecast population growth, criticised this approach. In response, Ofwat has introduced an upward or downward adjustment to its BOTEX allowance for companies with a high or low forecast of population growth. This results in a small overall increase in the BOTEX, with the companies in high-growth regions benefiting by up to 1.3% and those in low-growth regions having their BOTEX allowance reduced by up to 1.1%.
- Ofwat now uses companies’ forecasts of water treatment complexity for setting base cost allowance, and has updated its forecasts for properties and weighted average density with the latest available data. This is particularly important for companies that are expecting significant increases in treatment complexity over 2020–25 and increases the allowance by 0.7% for the industry as a whole.
- At the DDs, Ofwat allowed additional enhancement expenditure funding for leakage if a company was forecast to perform beyond the forecast 2024–25 upper quartile level on both leakage per property and leakage per kilometre of main. In their representations, some companies pointed out that these two drivers are dependent on a region’s sparsity/density. In its FDs, Ofwat has created a composite metric based on both measures. As a result, one more company has received an allowance for leakage enhancement and the allowances have increased for all other companies.
- For residential retail costs, Ofwat now assigns a 75% weight to results from its total costs models relative to its bottom-up models (25% weight). This increases the efficient TOTEX allowances by 0.3% (c. £10m) compared with the DDs approach, where total costs models and bottom-up models had the same weight.
In the DDs, Ofwat set out its initial position on performance commitments (PCs) and ODIs. In their representations, a number of companies argued that these were too stretching, and that they exposed the industry to significant downside risk.
In the round, the DDs represent a significant movement compared with the DDs: reduced stretch, greater protections, and more symmetry in ODI downside/upside risk. However, each company’s situation will be different.6
In exploring what has changed since the DDs, of particular interest is Ofwat’s approach to ‘common’ and ‘comparable’ PCs (and the associated ODIs), which cover:
- upper-quartile (UQ) measures—supply interruptions, pollution incidents and internal sewer flooding;
- reducing water demand—leakage and per capita consumption;
- statutory measures—the compliance risk index (CRI) and treatment works compliance;
- asset health measures—mains repairs, unplanned outage, sewer collapses, external sewer flooding, sewer blockages, water quality, and low pressure;
- resilience measures—risks of sewer flooding in a storm and severe restriction in a drought;
- vulnerability measures—the priority services register;
- customer experience—the customer measure of experience (C-MeX) and the developer services measure of experience (D-MeX).
For the three UQ measures, some companies had argued that Ofwat’s use of a combined historical and future UQ target led to over-stretching targets, compared with PR14 (in which historical data only was used to assess the UQ). In the FDs, Ofwat has stuck to its DDs position on this issue, pointing out that companies have outperformed the historical UQ target.
However, Ofwat has reduced the stretch target for water supply interruptions. At the DDs, the PC for 2024–25 was set at three minutes (the forecast UQ), with a glide path. However, Ofwat now considers that the forecast UQ is not an appropriate expectation for the sector by 2024–25, and has adjusted the 2024–25 target to five minutes, with an amended glide path in the first four years. In addition, at the FDs, underperformance payment rates are now symmetrical with outperformance payment rates, so as to ‘provide a more balanced spread of incentives and risks’.
In the case of internal sewer flooding, Ofwat has reduced the downside ODI payments for three companies it regards as performing poorly, by introducing collars for the early years of 2020–25. Moreover, ODI payment rates are now symmetrical (again, having previously been downward-skewed).
On leakage, Ofwat has made the leakage PC less stretching for some companies (while there remains a requirement for a minimum 15% improvement). In addition, for companies that are already performing very well on leakage, Ofwat has recognised that these may find it more challenging to reduce leakage further. The regulator has provided some additional funding for enhancement expenditure that takes leakage beyond the forecast UQ—although whether this is enough will need to be assessed by each of the companies concerned.
Water quality compliance
On the CRI, at the FDs Ofwat has amended the deadband to a score of 2.00 throughout the 2020–25 period, in part to allow more flexibility in performance to take into account the uncertainty created by the ban on the use of metaldehyde (which has been overturned by the High Court).
On asset health, Ofwat has reduced the stretch on mains repairs for all companies in all years. This is to allow all companies the flexibility to deliver the reduction in leakage described above, by allowing more flexibility in the earlier years to use proactive mains repairs as a means of doing this. The underperformance payment rate on mains repairs for all companies has been set to the industry average, to provide a more balanced spread of incentives and risks.
Caps and collars
Ofwat has also applied caps and collars to financially material and/or highly uncertain PCs and allowed caps and collars on other PCs where company proposals are supported by high-quality customer engagement. Where the vast majority of companies have caps and collars on a common PC, Ofwat has applied caps and collars to all companies.
Cost–service quality link
Interestingly, Ofwat also discusses the links between cost efficiency performance and service performance. A number of companies had argued that there was a trade-off between the two and that it was not possible to deliver UQ performance on both.
In the FDs, Ofwat presents its own new evidence. While the regulator notes that it does not expect a company to be good at everything, it presents evidence which, in its view, shows that companies that perform well on costs are also good performers on service. This is through various higher-level heat-map analyses of rankings, and is worth further analysis.
It will now be for the water companies to assess the FD proposals and ascertain whether they can live with the settlement or whether they will choose to refer the package to the Competition and Markets Authority for further scrutiny. Companies have two months to decide.
3 This is an approximation as it is based on the difference between the benchmark estimated over 2014–18 and the benchmark estimated over 2015–19 and does not account for the change in the estimated coefficients.
4 Ofwat has slightly amended its estimation of real price effects; however, the impact is very small. Ofwat has continued to make an allowance for labour costs, but has increased the proportion of costs accounted for by labour costs from 35% to 39%, such that its real price effects assumption is very slightly higher by 0.04%.
5 That is, the frontier-shift challenge is now also applied to unmodelled costs, such as business rates, Traffic Management Act costs and abstraction charges.
6 What follows is based on our review of the following supplementary FD documents: Ofwat (2019), ‘PR19 final determinations: Delivering outcomes for customers policy appendix’, December; and Ofwat (2019), ‘PR19 final determinations: Overall level of stretch across costs, outcomes and allowed return on capital appendix’ December.