RPI reform: The implications for real assets

The Retail Price Index (RPI) is the UK’s longest running measure of consumer price inflation, dating back to 1947. However, the flaws in RPI – chiefly its upward bias relative to actual inflation due to the way it is calculated – have been widely criticised for more than a decade, with reform seen as long overdue. 

Contracts have gradually shifted away from RPI-linkage to other measures of inflation over a number of years, but legacy contracts linked to RPI are still prevalent and some markets, such as index-linked gilts, remain indexed to RPI.

The flaws in RPI have been the focus of various investigations and consultations, but until 2019 these did not result in any changes.^1,2 Uncertainty around the future calculation of RPI became heightened in September 2019, following the publication of a letter from then-Chancellor Sajid Javid to the UK Statistics Authority (UKSA).

The long-running saga finally reached its conclusion last November when, alongside the Spending Review, the UK government and UKSA published their response to their consultation on future changes to RPI. It was confirmed that RPI will continue to be published as an index but will be aligned with the Consumer Prices Index including owner occupiers’ housing costs (CPIH) from February 2030, with no compensation given to holders of index-linked gilts. CPIH is generally expected to average 0.75 to one per cent lower than RPI.

As we previously commented³, while agreeing with the criticisms of the way RPI is calculated, we believe that the proposed reforms will lead to an unequitable re-distribution of wealth.

Given our view, shared by many other commentators, that such an outcome could not have been the intention, we thought there was a chance the reform would be delayed further or that the calculation of RPI would be changed to CPIH plus a fixed margin. 

The RPI reform will have a varied impact on real assets

One of the benefits of some real assets is the long-term inflation linkage they provide, protecting the real value of investors’ capital against inflationary surprises. This can be derived in various ways, ranging from inflation-linkage in long-term contracts to quasi-inflation linkage resulting from construction costs, wage growth and therefore affordability tending to rise with inflation.

The reform will only impact the revenues of assets with cashflows contractually linked to RPI, so most of the real asset universe will be unaffected. Within real assets, contractual RPI linkage is commonly found in long-lease real estate, income strips, ground rents, subsidised renewables, private finance initiative (PFI) structures and inflation-linked private debt.

For assets in these sectors, the impact on expected returns beyond 2030 is dependent on various factors, including its duration; the proportion of revenues that are RPI-linked; whether there are caps or floors to RPI linkage; whether any costs are also RPI-linked as well as tax and leverage impacts. Figure 1 summarises these factors for each of the main affected real asset sectors.

Figure 1: Real assets: Characteristics and inflation linkage

To date there has not been any noticeable change in the valuation of RPI-linked real assets resulting from the RPI reform announcement, albeit there has been limited transaction activity. Figure 2 compares the change in expected returns for theoretical RPI-linked assets across various sectors.

Figure 2: Expected impact on full life IRR for theoretical assets with RPI-linkage from a 1% reduction in RPI from 2030 (per cent)

The chart shows a relatively minor impact on returns across most infrastructure sectors and reversionary long-lease assets from a reduction in RPI in 2030. For assets with less than 25 years of RPI-linked cashflows, the impact tends to be low, as inflation in the early years is far more important than the latter years. This is not the case, however, for index-linked debt, where cashflows are back-loaded. Income strips and ground rents are also more affected, as all cashflows are RPI-linked and the tenure of these assets is much longer.

Moreover, for assets that derive inflation linkage from private contracts, such as across long-income real estate, there may be contractual protection from the change in RPI. The fact that RPI will continue to be published as an index is likely to mean some RPI protection clauses will not be triggered, depending on the specific wording in each contract. It is essential for asset owners to assess this on a case-by-case basis.

This change has yet to be reflected in asset values

There was relatively minimal change in market-implied RPI following the Spending Review announcement, as observed from index-linked gilt and nominal gilt yields. Long duration-implied inflation had already drifted lower since the risk of reform became heightened in 2019 (see Figure 3); however, the change in market-implied RPI in the early 2030s has only decreased by roughly 0.4 per cent since the end of 2018, far less than the expected wedge between RPI and CPIH (0.75 to 1.0 per cent).

It may have been expected that implied RPI beyond 2030 would fall following the November announcement. However, this has not happened, as shown in Figure 4. The lack of change in pricing of both market-implied RPI and RPI-linked real assets means that expected returns are reduced for both, but the premium over index-linked gilts offered by real assets remains unchanged. 

Figure 3: Evolution over time of market-implied average RPI from 2025 to 2029 and from 2030 to 2034 (per cent)

Figure 4: Market implied RPI from 2025 to 2050, as at various dates (per cent)

The lack of reduction in long-term inflation pricing could partly be due to a growing investor belief that inflation will increase in the long term. However, it is also likely driven by the dominance of liability driven investors in the index-linked gilt market. UK private sector defined benefit pension (DB) schemes had combined liabilities of £1.9 trillion⁴ as at the end of December 2020, compared to the market value of long duration⁵ index-linked gilts outstanding of just £619 billion⁶. This supply-demand imbalance appears unlikely to reverse over the medium term.

These investors are reasonably price insensitive to changes in implied inflation, as both their assets and liabilities are typically similarly affected by changes in inflation pricing. Moreover, many UK DB schemes are de-risking, and looking to increase their exposure to long-duration inflation-linked assets. A lack of supply has long meant that investors are willing to pay a premium for such assets.

Real assets that offer strong inflation linkage are due to benefit from this sustained demand. As such, we do not believe that the impact of RPI reform on expected returns will necessarily be fully reflected in changes in asset values.

Illiquidity premium remains intact

The lack of market reaction to the announcement that RPI would be brought in line with CPIH from 2030 may have surprised many. Despite CPIH averaging 0.75 to one per cent below RPI, market pricing of inflation-linked assets has not significantly changed.

Similarly, the impact on most real asset sectors with RPI-linked cashflows is perhaps less than some expected, and lower than for index-linked debt of similar duration. This is a result of not all cashflows being RPI-linked, the benefit of floors to rent reviews, and tax impacts.

The full extent of protection from the change being embedded into contracts remains to be seen, however, and should be a key focus for owners of such assets over the coming year. Liability driven investors’ appetite for a limited supply of long duration inflation-linked assets appears set to support their value. One thing seems clear: the illiquidity premium offered by inflation-linked real assets over index-linked gilts remains unaffected by the reform.