Regulatory financeability tests: early or late warning systems?

This bulletin explains why it is essential for regulators in Australia to adopt financeability tests as standard practice whenever they make a regulatory determination, and to take the results of those tests seriously.

Financeability tests are used by regulators in the UK to assess whether the revenues they allow a business to earn over a price control period will be sufficient to finance the business’s operations efficiently. These tests act as an early warning against a regulated business becoming financially constrained or insolvent—an outcome that  ultimately harms consumers, or taxpayers, who may be called on to rescue a regulated business providing essential public services from collapse.

By contrast, in Australia, regulators have either refused to employ financeability tests or run them but only pay lip service to the results.

A Case Study: The Thames Water ‘financeability’ problem

Thames Water is Britain’s largest water company, providing water and wastewater services to 15 million customers in London and the Thames Valley.

In late June 2023, alarming reports began to circulate that Thames Water was struggling to meet its debt obligations and may be on the brink of collapse. The crisis deepened as the CEO stepped down suddenly following criticisms about the company’s poor environmental performance and record leakage rates, and a new Chair was appointed hastily to regain control of the situation.

The UK Government and the water sector regulator, Ofwat, began drawing up rescue plans for Thames Water amidst fears that the company might become insolvent. The crisis ended when shareholders agreed to inject £750 million of equity capital into the business, to reduce its debt burden.

The media was quick to blame the financing decisions of the company’s private owners for its woes. Previous owners of Thames Water, including Australian bank Macquarie (dubbed the ‘vampire kangaroo’ by some in the British press), were accused of loading the business with excessive debt, while extracting huge dividends, leaving it unable to meet its financial obligations.

The Thames Water debt crisis

Thames Water is one of the most indebted water companies in Britain. At the time of the crisis, it had £16 billion of debt and a gearing ratio (the proportion of its assets that are financed with debt rather than equity) of around 80%. Ofwat noted that Thames Water (and some other companies) had “borrowed too much.”[1] This meant that the company faced large debt repayments.

The situation was worsened by the fact that nearly 60% of Thames Water’s debt was inflation-linked. That is, a significant portion of the interest Thames Water needed to pay was linked to the retail price index (RPI) measure of inflation. The RPI is similar to the consumer price index (CPI). However, it also includes mortgage interest payments, which makes it more sensitive to changes in interest rates. Hence, Thames Water’s interest bill grew significantly as inflation rose sharply over 2022 and 2023.

But the size of Thames Water’s debt obligations is only one part of the equation.

Inflation and the regulator’s revenue limits

As a natural monopoly, the amount of revenue that Thames Water is allowed to earn is capped through periodic price controls by Ofwat.

A part of that revenue allowance is an amount to cover real interest payments. However, to the extent that a company holds inflation-linked debt, its actual interest bill in each year will be the sum of a base (real) rate plus outturn inflation that year.

Ofwat allows the regulatory asset base (RAB) of the businesses to grow each year in line with actual inflation. Recovery of this inflation-indexed RAB, over 50 years or more, provides the business with compensation for the inflation component of  its interest costs.

Herein lies a problem: the business is contractually obliged to pay the entire interest bill every year. But, it will take decades to recover through regulated prices the inflation component of that bill. When inflation starts rising, the mismatch between the business’s regulated cash flows in each price control period and its interest obligations widens.

In short, Thames Water faced a perfect storm: a large pile of debt, rapidly rising interest repayments linked to inflation, and regulated revenues that were insufficient to cover these increasing costs. All of this added up to material risk of default on its debt obligations.

UK regulatory financeability tests: an early warning system

Ofwat has a statutory duty to:

“secure that companies…are able (in particular, by securing reasonable returns on their capital) to finance the proper carrying out of their functions.”[2]

If the annual revenue allowance set by Ofwat is insufficient to pay the business’s interest bill or to attract equity finance, the business will be unable to finance its activities properly or invest in assets. In these circumstances, the business would be unable to deliver services of proper quality to consumers.

Moreover, the financial collapse of essential service providers such as water companies, energy networks or communications firms can cause catastrophic disruptions and economic costs to consumers.

