LDI Swap Spread Outlook

BMO's LDI team take a look at the many drivers of asset swap spreads and their potential range for 2019

Share
Subscribe to our Insights

Risk Disclaimer

The value of investments and the income derived from them can go down as well as up as a result of market or currency movements and investors may not get back the original amount invested.

Past performance should not be seen as an indication of future performance.

 
 

In recent months, volatility has returned from quite a long and hopefully relaxing holiday. Its impact on various asset classes has elicited both cheers and groans depending on one’s viewpoint and positioning. However, volatility can also be framed as opportunity. In this article asset swap spreads are discussed, reviewing their many drivers (often with nuanced impacts) and potential range for the upcoming year. Additionally, we explore LDI exposure to asset swaps and discuss unrewarded risk.

The chart below shows the progression of relative value between the December 2049 government bond vs both SONIA (Sterling Overnight Index Average) and LIBOR (London Interbank Offered Rate) swaps. In the past LIBOR asset swap spreads were the norm and a simple and liquid method of expressing sentiment, however as LIBOR is soon to be retired into the fallback, volumes have been increasing steadily in SONIA asset swaps. It should be noted that many banks are lagging this trend as they would hedge the component parts as a LIBOR asset swap and the SONIA-LIBOR basis which can result in a) differentiated mid pricing between banks and b) more volatility in the basis post large asset swap programs.

Risk Disclaimer

The value of investments and the income derived from them can go down as well as up as a result of market or currency movements and investors may not get back the original amount invested.

Past performance should not be seen as an indication of future performance.

Asset swap spreads from 1 January 2018 (by basis points)

Source: Barclays Live, as at 6 February 2019

Related capability

LDI

Use our handy glossary to look up any technical jargon you are unfamiliar with. 

From the chart it is possible to identify a long period of gilts trending to be more expensive vs swaps (more pronounced vs LIBOR), which is then broken up by the return of volatility in the fourth quarter. The trend was driven by two major factors; firstly, the demise of LIBOR prompted relentless switching out of LIBOR swaps and into bonds and SONIA swaps and secondly, attractive buy-out pricing permitted more pension funds to transition to buy-out ready portfolios. There is an element of feed-through from US swap spreads as well.

An interesting corollary of the buy-out activity is that when prepping for their transition, pension funds typically switch into government bonds – purely due to ease of transfer, as once the insurance provider receives them they sell them to switch into credit and swaps! Particularly busy periods of this type of cycle show gilts becoming more expensive and then sharply cheaper vs swaps – up until the end of the third quarter the cheapening was opposed by opportunistic buyers which muted the impact. However, whilst both these factors remained throughout the year, the resultant trend did not persist into the fourth quarter.

The cat among the pigeons was of course Brexit and the scrambling to understand and position for the constant flux of the end game. The large moves were exacerbated by a more technical factor – that of market participant positioning into supply events and the final quarter of the year. The 2071 ultra-long syndication was a case in point as demand fell far short of what was expected and risk appetite hence contracted as many were caught on the wrong side of significant outright moves. As a result, the market reacted more strongly to risk events and significant positions. It was also a case of politics and supply events not being easy bed partners – the 2049 auction in early December drove a c. 0.25% drop in outright rates and a 0.13% shift in asset swap spreads.

Buy-out Bonanza

Buy-outs and buy-ins had a boom year in 2018 with volumes estimated up 20% on 2017. This has its roots in the following:

  • Insurance entities price using swap and credit discounting. Hence the recent softening in long dated credit has resulted in more attractive pricing.  Additionally, as gilts have trended towards swaps in yield terms there is less gap risk between gilt based and solvency discounting.
  • Pension funding levels have improved due to positive asset (equity) performance and some recovery in rates.
  • Availability of credit assets.
  • Health of sponsors supportive of buyout activity.

Looking forward, we asked our counterparties what they viewed to be the potential highs and lows of the 30yr SONIA asset swap spread over 2019. The results are shown in the radar chart. The numbers 1-13 represent the responses of each bank, with the lines showing their low and high estimates, with the spread increasing further from the centre of the chart.

