In its current state, Solvency II presents a barrier to investments in ‘core’ infrastructure by European insurers, write William Coatesworth, Serkan Bahceci and Edward Collinge

The draft proposal for Solvency II paints infrastructure investing with an overly broad brush and misses an opportunity to distinguish between the diverse styles of infrastructure investing that carry very different expected risk-return profiles.

The simplistic approach to the entire infrastructure sector under the draft proposal makes it more difficult for European insurers to access the stable and relatively predictable long-term cash flows provided by infrastructure assets at the lower end of the risk spectrum.

While investment by insurers in this area has been cautious to date, the proposed Solvency II regime removes the quantitative restrictions on investment in specific asset classes, replacing these with the need for insurers to invest in accordance with the prudent-person principle. As a result, increasing numbers of insurers are seeking infrastructure assets as an investment opportunity, with a number of large annuity providers recently announcing that they intend to increase their exposure to infrastructure.

However, as such investments are generally unlisted, an investor making a portfolio allocation decision, will not be able to compare and rank infrastructure assets against other investment alternatives simply by looking at historical rates of return and volatility, since such long-term rate-of-return data do not exist.

Bottom-up approach
An alternative approach to assessing infrastructure assets is bottom-up, determining the rate of growth of cash flows (or cash-flow proxies) by looking at annual operating incomes and costs over long periods of time. This approach allows historical cash-flow performance of infrastructure assets to be compared with the cash-flow performance of other asset classes.

Under current draft Solvency II requirements, the capital requirements in respect of both infrastructure debt and equity are captured under the market risk sub-module of the standard formula.

Due to its characteristics, infrastructure equity would generally be considered as ‘type 2 equity’ under the standard formula and, as such, capital requirements are calculated under a 49% fall in market values (plus or minus any symmetric adjustment).

The capital requirements for infrastructure debt are captured under the spread risk sub-module, irrespective of whether the debt is held in the form of bonds or where insurers are providing investment through long-term loans. Under this, the capital requirements are calculated in relation to the duration and credit rating of the instrument.

Where the infrastructure debt is unrated, the spread risk charge to be applied falls between that for A- and BBB-rated bonds and loans.

On 8 April 2013, the European Insurance and Occupational Pensions Authority (EIOPA) published a discussion paper on ‘Standard Formula Design and Calibration for Certain Long-Term Investments’.

The paper sets out EIOPA’s findings from an in-depth analysis on whether the calibration and design of the regulatory capital requirements for certain long-term investments held by insurers under the envisaged Solvency II regime (as calculated using the standard formula) should be adjusted or reduced under the current economic conditions, without jeopardising the prudential nature of the regime. This analysis covers private equity and venture capital, socially responsible investments (SRI) and social business debt and equity finance, infrastructure project debt and equity, and securitisations of SME debt.

Throughout the paper, EIOPA has highlighted the need to ensure that the standard formula provides an appropriate trade-off between ‘risk-sensitivity and simplicity’ which would need to be considered before introducing more granular treatment of particular asset classes.

EIOPA has highlighted the challenges of performing calibrations for infrastructure equity, which is generally unlisted, and debt, which generally takes the form of loans and as such has no market prices available.

EIOPA concludes that it has found no evidence that “the spread risk for infrastructure project debt... differs significantly from the spread risk of corporate debt with the same rating”.

Interesting findings
Studies looking at historical cash-flow performance of infrastructure assets within the banking sector and comparing these with the cash-flow performance of other asset classes have highlighted a number of interesting findings, including:
• The volatility of infrastructure cash flows is materially lower than those of equities and property;
• Infrastructure cash flows are not highly correlated to those of equities and property;
• The cash flows of infrastructure assets grow faster than the CPI over time;
• Diversification opportunities exist within the infrastructure asset class itself.

Furthermore, while the cumulative credit losses for infrastructure debt approximate those seen for A-rated corporate bonds, figure 2 illustrates that the spreads on infrastructure bonds have been sustained at long-term average levels comparable with BBB-rated corporate bonds, but with significantly lower volatility than exhibited for corporate bonds.

As such, EIOPA’s proposed calibrations may materially overestimate the capital requirements for insurers looking to hold these assets. The use of an internal model to calculate Solvency II capital requirements allows firms to adopt a more granular treatment than set out in the current draft of the Solvency II standard formula. Therefore, standard formula firms looking to hold relatively significant levels of investment in infrastructure may consider moving to an internal model approach to more accurately reflect the risks inherent in this asset class.

This is an abridged version of a white paper produced by JP Morgan Asset Management and Milliman.

William Coatesworth is a consulting actuary at Milliman. Serkan Bahceci is head of infrastructure research JP Morgan Asset Management. Edward Collinge is executive director at JP Morgan