Commercial real estate finance: opportunities for insurers

4 minute read

The low yield world – where around £7trn ($9.22trn) of bonds have negative yields¹ – has propelled insurers to look more closely at alternatives to core fixed income. In this environment, commercial real estate finance has distinctive advantages, with predictable returns comfortably above UK gilts, combined with the security of recourse to the underlying property and covenant protection for lenders.

The relative security of these assets is reflected in their capital treatment under Solvency II regulations, giving them an edge at a time when the insurance industry is focused on improving capital efficiency.  

Insurers are already active market participants, holding around 14% of drawn-down debt in the UK’s £166bn commercial property lending market, and originating over £1.5bn of new lending in the first half of the year.² But there are further opportunities available, outside the interests of those building annuity books, given the way the changing regulatory environment has forced banks to rein in their activities.   

Real estate finance: market overview

For those assessing opportunities, it’s important to be mindful the UK is in the late stages of the property cycle. Real estate values have been on a mainly upward trajectory since the financial crisis, although experiencing some weakness recently associated with Brexit uncertainty. In some cases, real estate income return has fallen below the real estate debt yield, and five-year total return forecasts for direct investments in real estate equity are modest. 

This is supportive of a debt-oriented approach. Taking real estate exposure suitably high up in the capital structure – in senior debt with moderate loan-to-value ratios – recognises the potential downside for capital values in the uncertain Brexit period and as interest rates normalise.

Financing prime assets is competitive; recent indicative pricing includes funding for prime UK regional space as low as 130 basis points (bps) over gilts or 120bps in central London. In our view, there may be better opportunities in investment grade debt that falls outside banks’ appetite – for example, in secondary markets or regional assets with high quality sponsors. Indicative spreads are shown below.³

Real estate finance: indicative spreads

Attractive opportunities for insurers

There are a number of areas where insurers can find attractive opportunities. One is in providing longer-dated financing than typically provided by banks, with terms that might stretch beyond 10 years. There are also opportunities in the five- to 10-year space, which might provide yield pick-up relative to shorter-dated real estate finance deals.

For insurers, these assets could be used in long-term savings business, such as with-profits, or to back long-dated general insurance liabilities. Alternatively, these assets could be used within shareholder funds, particularly as assets backing the on-going capital requirements of the business.

Insurers writing annuity business in the UK are limited to a particular sub-set of opportunities. This is due to the restrictions imposed by Solvency II’s matching adjustment (MA) framework – in particular, the requirement for protection against prepayment risk. Here, significant demand for assets with the appropriate protections has led to spread compression.

Insurers are also looking at broader opportunities within real-estate backed debt, such as long-lease commercial real estate (including student accommodation and social housing), ground rents and equity release mortgage loans. 

Finding illiquidity premia in private markets

When investing, insurers need to be able to assess relative risk and return across both public and private markets. It is important that potential investors recognise that premia from non-traditional investments vary considerably. They change over time, and are not immune to developments in the public markets.

Measuring premia is not necessarily straightforward, as most alternative assets do not have a close parallel in public markets. Nevertheless, a pragmatic approach utilising similar publicly-traded bonds can give an indication of the additional illiquidity premia available in real estate finance.  

Drilling down in this way shows premia for real estate debt at low levels immediately following the financial crisis. By 2012, the situation had reversed, with premia reaching almost 200bps over comparable liquid credits as banks were deleveraging, withdrawing some longer duration facilities, and spreads on public credit reduced. Over the past few years, the illiquidity premium has fallen back again, but the trend now appears to be stabilising.

Real estate finance: illiquidity premia over time 

Considerations for insurers under Solvency II

The underlying premise of the Prudent Person Principle within Solvency II is that insurers need a thorough understanding of their investments. For real estate finance, this includes considerations around fair valuation, credit assessment and capital modelling. It also requires insurers to assess their own expertise in managing potential default events – for example, the breach of loan-to-value or income cover ratio covenants.

In the UK, these covenants provide significant value to lenders, who can use their ability to revalue assets annually or bi-annually, and call up loans in breach of loan-to-value covenants. This differentiates the UK market from the US and a number of other European countries.

When it comes to credit assessment, insurers or their asset managers need robust governance in place around the internal credit rating processes. This would include ensuring that the team responsible for origination is independent from the team(s) responsible for credit assessment methodologies and carrying out the assessments themselves.    

For UK insurers using the MA, the internal credit rating (and mapping to Credit Quality Step) has a direct impact on the valuation of liabilities. In light of this, the Prudential Regulation Authority has set out its expectations for illiquid assets within MA portfolios in a recent supervisory statement.⁴

Given the need for ‘one size fits all’ treatment, the standard formula solvency capital requirements⁵ for real estate finance are inevitably quite broad brush. For instance, the standard formula suggests the same capital treatment for a 75% loan-to-value loan and a 50% loan-to-value loan (shown below). Charges are higher than those recently approved for A-rated infrastructure corporate bonds, but below those for BBB-rated infrastructure corporate bonds. This assumes that the underlying real estate assets meet the collateral requirements set out in the Solvency II framework – if not, the capital charge would double, implying a level of risk above that of a BBB-rated credit.

Comparing regulatory capital charges across asset class under the standard formula           

A more finely tuned, risk-sensitive approach is possible within internal models. Insurers need to take account of the time and cost involved in developing these models, which require regulatory approval. However, the key advantage should be an outcome that takes a more sensitive view of risk, allowing insurers to adopt a more targeted investment strategy, particularly in respect of comparatively secure assets with low LTVs.

Conclusion

Current interest in real estate finance reflects the attractive yield, diversification benefits, security that comes with recourse to underlying assets, and subsequent favourable capital treatment under Solvency II. This is presenting something of a sweet spot for insurers at a time when banks have been forced to rein in the scope of their activities. But, with risk at front of mind, recognising that the UK property cycle is advanced suggests that investing high up in the capital structure might be prudent. 

This article originally appeared in Insurance Asset Risk. 

References

¹ Source: Financial Times. 22 August 12 2017

² Source: DMU. Mid-year summary H1 2017

³ Source: Aviva Investors. 30 September 2017

⁴ http://www.bankofengland.co.uk/pra/Pages/publications/ss/2017/ss317.aspx

[5] Based on calculating Value-at-Risk to a 99.5% confidence level over a 1-year time horizon.