Open-ended funds investing in less liquid assets

Speaker: Edwin Schooling Latter, Director of Markets and Wholesale Policy
Location: Investment Association 
Delivered on: 19 March 2020

Highlights:

• The FCA and Bank of England have been working together on the risks arising when there is a liquidity mismatch between the assets and liabilities of an open-ended fund.
• Swing pricing and notice periods can help protect fund investors and reduce risks to the wider financial system, when there is pressure to sell fund assets to meet redemption requests.
• The FCA invites industry and stakeholder comment on the best mix of tools to ensure open-ended funds appropriately manage liquidity risks while enabling investments that can benefit fund holders and the wider economy.

Note: this is the speech as drafted and may differ from delivered version.

---

Between 2016 and end-2019 we had a number of UK examples – including various commercial real estate funds, and the LF Woodford Equity Income Fund – of open-ended funds needing to suspend dealing in response to the volume of redemption requests.

These suspension decisions served to protect remaining investors. Had these funds not suspended, fund assets would have had to be sold at potentially severely reduced prices to raise cash quickly. That would almost certainly have been a worse outcome for investors than suspension. An important principle in authorised funds – which are mutual investment vehicles, not individual deposit accounts – is that all investors, not just redeeming investors, are treated equitably.

We are now in the midst of an unprecedentedly difficult time for funds, as for others. The suspensions of property funds announced this week are different in root cause from some of the others I referred to – driven by valuation uncertainty rather than redemptions. What is common, however, is the fact that suspension can serve investors’ best interests in difficult market conditions. 

Nevertheless, these examples have raised legitimate questions about the wisdom and appropriateness of promising daily liquidity to investors when funds invest in illiquid or less liquid assets.

The issue of liquidity mismatch is not confined to the UK. In its 2019 ‘Stress simulation for investment funds’[2] report, European Securities and Markets Authority (ESMA) found that, based on calculations set out in its report, up to 40% of high-yield bond funds within the EU could experience a liquidity shortfall in a severe redemption shock scenario.

As you know, the FCA and Bank of England have been working on a joint review of these issues. The progress of this work has been detailed in the Bank of England’s Financial Stability Report[3]. The FCA fully supports that joint work.

Today, I would like to invite your thoughts on what the optimal policy response should be.

The risks

Where an open-ended fund offers daily redemptions, but a significant proportion of the assets in which the fund invests cannot be liquidated within a day without material loss of value, there is an asset-liability mismatch.

That mismatch gives rise to several risks.

First it is a risk to the fund’s own investors.

Investors who take the daily redemption promise at face value, and redeem without understanding that this could force a sale of fund assets at a lower price than could be obtained through a slower sale, will achieve lower returns (or losses) compared with what could have been achieved.

Asset-liability mismatch in the fund’s structure, and the first mover incentive, introduces a source of instability in the fund.

Typically, a fund will have a mix of assets. Diversification is often a main purpose of investing through fund structures rather than in individual assets. Some fund assets will be more liquid than others. There might be a cash buffer. This might be held in part specifically for the purpose of meeting redemptions without having to sell less liquid assets at speed. A cash buffer is not, however, necessarily enough to solve the problem. Sometimes, the incentive for investors to redeem before the cash buffer is exhausted, could even exacerbate redemption pressure.

Let’s say that investors look at the current dealing price of the fund, and consider it higher than the price that could be achieved by selling a representative sample of the fund’s assets quickly – by ‘quickly’ I mean fast enough to meet others’ expected redemption requests as they fall due. There could be an incentive for investors to make redemption requests now, before others do. If the fund’s redemption arrangements do not counterbalance this ‘first mover advantage’, the lower returns or losses could instead fall on those who remain in the fund. This asset-liability mismatch in the fund’s structure, and the first mover incentive, introduces a source of instability in the fund.

Second, this structure could in extremis amplify systemic risk. If the redemptions, perhaps driven in part by investors pursuing this first mover advantage, force a 'fire sale' of the fund’s assets, this impacts the market price of these assets. That could trigger losses in other funds and on others’ balance sheets, in turn potentially catalysing other redemption requests or forced asset sales. And so on.

As the amount of assets under management in open-ended-funds increases, it becomes increasingly important to make sure that this potential systemic risk is appropriately mitigated.

How can these risks be addressed?

In principle, this mismatch can be addressed either through the asset side of the balance sheet, or through managing liabilities, i.e. redemption arrangements, in a way that better matches asset liquidity. We already have rules and guidance that speak to liquidity risks arising from mismatches, though recent experiences do raise the question of whether they need to be tighter.

