Q&A: Mark Horsfield on Brexit, risk and local-to-local lending
0In the run up to the EU referendum, we began an in-depth interview over the course of the week with Arlingclose’s director, Mark Horsfield, about the direction their investment advice has taken in recent years. Half way through the interview there was a dramatic turn of events when Brexit, and Arlingclose’s reaction to it, came sharply into focus…
Room151: Diversification away from unsecured bank deposits has been a cornerstone of your advice in recent years, is that advice translating into real positions in client portfolios?
Mark Horsfield: That is correct, since 2013 we have been highlighting the implications of legislative change and EU Directives now in force on the resolution arrangements for many banks and the impact on unsecured deposits held by local authorities. We believe that the changes this represents should be highlighted within treasury management strategies and approved by Councillors, as should appropriate steps to be taken to mitigate the fluid nature of risk and credit risk in this instance.
The steps we have recommended and worked with local authorities to implement involve increased diversification of counterparties (achieved by progressively lowering recommended counterparty limits within the Annual Investment Strategy) and the utilisation of non-bank counterparties such as investment grade corporate bonds as well as other local authorities. But we also recommend secured investments with banks that are exempt from the bail-in regime altogether and these include covered bonds and repo.
Our work on a tri-party repo solution has taken longer than initially expected due to changes in documentation required by the collateral agent we have been working with. These changes have been as a result of EU Directives and, as global documents, the onset of negative interest rates in other parts of the world outside the UK. But there is strong interest from our clients, bank counterparties and the collateral agent alike and we are very confident that all these pieces will fit together soon.
In the meantime, covered bonds provide a very good way of achieving exposure to secured bank investments and client portfolios now include £700m of the covered bonds specifically monitored by Arlingclose. Including corporate, government, local authority and property exposures, bail-in exempt investments now represent over half of the aggregate investment balances held by our local authority clients – and with investment balances remaining resilient, we expect this proportion to continue to rise in future.
R151: For those that don’t know, could you briefly explain what tri-party repo is and how it works?
MH: A tri-party repo involves, by definition, three parties: the lender (in this case local authority); the borrower (a bank) and, finally, a collateral agent.
The local authority lends money to the bank in much the same way as they would with a normal unsecured deposit but in this instance the bank provides a pre-agreed level of collateral or security. This collateral is managed by the collateral agent.
Sounds complicated? Well let’s say that the pre-agreed collateral is a basket of UK Gilts that match the principal sum lent/borrowed on a 100% basis. The local authority lends the bank, say £5m for 6 months, and the bank deposits £5m of Gilts with the collateral agent so that in the worst case scenario, the bank defaults, the local authority takes delivery of an equivalent amount of Gilts. Over the course of the six months the value of Gilts will rise and fall and the agent will adjust the value of the basket of gilts or ‘collateral basket’ to reflect those change in valuation. Some of the Gilts in the collateral basket may pay a coupon or interest payment which is due to the owner of the Gilts, i.e. the bank, and these coupons will be collected and repaid to it. So the collateral agent manages the basket on behalf of both parties.
In the unlikely event of default by the bank then the collateral agent will deliver the collateral to the lender as security. At the normal conclusion of the transaction the borrower will repay the lender and simultaneously the collateral will be returned to the borrower.
Repos are a large market and they can also operate on a bi-lateral basis as well where the lender manages the collateral basket and the day-to-day administration of the margin and coupon payments falling due. Whilst this meets the same objective – secured lending – we prefer the tri-party arrangement and the utilisation of an experienced collateral agent who is remunerated by the borrower. That cost (0.01% to 0.02%) is reflected in the rate offered by the borrower. Repo rates will also be influenced by the credit quality of the securities within the pre-agreed collateral basket, and any over-collateralisation.
With authorities forcing banks to secure longer duration lines of funding – as one of a range of measures to address the Northern Rock scenario of borrowing short and lending long – we view tri-party repo as one of a number of solutions to address investment strategy and a situation where local authorities lending to banks on an unsecured basis bears increased credit risk as a result of bail-in but where unsecured rates of return have continued to decline.
R151: How would the yield on a repo loan compare to an unsecured deposit with the same bank?
MH: It will depend upon the credit quality of the secured assets in the pre-agreed collateral basket and the extent of any over collateralisation. For example, a collateral basket containing corporate bonds will, other things being equal, yield more than a collateral basket consisting of Gilts.