Predictions of the highs and lows of the 30yr asset swap spread over 2019

Source: BMO Global Asset Management, as at 31 December 2018

The outliers in these predictions are related to either overly optimistic or pessimistic Brexit scenarios. Looking at the full set of data, the range is predicted to be around 0.30%, or if those two extreme scenarios are excluded then it gives a muted average range of 0.23%. The primary risk factors for bonds to cheapen versus swaps is the (ever increasing) possibility of a disorderly Brexit, including but not limited to marked GDP falls, rising unemployment, inflationary currency and falls in property prices which have both a negative emotional and financial impact on the average UK homeowner. A potential general election leading to a Labour Government is seen as extremely negative for gilts as many question the fiscal rectitude of some of their stated policies. Consequently, a knock-on impact could be a revision of official borrowing requirements to meet a fiscal expansion. Other factors include reduced demand for gilts as swaps regain popularity, repo costs increasing or transfers out into DC schemes and falls in longevity assumptions.

On the other hand, a hard (yet not disorderly) Brexit could result in lower rates as the BoE push back their plans to normalise the base rate which would likely push gilts to more expensive levels versus swaps. The BoE may even feel moved to restart QE; certainly the expected reinvestment of maturing bonds and coupons is at extreme levels (c. £21bn) and due to be invested over the (currently scheduled) Brexit date which will put some pressure on asset swap spreads. 

Much of the buy-out activity is dependent on funding ratios and the appetite of corporates to pay the premium: if these take a turn for the worse (global slowdown) then those pension funds in phase one could switch into bonds yet not complete the transfer. With easy access to repo balance sheet demand for gilts will likely outstrip equivalent swaps particularly if there are lingering concerns over central clearing and the LIBOR fall-back. The impact of foreign investors could again tend towards purchases, particularly if currency rebalances and global tensions rise.

The volatility in asset swap spreads, the current ease of obtaining repo funding and some discomfort with swap-based hedging (LIBOR fall-back, central clearing etc) have prompted many pension funds to remove perceived basis risk and switch their LDI hedge into leveraged gilts to match their gilt-based liability discounting. One of the fundamental tenets of LDI hedging is to reduce funding ratio volatility and this accomplishes this admirably (with a slight question mark over the future of repo funding). If gilt-based hedging was put in place a few years ago then the trend in gilt richening has also provided elevated returns versus a swap-based hedge. However, if a pension fund’s ultimate goal is to head to some form of buy-out, what then? Given that insurance firms utilise swaps and credit in their solvency discounting calculations and value the buy-out premium based off a risk-free-rate, (often swaps) does this cause end-game basis-risk? From the highs and lows that our counterparties provided, the average predicted level for 30yr SONIA asset swap is 0.49% in 2019 (current level is 0.43%). This implies that the risk is skewed towards gilts cheapening versus swaps, potentially increasing buy-out premium vs liability valuations. This is unrewarded risk and would suggest that it could be preferable to use a swaps-based hedge – however without the benefit of a crystal ball it is impossible to predict the most attractive form of hedging prior to the fact. However, a way to approach this problem of which asset to use could be to hold a mix of both gilts and swaps, thus mitigating sharp drops in value such as that seen in gilts up until February 2016. As an illustration, in the three years up to February 2016 a typical nominal gilt-based hedge would have risen by 29% compared to 38% for an equivalent swap-based hedge. Conversely a nominal gilt-based hedge would have returned 9% more than a swap-based hedge in the three year period up to end 2018. These are material differences purely based on interchangeable asset types.

Given the issue of significant differences in returns between the two hedging assets and the unrewarded risk taken versus buy-out pricing, is it practical to use the volatility in the asset swap to obtain some reward? The answer is potentially – although of course there are many approaches which may suit differing schemes and risk appetites. In theory a roughly 0.10% parallel move in the asset swap could imply a turnover of 10% of the portfolio (gaining 10x interest rate risk traded). An example scheme with £50m of liabilities and a 20yr duration would hence be able to lock in £100k in extra value for each such move (0.20% on liabilities). A similar approach is possible for real liabilities which again have the potential to add extra value over the longer term compared to either a pure swaps- or gilts-based hedge.

In conclusion, a confluence of demand and supply factors muddled with politics mean that volatility is likely here to stay in asset swap positions with potentially large moves expected either way. The eventual outcome of Brexit will tilt the range towards the higher or lower end. For schemes whose end game is buy-out in one to five years, it is tricky to choose the appropriate asset selection for the optimal outcome. However, for those with sufficient risk appetite it is possible to be rewarded by strategic and systematic switching into the cheaper asset, with the ultimate potential to lock in gains beyond either a gilt- or swap-based hedge return.

News & Insights