The liquidity of a fund’s assets

Addressing the mismatch through the asset side, would mean restricting the assets in which daily dealing open-ended funds may invest to those that can be sold same day without material loss of value – for example highly liquid shares or the highest quality liquid bonds.

We do not want to introduce new restrictions on these investors’ access through authorised funds to investments that may offer a higher expected return.

Our regulatory framework for open-ended authorised funds offered to retail already contains some restrictions on the assets in which funds can invest. For example UCITS, which comprise the largest share of open-ended funds marketed to UK investors by value, may only invest 10% or less of their funds in unlisted securities.

But liquid assets generally offer lower expected returns than less liquid ones precisely because investors must pay a premium (i.e. accept a lower return) for the liquidity. Yet many investors in funds do not need immediate liquidity. Many are investing for the long term, hoping to maximise returns over that long term. We do not want, therefore, to introduce new restrictions on these investors’ access through authorised funds to investments that may offer a higher expected return.

Moreover, it is important for overall economic growth, employment and prosperity, that the financial system supports investment that may take time to deliver a return. New businesses, or infrastructure projects, that require up front capital to invest, may not deliver a return for some years. So, beyond the potential benefit to investors themselves, it is important to the economy as a whole that the financial system can support such investment.

There are ways of making such long-term investments which still enable investors to redeem investments without forcing the sale of the underlying assets. Shares in closed-ended investment companies – investment trusts – are an obvious example. These may offer a better choice for those end-investors wanting exposure to illiquid assets, such as property or infrastructure, while retaining a right to try and sell their shares 'on demand'. Those who want to exercise that right at times of market stress may, however, have to accept a steep discount to net asset value. In other words, this liquidity will still come at a cost. Moreover, buying or selling shares in a closed-ended vehicle each time that the investor wishes to increase or redeem their investment, may not suit all investors.

Overall, we think that imposing additional restrictions on investment in illiquid assets is unlikely to be the best solution for fund investors or the wider economy. That said, for fund structures which should fit a risk profile designed with retail markets in mind, such as UCITS, it is important that the rules are crafted in a way that achieves that risk objective. We will, for example, look for an opportunity to review the UCITS rule that creates a presumption – unless there is information to the contrary – that listing on an exchange means an asset is sufficiently liquid to meet short-notice redemption requests. Some listed assets are still illiquid.

Managing liabilities to match the liquidity of assets

The alternative to restrictions on the asset side is better to align redemption arrangements with the liquidity of assets. One tool which very directly addresses first mover advantage is swing pricing. Another obvious liquidity management tool is notice periods. These are the two main types of tool on which the FCA and Bank’s joint work has focused.

Swing pricing

Research[4] we published in May 2019 presented evidence that the use of swing or dual pricing adjustments leads to significantly reduced redemptions from bond funds during periods of market stress.

Swing pricing is already allowed in our rules. Many single-priced funds already use partial swing pricing. Some investors may consider it the best way to remove a first mover advantage without restricting redemption. They may place trust in the fund manager using swing pricing fairly, with the depositary overseeing this. Other investors may be uncomfortable about the variable nature of the swing. Some fund managers may be uncomfortable with the challenges of explaining a complex tool to less sophisticated investors, or the complexity of using it in practice, or concerned about how to manage investor perceptions of whether the swing price was calculated fairly.

Swing or dual pricing adjustments leads to significantly reduced redemptions from bond funds during periods of market stress

There are challenges, however, to applying such a mechanism for more illiquid asset portfolios. It requires an assessment of the potential sale price of underlying assets before a sale and redemption is made. The magnitude of swing pricing would vary with market conditions and overall redemption pressure, and so may be difficult to calibrate in practice. If swing pricing is used, it must be used fairly. It is intended to help avoid remaining investors bearing an unfair proportion of costs, risks, or loss of value, as the fund’s most liquid assets are sold to meet redemptions by exiting investors. But fairness also means not discounting redemption value so severely that it is unfair on those exiting the fund. It should not be used as an attempt to gate or suspend redemptions “by other means”.

Given that fair and appropriate use of swing pricing can benefit investors by preserving liquidity while removing first mover advantage, we think there is a place for this tool where funds hold less liquid assets. We doubt all will choose this option, and it will not be the best tool for some asset types. But we would welcome stakeholder input on whether stronger regulatory guidance around the fair and transparent use of swing pricing tools would be useful.