Recognising this, Ofwat and other UK regulators with similar financing duties have developed ‘financeability tests.’ These act as an early warning to detect whether the revenue allowances set by the regulator would be insufficient for an efficient business to remain financeable. If the tests show a likely deterioration in financeability, the regulator can adjust revenue allowances to avert the problem.

Ofwat sets revenue allowances based on a ‘benchmark’ business that uses debt to finance 60% of its assets and maintains a BBB+ credit rating.[3]

The purpose of Ofwat’s financeability test is to assess whether the revenue allowances set in this way are in fact sufficient to support that BBB+ rating at the benchmark 60% gearing. It is effectively a test of the internal consistency of the regulatory decision.

Ofwat and Thames Water

Ofwat decided in its 2019 revenue determination for Thames Water that it needed to accelerate the recovery of costs by £125 million in order for a benchmark business in Thames Water’s circumstances to maintain a BBB+ rating at 60% gearing.

Thames Water argued that, even after bringing forward these allowed revenues, a benchmark business with 60% debt finance would be unable to finance the delivery of its statutory obligations or attract the required amount of equity finance. Ofwat did not accept Thames Water’s representations.

Ofwat also performed a separate test of ‘financial resilience’ based on Thames Water’s actual gearing (at the time) of 80%. Ofwat expressed concerns that Thames Water was very highly geared and noted the company should take steps to improve its financial resilience.

Under this two-pronged test:

  • Any problem found using the regulator’s benchmark settings is a problem of the regulator’s making and requires a regulatory remedy; and
  • Any problem that is due to the actual business departing from the regulatory benchmark is the responsibility of the company and its owners to fix. Consumers should not be expected to pay more for poor financing decisions taken by the business.

Australian regulators seem to prefer a late warning system

Whereas Ofwat and Thames Water differed in their views about whether the regulatory allowance was sufficient to support financeability, there was strong agreement that consideration of financeability is a vital component of regulatory best practice.

By contrast, Australian regulators have been slow to embrace financeability tests. Regulators in Australia fall into three groups:

  • Those that perform no financeability test at all;
  • Those that perform a test, but invariably conclude that no action is needed regardless of the outcome of that test; and
  • Those that perform a test and conclude that action is needed – but not by them.

The Thames Water case provides a timely reminder that even the largest and most well-resourced regulated businesses can face serious risk of insolvency, which can be disastrous for consumers. Properly implemented financeability tests can avoid such outcomes. Therefore, it is important to consider:

  • The essential features of a useful financeability test; and
  • The right regulatory responses to the outcomes of financeability tests. (Spoiler alert: The right regulatory response is not to always do nothing.)

Features of a good regulatory financeability test

Financeability tests are an essential element of best practice regulation

It is in the long term interests of customers that an efficient provider of essential services is financeable, so that it can provide services desired by customers. This requires that regulated revenues are sufficient for an efficient business to meet its financial obligations as they fall due and to invest what is needed to deliver regulated services.

If regulated businesses cannot provide their services because they are financially constrained or insolvent, then customers or taxpayers will bear the cost. If the regulated business is financially constrained, it may be unable to make the investments needed to provide the level of service today or into the future that customers want. If a regulated business becomes insolvent, then it will likely be taxpayers that step in to ensure the essential services continue to be provided.

This is why financeability tests are a key part of best practice regulation. The purpose of a financeability test is to ensure that the revenue allowance is sufficient for an efficient business to meet its financial obligations over the forthcoming regulatory period.

The test should also allow the regulator to diagnose the source of the problem, so that appropriate corrective action can be taken.

A two-pronged test is needed to determine appropriate action

Financeability problems can arise because:

  • The regulatory allowance is too low to support the benchmark financing parameters (gearing and credit rating) that the regulator assumed when setting that allowance; or
  • The regulated business has departed from the benchmark financing parameters (e.g., by adopting higher gearing).

The first is a case of internal inconsistency in the regulator’s process and is for the regulator to resolve. The second is a case of risky financing practices and is for the firm and its owners to resolve.