For a collateral basket of Gilts reflecting 100% of the value of principal, the yields are similar to unsecured bank deposit rates out to tenors of up to 6 months. Beyond that there is a larger credit risk premium built into the unsecured deposit rate meaning that the yield will be modestly higher relative to the secured alternative via repo. We like the risk-v-reward dynamic that this presents for treasury management strategies with capital preservation as the primary objective and it also enables us to implement our strategic advice of lengthening investment durations on a secured, as opposed to unsecured, basis with banks. Translating this into practice means that we advise our clients to have different counterparty limits for unsecured and secured lending activity.
R151: So where do you stand on asset-backed or mortgage-backed securities as source of secured investment?
MH: ABS/MBS are a little more complicated than covered bonds. Your counterparty is a special purpose vehicle, not the bank, meaning there is a separate company to review. One of the things we like about covered bonds is that banks have relatively few secured creditors, but in a MBS company, most investors are secured. So it’s a bit like priority boarding at the airport – it’s not so useful if everyone else has it too.
The other feature of MBS is that they don’t have fixed repayment amounts; as an investor you receive your principal and interest as the underlying borrowers repay their mortgages – including lump sums for early redemption. You therefore suffer from re-investment risk.
So we believe that ABS/MBS can have a place in secure, diversified investment portfolio. But due to these complications a pooled fund or a segregated fund are probably the avenues to gain exposure to this asset class.
R151: One of the other areas where we’ve seen a lot of growth in recent years is local-to-local lending. There seems to be an assumption among investors that there’s zero counterparty risk. What’s your view on that?
MH: I would not advocate the existence of zero counterparty risk other than where a local authority lends money to itself through what has become known as internal borrowing which is a very effective component of local authority treasury management strategy in the low for longer interest rate environment that has existed since 2009 and will last for some considerable time to come.
We do not currently differentiate between any UK local authority in terms of their individual capacity to be considered an acceptable counterparty for short- and medium-term loans and this extends to PCCs and Northern Irish local authorities. The basis for this is the regulatory framework and the strong interlocking credit metrics that include access to government loans (currently from the PWLB) and first call on the revenues of the local authority in the event of default.
As we know there are a relatively small block of local authorities that have obtained their own individual credit rating, and whilst these are not uniform ratings, they do provide some external reference point in terms of credit risk assessment and issues. Having said that, we are aware that some authorities do not meet the credit requirements of some lending banks largely due to the extent of existing external debt commitments and this reflects their own assessment of the credit risks associated with local authorities.
As the result of the EU Referendum has recently shown, unexpected events can happen and quickly create a sequence of other issues. It is for this reason that, in the absence of any due diligence, we advise clients to limit the duration of fixed term investments with other local authorities to 4 years. We have selected 4 years as a prudent reflection of the term of a political cycle although, as we know, a week is a long time in politics.
We are comfortable with investments longer than that provided that the duration is supported by financial analysis for all parties but we would also suggest that inter-local authority borrowing and lending in excess of 4 years requires more robust documentation. We view this as a sensible safeguard against the risk of material, unexpected change in circumstances over the medium and longer term horizon and if it does then there is suitable documentation in place that includes flexibility and transparency for both parties which is signed-off by the Chief Financial Officers of the authorities involved.
I would make a final point about the recent referendum result. In 2014 ahead of the Scottish Independence Referendum, Arlingclose advised its clients to limit the duration of their investment and borrowing activities where the parties were on different sides of the border. That limit was set with an April 2016 date in mind because had Scotland voted for independence, the expected start date of a fully independent and operational Scottish parliament was May 2016. What currency a fully independent Scotland would be dealing in was subject to a lot of debate and introduced currency risk into the equation.
For example, a Scottish authority could have invested with an English authority in August 2014 for 2 years in a transaction in sterling only to have received the proceeds back in sterling that it then needed to convert back into Euros or other base currency upon its maturity in August 2016. We thought it sensible to highlight this potential risk and with the uncertainty that has been generated from the recent EU Referendum result similar advice may be forthcoming in the future but for now we are very comfortable with UK local authorities as investment counterparties and they remain a key part of our strategic advice to address the bail-in risk associated with unsecured bank lending.
R151: So we’ve been doing this interview over the last week or so. When we started it we all expected a Remain vote and now we are where we are. What next for local authority treasurers?
MH: The outcome was quite a surprise. It was not surprising that it was close, there was inevitability about that, but it almost appeared that few in the City or even in the Leave camp really believed that a majority vote for Brexit would materialise.