In theory at least a fund could choose to offer daily redemption at a fully “swung” price as an alternative alongside redemption at the end of a notice period – i.e. offering the investor a choice. We would be interested in whether any of you think that is a choice investors would welcome, or whether practical, administrative, complexity or other complications are too great.

Notice periods

The other intuitively obvious tool to address liquidity mismatch is the use of notice periods. Conceptually, they could offer several benefits. By notice period we refer to a gap (e.g. weeks or months) between the point at which a redeeming investor submits a redemption request and the subsequent pricing and execution of that transaction. We do not use the term to refer to redemption frequency (for example, only being able to redeem once a month).

First, they are arguably a simpler and therefore more effective way of promoting investor understanding of the risks associated with investing in less liquid assets through an open-ended structure than the most comprehensive descriptions of liquidity risk and risk management tools in a lengthy prospectus.

The other intuitively obvious tool to address liquidity mismatch is the use of notice periods.

Second, they may reduce the likelihood of redemption requests at times of market stress. As when redeeming daily-dealing funds, the actual sale value is only determined at the point when the transaction is priced and executed. But, with a notice period, the gap between the last published Net Asset Value (NAV) before receiving the redemption request and the NAV determined at execution of the transaction would be longer. Trying to sell at a time of stress would mean the investor accepting significantly greater uncertainty about the value they would achieve than if they gave notice in a period of more 'normal' market volatility. As with swing pricing, the choice faced by the investor looking to sell at a time of stress would be more like the one they would face if they held the fund’s underlying assets directly.

Third, the existence of the notice period reduces – but does not wholly eliminate – fire sale risks. If the fund could, for example, sell its relatively less liquid bonds over a 2-week or 1-month period, rather than having to sell on a single day when markets may be most stressed, this could both be better for investors and reduce amplification of price moves affecting the remainder of the financial system.

How long should a notice period be?

Determining the optimal length of a notice period requires careful judgment and balance. Setting the notice period long enough to allow all redemption requests to be honoured regardless of the level of redemptions, and regardless of market conditions, would mean a notice period equal to the time necessary to sell the fund’s least liquid asset in the most unfavourable foreseeable market conditions. Yet in some cases that could involve locking in investors for longer than is needed in almost all circumstances.

It is not clear it is in investors’ immediate best interests to deny them redemption when this can be achieved without loss of value either to them or to remaining investors, and without immediate systemic risk. You could advance an argument that this is a justification for allowing daily redemption in 'normal' times while withdrawing this right in the face of high volume redemptions, or extreme market stress. That is, in effect, suspension – ie more or less the model that we have now.

When extraordinary circumstances do occur suspensions may be the best remaining option both for the fund’s own investors, and for the wider financial system.

But if investors place trust in a promise of daily redemption that is not honoured, even if that happens only rarely – perhaps for idiosyncratic reasons or in particularly adverse market conditions – this can damage investor confidence in an asset class that includes thousands of other funds, over a much longer time period.

Fund structures need to be such that they work through the economic cycle.

Within that overarching constraint, there is still a balance to strike. Aligning notice periods to the fund’s least liquid asset in the most stressed market conditions may remove the risk that redemption pressures ever lead to suspension decisions, but may be more than is necessary. Fund design and choices must ensure that resorting to recovery and resolution type tools – arguably suspension is such a tool – is rare. But we should recognise that a regime where there are no suspensions may not be achievable, or, if it were, the costs and lost benefits might exceed the advantages. We would like an open debate with stakeholders about where the right balancing point lies.

We should remember that when extraordinary circumstances do occur – when it is not possible to sell assets to meet redemptions without major loss of value – suspensions may be the best remaining option both for the fund’s own investors, and for the wider financial system.

Measuring liquidity and transparency

If there were new regulatory requirements to have notice periods or swing pricing, these would be unnecessary for funds invested wholly in highly-liquid assets. The optimal choice of how much to rely on each tool could also look quite different for different types of fund – some will have assets for which calculating a fair swing price is feasible. For other assets this will be much more difficult.

This raises the question of the appropriate liquidity threshold, or thresholds, at which additional requirements would apply. That, in turn, takes us to the question of how to measure liquidity. In the United States, the Securities and Exchange Commission (SEC) has already implemented a liquidity measurement regime based on 'bucketing' assets into 4 categories. The SEC’s approach offers useful experience of the advantages and difficulties of asset liquidity measurement. The joint Bank of England – FCA review has also been considering this challenge, and has been working with firms on piloting some data gathering. We are planning to survey a larger set of firms at a suitable time when firms will be able to engage with such an exercise. We are looking for an approach that has synergies with the way good managers already identify and track liquidity risk, and therefore would minimise additional costs.