Regulated businesses should be free to depart from the regulatory benchmark if they choose. The business then keeps any benefits, and bears any costs, of such a departure. Neither the regulator nor consumers should be called on to immunise regulated businesses against bad outcomes arising from such decisions.

The Ofwat approach described above involves a two-part test:

  • The ‘financeability’ test is performed using the regulator’s benchmark financing parameters. The test provides an indication of whether the regulatory allowance is sufficient to support those parameters.
  • The ‘financial resilience’ test is performed using the firm’s actual financing parameters. It assesses whether the firm’s departure from the regulatory benchmark may be causing problems.

A similar two-pronged test is applied by the NSW Independent Pricing and Regulatory Tribunal (IPART).

The source of any financeability concern can be identified more easily by using two separate tests:

  • If a regulated business fails IPART’s ‘benchmark’ test, that implies that the revenue allowance is too low to service what the regulator itself considers is the efficient debt obligation; and
  • If a regulated business fails IPART’s ‘actual’ test but passes the benchmark test, that implies that the business is not financially resilient due to its own financing choices.

Recent financeability test practices in Australia

There are almost no examples of Australian regulators taking action to address financeability concerns arising from their decisions to set inadequately low revenue allowances. Some regulators perform no test at all, so have no way to detect the existence of a problem. But, many that do perform financeability tests have chosen to take no action, even when the test has clearly identified a financeability problem.

‘Narrating away’ the outcomes of a financeability test

It is important that regulators acknowledge the outcomes of their financeability tests, respond consistently and do not try to apply a narrative that seeks to ‘explain away’ a potential problem.


The most important financeability metric considered currently by rating agencies for Australian regulated energy and water businesses is the funds from operations to net debt (FFO/Net Debt) ratio

In IPART’s 2020 determination for Sydney Water, the FFO/Net Debt ratio fell below of the minimum 7.0% threshold specified in IPART’s benchmark test in every year of the regulatory period. This suggested strongly a failure of the benchmark test. IPART should have adjusted Sydney Water’s revenue to ensure an efficient, benchmark business could remain financeable.

However, IPART concluded that the failure of the test on the FFO/Net debt ratio did not indicate a problem. This is because, amongst other reasons, that ratio was expected to improve over the regulatory period towards the target ratio (from 6.6% in 2020-21 to 6.8% in 2023-24). However, it would remain below the required threshold in every year.[4]

According to IPART, there was no problem to fix because the failure of the test was forecast to become less severe over time.

Australian Energy Regulator (AER)

In 2022, the AER undertook a major review of its methodology for setting the allowed rate of return for electricity and gas networks. As part of that review, the AER presented a simplified financeability test to assess the impact of its proposed methodology.

The AER’s test used the FFO/Net debt ratio as the sole metric and set a threshold of ≥7.0% for a ‘pass’.

The AER found that the FFO/Net debt ratio for 25% of the businesses fell below the threshold for a pass.

The AER concluded that there was no evidence of a financeability concern (at the industry level) under its proposed rate of return methodology because 75% of the firms appeared to pass the test.

Of course, this ignored the fact that a quarter of the industry had failed the regulator’s own test.[5]

Not adjusting regulatory decisions in response to a failure of the benchmark financeability test and instead narrating away the outcome is as good as having no test at all.

Applying the wrong solution to the problem

The regulatory response to a financeability problem should properly address the underlying cause of the problem. Without addressing the cause, remedial actions may be misdirected, leading to inefficient outcomes.

If the financeability problem was caused by the business taking imprudent or risky financing decisions (e.g., by gearing up more than the benchmark level), it should not fall to consumers to ‘bail out’ the business for those poor decisions. That would impose unnecessary costs on consumers and create poor incentives for the business to avoid bad financing decisions in future.

However, if the financeability problem was caused by the regulatory allowance being set too low for a efficient business to remain financeable, then the regulator, not shareholders, should fix the problem.

IPART has made this distinction clear:

“If the source of the concern is that prices are too low even for a benchmark efficient business, we think the appropriate remedy is to review our pricing decision. In essence, this step would involve correcting a regulatory error. The financeability test could help identify any such error by applying additional information that may not have been available in the building block model used to set prices.