The subsequent reaction in the financial markets has been predictably volatile and this has been exacerbated by the political implications and the delayed triggering of Article 50 that formally sets the process and timetable of the UK’s exit from the EU rolling. The uncertainty has not been helped by the entirely predictable point-scoring emanating from some EU officials regarding a desire for near immediate and tough negotiations to commence without any delay whilst others and, most notably, the German Chancellor Angela Merkel have adopted a more realistic yet understandably firm tone. Whilst much attention has fallen on UK financial markets and the value of sterling it has not gone without notice that European domestic stock markets and the value of the Euro have also been under similar pressures. This suggests to us that it will be more likely than not that there will be compromise towards an acceptable deal because bigger forces are at play and at the forefront is something called globalisation that goes way beyond the EU or UK.
But what do we specifically think are the treasury issues from where we now find ourselves?
Firstly, interest rates. We will be reviewing our interest rate forecast in the coming days as we always do during the first week of each month. The economists and research that is considered by our Interest Rate Group is now including an imminent 25bp cut in official UK interest rates at the MPC’s July meeting on 14th July. We are minded that the June minutes of the MPC were non-committal on the direction of monetary policy reaction to Brexit and that the inflationary impact of a fall in sterling would be temporary. Over the medium term, downward pressure on inflation from weaker activity is likely to support an easing in monetary policy and given the level of official interest rates it is also likely to be accompanied by a further round of asset purchases or QE in the Gilt market. The combination of these forces has the potential to drive Gilt yields lower than the record low levels seen in the last few days and sustain them for some time into the future.
In our view, this continues to vindicate a strategy of deferring longer term fixed rate borrowing, the utilisation of internal resources and the plentiful supply of variable rate/short-term funding available in the money markets.
Secondly, credit risk. The credit rating agencies (CRAs) have delivered on expectations and reduced or given negative outlooks on a variety of their credit ratings for the UK. Standard & Poor’s (S&P) and Fitch have both reduced the UK’s long-term to AA (from AAA and AA+ respectively). All three CRAs have the UK on a negative credit outlook. Moody’s reduced its long-term rating for the UK from Aaa to Aa1 in February 2013 in response to weakening GDP growth.
Whilst this would typically result in higher government borrowing costs, we are not in anything like typical times. The decision by Moody’s in 2013 led to an immediate increase in Gilt yields of 4bps (0.04%). The announcement by the CRAs this week has not stopped the decline in yields because greater market forces are in play and the agencies had already announced the likelihood of a ratings downgrade in the event of a Leave outcome. To all intents and purposes, they were catching up with events.
The downgrade in the UK rating is likely to feed through into other institutions and on bonds with an explicit Government guarantee, such as those issued by Network Rail Infrastructure plc and LCR Finance plc, this has already been confirmed by S&P. We have reflected these changes in our recommended maximum duration on the bonds issued by these institutions and those issued by Transport for London.
It is possible that the CRAs will extend the potential for downgrades into other institutions, including banks, since the action is largely driven by an expected deterioration in the UK’s economic growth outlook. Credit ratings are one of 8 credit indicators that we monitor and our Credit Group is currently reviewing all of those indicators and that investment advice. It appears likely and appropriate to shorten some of our recommended maximum investment durations on unsecured investments for some bank counterparties whilst noting that those durations have been progressively and deliberately shortened from the maximum levels recommended in our treasury strategy template anyway.
Reducing durations will mean that interest rate earnings from unsecured bank deposits are likely to fall, particularly if interest rates also decline, which is why we continue to recommend secured investments such as covered bonds for duration plays.
Thirdly, public finances. The current Chancellor of the Exchequer, in his first interview since the EU Referendum result, stated that whilst there would be no ‘emergency budget’ as he had suggested would be the case in the run up to last Thursday’s vote did provide a stark warning for the future in terms of there “absolutely” need to be [further] tax rises and spending cuts in order to deal with the economic consequences of Brexit whilst respecting the verdict of the British people.
I am not sure much more needs to be added here other than to suggest that the financial climate for local authority treasurers is going to remain very challenging. I have a hunch that this is not really news, treasurers know it already and it is more of the same. But the issue is whether it de-rails any government investment plans achieved through its devolution plans, LEPs and City Deals is another matter and suggests flexibility and affordability are likely to remain key issues alongside more traditional components of treasury management strategy in this period of uncertainty.
