31 August 2016
In the second part of this InsuranceERM/Insight Investment roundtable, participants discuss some practical aspects of investing in illiquid assets: valuation, managing liquidity, putting the money to work and fee structures
- Michael Leonard - Head of Insurance Solutions Group, LV=
- David Leach - Deputy CRO, Guardian Assurance
- Philip Howard - Senior Advisor, Beazley
- Gareth Quantrill - Group Investment Manager, RSA
- Scott Robertson - Head of Financial Management Group, Phoenix Group
- Heneg Parthenay - Head of Insurance, Insight Investment
- Simon Richards - Head of Insurance Solutions, Insight Investment
Chaired by Christopher Cundy - Managing Editor, InsuranceERM
Chris Cundy: Valuing illiquid investments has been a perpetual challenge for insurers. Is it becoming any easier?
Scott Robertson: We are in an environment where there is a drive towards market consistency, which means holding the assets at fair value on the balance sheet. What is the fair value in the context of an illiquid asset that seldom if ever trades?
This is the wrong question, because if you are holding illiquid assets on a buy-and-hold basis backing similarly illiquid liabilities, then the fair value of the asset is not relevant from an economic point of view. The right questions to ask are how certain are the future cash flows and the probability of default, and what actions are available to mitigate these risks.
David Leach: If you have a price at the outset, it is not uncommon to use an approach which finds some related indices that you can track and give an idea as to how the value of your illiquid asset would have moved. But again, the asset manager is closest to these assets and best-placed to value them. In the end, price verification raises questions that you can pose to the manager – this discussion is possibly the most useful aspect of the exercise.
Gareth Quantrill: Yes, we use much the same approach. The difficulty will arise when you have an asset that is not performing, because you do not have the same index proxy for non-performing loans.
Michael Leonard: Sometimes there is inconsistency from the regulator. For example, they have pushed back against our process for pricing equity release assets, but the same process is acceptable for a commercial mortgage loan portfolio.
Heneg Parthenay: Can the information you have on illiquid credit for reporting be used to help with valuation? There may be no market price, but you have a lot of information about the loans underlying the assets.
Philip Howard: Yes. We get a quarterly report from our managers and we like there to be an independent valuation component. Once a year we have a detailed audit. We have the right to query anything – and we do.
"There is some risk around whether your liabilities are actually as illiquid as you think they are."
Chris Cundy: How can you create liquidity in an illiquid portfolio?
Gareth Quantrill: There are two elements. First, if you go into illiquid credit, that you have to prove that you can afford to be in illiquid anything. So we look at our peak insurance risks and cover those with high-quality liquid assets, that gives us the freedom that we can apply to illiquids.
The second is just about laddering. I can create liquidity by ensuring that not all of my illiquids are ten-year lock-ins, but I have some amortised structures: fund investments over different periods, drawdowns within funds and harvesting periods which are diverse.
Simon Richards: Agreed - it is about creating liquidity through the cash flows arising from those investments, rather than the direct liquidity of the investments themselves.
Gareth Quantrill: ... which is the only true liquidity. We only count government bonds as 'high-quality liquid assets' when we do our analysis.
Philip Howard: There are other ways to think about this. If you invest in student housing, for example, you want to know how you can sell it. Most of the time there is a good market in blocks of student properties, but sometimes there is not and so you could set up a strategy to securitise it, or structure it into a REIT [real estate investment trust], and get to a critical size where you can finance it in the public market.
David Leach: If you think about the full run-off of a big annuity book, your assets may not be as long a duration as the liabilities. You may in fact have an excess of liquidity in the earlier years. Duration can be extended using interest-rate swaps, but then you may end up being exposed to interest rate rises and you may need to post collateral.
There is also some risk around whether your liabilities are actually as illiquid as you think they are. The recent political intervention in the annuity market is one example.
Michael Leonard: As an institution, our main concern on illiquidity is regulatory i.e. a change in the regulatory environment that means a significant part of your portfolio is now penal to hold and, if it is illiquid, you cannot get out of it. That is something you cannot hedge against.
Scott Robertson: It is not just about your liquidity position today, but also about what the position will be in 15 years' time. You have to be careful that you do not "chase yield", because you could get a situation in your projection where in 2032, for example, you are stuck with 80% of your book in illiquids, and then you are more exposed to liquidity risk due to the variance in liability cash flows.
"You have to be very careful not to pay fees on committed money as opposed to actual money invested"
Michael Leonard: Regulation has created illiquidity in the market and it is stoking up a problem for ten or 15 years' time. For example, the [Solvency II] matching adjustment portfolio is buy and hold. You cannot lend securities without guaranteeing you can get those specific securities back, so you are not even allowing the banks to have repo to trade these positions. Without constant issuance of liquid assets, what are you going to replace it with? You will end up in 2035 with a portfolio that is 80% in equity release, 15% in commercial mortgages and 5% in cash.
Scott Robertson: The matching adjustment rules in respect of equity release have institutionalised illiquidity. We have to buy-and-hold the restructured equity release assets or else lose the matching adjustment benefit. This makes a secondary market in equity release mortgages very difficult indeed.
Michael Leonard: All it does is increase default risk, because once the market is institutionalised, the only people who are going to buy an asset that has tanked are the hedge funds.
"When we are investing in funds, we insist on a right to call back the capital"
Time to market
Chris Cundy: If you have a mandate to invest in illiquids, it can take time to fully allocate your investment. What are the considerations around that?
Philip Howard: When working with outside managers, two things we ask ourselves are how quickly will they put the money to work, but also how sensibly will they put the money to work. I would much prefer a sensible manager who does not buy if the market is too high.
The other point is that you have to be very careful not to pay fees on committed money as opposed to actual money invested. But you have to balance that if the fund is just starting out and needs some money to keep going.
Gareth Quantrill: You're right, deployment windows for funds are very lengthy. I think one significant difference between a fund originating versus a bank is that banks seem less afraid of having a default. A lot of funds set a standard of almost a zero-default level, we've all seen the pipeline filter where firms typically only invest in 2-5% of all opportunities and I am not sure on an IRR [internal rate of return] basis whether that is necessarily the right approach.
Michael Leonard: And obviously the long ramp-up period exposes you to that market because you have committed the capital. When we are investing in funds, we insist on a right to call back the capital.
Philip Howard: Ultimately, you have to make a qualitative judgement about the sourcing ability of the fund, the deal flow they see and the intensity with which they look at opportunities.
Michael Leonard: Do others use a success fee? We feel it very much aligns managers' interests with ours.
Gareth Quantrill: Presumably you cannot do that in your real long-tail lending?
Michael Leonard: For some of the shorter-dated infrastructure, up to ten years, it will work. But post that, it becomes more difficult.
Philip Howard: The performance fee element that we pay is entirely based on the return. We usually end up with a private equity-like structure. There is the fee on capital that is invested, and then there is a preferred return, and then there is a catch-up, and then ultimately if you achieve the desired return or more, you pay a percentage of that when you achieve it. You do not pay anything up front.
Gareth Quantrill: It is quite difficult to achieve full alignment, given the demand for illiquids generally. So yes we have seen it in a few circumstances but if we cannot get that, we do not want to pay away some of our arrangement fees up front, which incentivises them to lend. So there is some balance in there.
This is the second of a two-part discussion. Click here to read part 1.