In this paper, we ask what investors in Thames Water – and its holding company Kemble Water – would have learned about the level of risk of their investment and its likely market value had they compared its characteristics to market and peer group data.

A large water and wastewater utility like Thames Water epitomises the “stable and predictable” cash flows that investors are attracted by in the infrastructure asset class. Yet, in December 2022, the value of this investment was impaired by almost 30%, an abrupt and unexpected loss of approximately GBP1.5bn (the company was previously valued at c.GBP5bn by its owners) for investors including UK, Japanese and Canadian pension plans. Only nine months earlier, in March 2022, some investors were still increasing the valuations of their stakes.

We show that a straightforward comparative analysis reveals the emergence of a high-risk, low-return profile that should have raised numerous red flags and prompted long-term investors seeking a ‘boring’ investment to reconsider. For a large water utility to lose so much value so fast, the investment must in fact have been mispriced for several years leading up to the impairment. Our own assessment is that its value had indeed been decreasing for years and will likely decline more from the current reported valuation.

Without this analysis, investors fell prey to a form of self-referencing or ‘absolute thinking’ that unfortunately remains very common in infrastructure investment: it’s about the one asset, not the market or peers. This narrow vision can obscure the big picture and the role played by market dynamics i.e., the systematic drivers of the fair market value of private infrastructure companies. Because infrastructure assets are large and illiquid, once invested, it can be hard not to ‘fall in love with your position’ since it is difficult to change easily or quickly. But taken in isolation, a single asset is often more of a story than a hard quantitative assessment.

We argue that benchmarking the key characteristics of the asset would have allowed a much better understanding of its risk profile.