Inflation Linked Debt: Analysis of Local Authority Bond Issuances
In the current environment of ongoing cuts to central government funding for local authorities, local authorities are increasingly looking to use capital spending to create revenues to fill this spending gap. For example, it has recently been reported that local authorities are likely to complete £1bn of commercial property deals by the end of the year, with real estate just one of a number of areas in which councils are increasingly investing. As such, it is no surprise that Local Authorities are looking to diversify their sources of funding away from PWLB and are increasingly looking to the capital markets to provide more flexible funding that better matches the profile of the inflation linked revenues that these investments will create. The most recent example of this is the Aberdeen City Council bond issue, which was linked to RPI.
Pricing of Inflation-Linked Debt
Due to the good credit quality of local authorities and the high demand for inflation linked products from insurance companies and pension funds, it is possible for councils to obtain funding that is significantly cheaper than PWLB by issuing inflation linked debt through the capital markets. However, in assessing the cost of index-linked debt, it is important to account correctly for features of the bonds that can differ from issue to issue. Local Authorities should make sure that they are aware of the different effects that need to be taken into account.
Below we have analysed the 3 inflation linked bonds issued by UK Local Authorities in the last 18 months:
Caps and Floors
The outstanding principal on an inflation linked bond will increase/decrease in line with the relevant index in a given year. As such, interest payments as well as the amount that will have to repaid at maturity are linked to inflation.
A cap or floor included in the terms of a bond has the effect of limiting the annual change in the underlying inflation index that can be passed through to the indexation of the bond payments. For example, the 3% cap on the Warrington Borough Council bond limits the indexation of the bond to only 3% in any given year, regardless of how high is the annual increase in the CPI Index. This is a valuable feature for the issuer of the bond. Correspondingly, a 0% floor prevents any annual decrease in the indexation of the bond in a scenario of negative inflation, and is thus the inclusion of a is a cost for the bond issuer.
In Warrington’s case, the value of the indexation collar (the cap and the floor together) was at the time of issue a net benefit to the issuer of 0.24% in yield. The fact the collar provides a net benefit to the issuer is intuitive since, given market expectations of future inflation, the 3% cap is more likely to come into effect than the 0% floor on average over the life of the bond.
In the case of Aberdeen, the 0% floor provides a dis-benefit to the issuer valued at 0.30% of yield. This reflects the scenarios where inflation is negative but in which the issuer will be required to make payments that are unchanged in nominal terms.
CPI vs RPI
The difference between CPI and RPI is of notable importance and crucial to assessing the cost of index-linked debt issuance. CPI is used as the reference for inflation-targeting by the Bank of England, and the medium-term target for CPI annual increases is currently set at a 2%. The older RPI index has now lost its status as an official national statistic, but continues to be used in the determination of many financial instruments including inflation-linked gilts.
There are methodological and compositional differences between CPI and RPI which mean that, on average over time, increases in CPI are expected to be less than those in RPI. The Office of Budget Responsibility (OBR) projected a long-term difference (often termed the “wedge”) of 1.4% in 2011. As of March 2016, the OBR revised this forecast of the long-term wedge to 1.1% based on latest evidence of fiscal changes. To be conservative, an assumption of 1.0% has been used in the table above analysing the bonds at the time of their issuance.
RPI-linked gilts are the benchmark gilts which are used for pricing inflation linked debt. Where a council issues a bond that is CPI-linked, rather than RPI-linked, the total spread over the gilt will incorporate an assumption on the long term difference between CPI and RPI. As CPI is predicted to be less than RPI in the long term, the principal of a bond linked to RPI is forecast to increase at a faster rate than that of a CPI linked bond. As the underlying principal for a RPI-linked bond, and therefore the interest payments, increase at a faster rate, the coupon will be less.
As such, when comparing spreads over the benchmark gilt of inflation linked bonds a long term forecast of the wedge should be subtracted from the apparent spread of a CPI-linked bond. The forecast wedge in 2015 was 100bps and this has been used in the table above analysing the bonds at the time of their issuance.
Inflation linked bonds can offer very good value for money to councils and produce significant savings relative to PWLB funding, but the pricing of the bonds needs to be properly understood.