If the source of the concern is that prices are adequate for a benchmark efficient business but too low for the actual business because its owners have been imprudent or inefficient, there are appropriate remedies. The owners could reduce the business’s level of debt by injecting more equity, accept a lower than market rate of return on their equity, or both. It is an important principle that an inefficient business should not be rewarded for its imprudent decisions at the expense of customers.”[6]

The best way to diagnose the source of a financeability problem is to apply a two-pronged test of the sort used by IPART and Ofwat. Because the benchmark test assumes that the regulated business always finances itself efficiently and prudently, the only explanation for a failure of that test would be a regulatory error that requires correction.

Calibrating the financeability tests incorrectly

Even when a test is structured correctly, it will only be useful if it accurately reflects the financial flows of an efficient benchmark business. If the thresholds for passing the test are set unrealistically low, then genuine financeability tests will go undiagnosed and uncorrected.

For example, when running its benchmark test, IPART assumes that businesses faces real, rather than nominal, debt obligations. But, in reality, companies in Australia issue nominal debt and therefore must pay nominal interest expenses.

Assuming that an efficient regulated business faces lower (i.e., real) interest expenses than it in fact does makes IPART’s benchmark test too easy to pass. Hence, the test is prone to ‘false negatives’—i.e., conclusions that there is no financeability problem, when there is one.

Recommendations for improvements to regulatory financeability tests

These examples show that regulators in Australia either do not apply financeability tests at all, or if they do, typically find reasons not to fix a problem identified by their own financeability tests.

The Thames Water example is a reminder that regulated businesses can face serious financeability problems.

Ofwat concluded at the last price control that Thames Water needed an uplift in allowed revenues of £125 million to support an efficient credit rating, at an efficient level of gearing. The crisis that Thames Water faced would probably have been worse than it was, had Ofwat adopted the approach that Australian regulators follow and ignored the results of its own financeability analysis.

Because regulated businesses deliver essential services, it is consumers that suffer if these businesses become insolvent or financially constrained. In the case of insolvency, taxpayers may also be called upon to rescue the business from collapse.

Therefore, regulators in Australia should view financeability tests as an essential part of best practice and incorporate them as a standard feature into their revenue setting processes.

A good framework for regulatory financeability tests should:
  • Be able to identify the source of a financeability problem.
If the cause is imprudent or risky financing decisions by the business, it should be left to investors to deal with. However, a failure of a benchmark test indicates that the problem is a regulatory one. In these situations, the regulator should take action to fix the problem, rather than explain it away or shift the onus on shareholders.
A two-pronged test allows regulators to pinpoint the cause of the problem and ensure that tailored action is taken.
  • Be specified in such a way that the outcome—i.e., a ‘pass’ or a ‘fail’—can be identified clearly and objectively.
Regulators should take the outcomes of these tests seriously and not dismiss a clear failure of the test by simply assuming that their regulatory decision is adequate.
  • Be calibrated in such a way that the tests are realistic and capable of identifying, rather than masking, genuine problems.

To find out more about how we advise on regulatory issues and financeability, pricing reviews, and more, please reach out to our team.

[1] Ofwat, Thames, debt and water sector finance, 24 July 2023. [2] UK Legistation, Water Industry Act 1991, section 2(2A)(c). [3] Ofwat, PR19 determinations: Thames Water final determination, December 2019. [4] IPART, Review of prices for Sydney Water from 1 July 2020, Final Report, June 2020, p 178. [5] AER, Draft rate of return instrument, Explanatory statement, June 2022, p. 24. [6] IPART, Review of our financeability test, Issues paper, May 2018, p. 32


Frontier Economics Pty Ltd is a member of the Frontier Economics network, and is headquartered in Australia with a subsidiary company, Frontier Economics Pte Ltd in Singapore. Our fellow network member, Frontier Economics Ltd, is headquartered in the United Kingdom. The companies are independently owned, and legal commitments entered into by any one company do not impose any obligations on other companies in the network. All views expressed in this document are the views of Frontier Economics Pty Ltd.

Start typing and press Enter to search