This leads to a further question about whether the liquidity profile should be disclosed. There are some arguments in principle for disclosure, but we also need to be mindful of downsides. For example, because measuring liquidity is subjective, there is potential for misleading comparisons. The US SEC decided against requiring public disclosure because of those downsides.

UK and overseas funds 

Around three quarters – 8,317 of 10,930 – of the Undertakings for the Collective Investment in Transferable Securities (UCITS) funds available for sale to UK investors are domiciled in EU jurisdictions rather than the UK. Any measures the FCA took to change the rules around pricing tools or notice periods for UK-authorised funds would not apply to non-UK funds sold here. That would decrease the effectiveness of those rules in addressing risks faced by UK investors, or systemic risks. Recognised non-UK funds with a less liquid asset mix could continue to offer a superficially tempting daily redemption promise to UK investors, even if UK domiciled funds could not. UK authorities will continue to work with our international counterparts to discuss the optimal approach to what are common risks. EU national competent authorities (NCAs) have been asking similar questions as to whether existing rules in respect of liquidity risk are sufficient. An international approach consistent enough to achieve common outcomes is likely to be the most effective one. But it will take time to realise.  

In the meantime, if UK rules on open-ended funds with illiquid assets are tightened, and those in other jurisdictions do not achieve the same outcomes, this will raise a question to consider when looking at additional conditions under which non-UK funds can continue to be marketed in the UK under future access arrangements. The Treasury on 11 March published a consultation paper about those arrangements – the overseas funds regime[5] (OFR). If we get our rules right, they should be to the advantage of fund managers and their investors. Under the Temporary Marketing Permissions Regime in force from the end of the EU withdrawal implementation period and until the OFR comes into place, there could, however, be a period in which rules applying to UK funds are tighter than those applying to others. Those offering investment in funds that would not meet UK requirements, for example through investment platforms, would need to think carefully about continuing to market or offer those funds to their customers.

Existing funds with illiquid assets and no notice period

In our view, the case for preventing the emergence of new open-ended funds with asset-liability mismatches and without the right tools to manage them is strong. But we need to be sensitive in addressing risks in existing funds. Their investors may not all welcome regulatory intervention to change the terms on which they can redeem their investments, even if these are intended to protect them. Such an intervention could even add to existing redemption pressure on those funds, exacerbating risks that intervention would be intended to reduce. This might argue for focusing first, for example, on limiting new retail investment in illiquid assets funds without the right tools, while avoiding abruptly changing existing investors’ choices.

Senior manager responsibility

Finally, I note that everyone on this call today will already be keenly aware that it’s the responsibility of relevant managers in authorised firms selling and operating funds to make sure they are designed and run so that the promises made to investors are met. Recent events have underlined that responsibility.

Compliance with regulatory requirements for authorised fund managers, such as managing liquidity and treating investors fairly, is part of obligations on relevant staff at firms operating funds to act with due skill, care and diligence. All firms should be clear where this responsibility lies.

From 9 December last year, more or less all FCA solo-authorised firms became subject to our Senior Managers & Certification Regime. That serves to reinforce the responsibility of senior management and boards at authorised fund managers, delegated portfolio managers, authorised depositaries, as well as at product distributors and platforms.

Conclusions

The FCA and Bank are continuing work to evaluate the costs and benefits of possible policy measures to achieve greater consistency of fund redemption terms with the liquidity of funds’ assets. We would welcome input into this work from investors and from industry – hence this meeting with you today.

We recognise that a daily redemption promise may be attractive for many fund investors. And for many funds investing overwhelmingly in highly liquid assets, this does not create a liquidity mismatch. We want open-ended funds to provide a structure through which investors can safely invest in less liquid assets which offer attractive expected returns and which supports investment that benefits the wider economy. Recent experiences may help rational and well-advised investors to recognise that daily redemption funds are unlikely, however, to be the safest or most profitable way of investing in less liquid assets. We think now is the right time to work with you to address this, and make progress on embedding the right structures for open-ended funds investing in such assets. We know the Investment Association is doing its own work on the Long-Term Asset Fund as part of that. We want to ensure the regulatory framework is right too. Getting this right should benefit investors, fund providers and the economy as a whole.