How to find the perfect pair

‘Pair trades’ work by going long one bond and short another. In our absolute return bond funds we use pair trades extensively in rates markets to add performance. There’s plenty of fish in the bond market sea; the breadth and depth of global rates markets allows us to find the perfect match.

The beauty of pair trades is that we can make money without taking exposure to the direction of interest rates – a source of performance that is particularly desirable as yields move higher. We do this by sizing the long and short side of each pair in equal duration, as well as finding well-matched assets.

For example, last month we paired together Australia and the US. Australian bonds in our view offered relative value versus the US as the Australian market had prematurely priced in a rate hike in late 2018. We went ‘long’ Australia versus a ‘short’ in the US so that we were exposed to the outperformance of Australian bonds, without being exposed to a generic move higher in yields.

But not all pairs would work out like this, so we have to manage our matchmaking carefully. If we had positioned long Italy versus the US for example (with Italy a ‘riskier’ market than ‘safe-haven’ US Treasuries), the pair would be likely to perform badly in a generic ‘risk-off’ environment and perform well in a ‘risk-on’ market – not a risk-neutral position, or a well-matched pair!

So we would caution, on Valentine’s Day, that blind dates may not be the path to domestic bliss! We would never walk blindly into a pair-trade relationship, rather make sure to understand the correlation between markets and how they change over time. We would also argue that there’s no harm in a bit of caution (expressed as our strict review discipline): always have an exit strategy in mind and remember, not everyone mates for life.

Sign up to receive our weekly BondTalk email

Low unemployment… and low gilt yields

The most recent unemployment rate in the UK came in at 4.6% – along with the highest ever employment rate achieved, at 74.6%.

Economic theory – namely the Phillips curve model – tells us that as the level of unemployment falls, the economy can expect a corresponding increase in the rate of inflation. But recent inflation increases in the UK have been the result of commodity price increases and a weaker sterling – not inflation generated by a booming labour market. Average earnings and productivity data have been relatively static; this has been attributed to the dampening effect of the gig economy.

‘Gig economy’ is a term used to describe a labour market characterised by flexible, less permanent jobs with more short-term contracts and freelance work. The chart below shows the extent to which the UK has transitioned to this, with a significant increase in self-employed workers and zero-hours contracts as a percentage of the total employed.

Source: ONS as at 31 January 2017

This affects average earnings. Recent data from Barclay’s researchers indicated that self-employed workers earn 20% less than full-time employed workers, while those on zero-hour contracts have little bargaining power and struggle to force wages higher. On top of this, the current tax regime in the UK favours two lower wage earners in one household over one higher wage earner. The same net income can be achieved by two people doing less or doing lower paid work and leaving spare time to work in the cash “black” economy too.

These dynamics directly influence central bank decision-making and so are important to consider when forming a long-term view on rates markets. The structure of employment has changed; this is no rigid 1950’s society and as such we expect the Bank of England to demand some hard data before taking action this time round. Investors calling for a rate hike still have a long wait.

Brexit is a clear risk to any firm view on the outlook for the UK despite the reasonably benign outcome since June last year. With immigration still a key political issue, labour coming from Europe to work in the UK remains a wild card for employment numbers as Brexit negotiations kick off. Nonetheless, Brexit implications are far more likely to see a central bank willing to support the economy, rather than tighten policy.

A50 Trigger – should we expect some Brexit buyers’ remorse?

Theresa May has confirmed that Article 50 will be triggered on the 29th March – will this prove to be a catalyst for showing some UK ‘buyers’ remorse’ following last year’s vote to leave the EU?

In the short term we’ve seen little evidence of such sentiment – business investment has held up surprisingly well and consumer spending has been resilient despite the pinch of imported inflation. The UK has also (so far) avoided signs of institutional instability – there has been no evidence of an overseas buyers’ strike for Gilts.

On a longer-term view we’re still assuming the mood music that we ‘ain’t seen nothing yet’. If this is the case, Bank of England support will become relevant again. Right now the hurdle to more QE is lower in the UK than in anywhere else – although next time we expect a more reactive stance, so we must see some poor economic data first.

With this backdrop we will at some point want to be long UK gilts versus other core markets like US Treasuries. But not yet. We still expect that gilts could underperform in the short term as the Bank of England steps away from the market. To make money we will implement active, tactical trades – to exploit the market’s under or overreaction in the next nine days, and beyond.

Overreaction, or complacency?

Italian government bonds are peripheral European investments and trade as such – they have embedded risk and are correlated to credit risk over the long term.

But this relationship has broken down over the last few months. The following graph tracks the Italian spread to German bunds against investment grade credit (as measured by iTraxx Main) over the last 12 months.









Italian spreads have widened significantly, reflecting a much weaker performance from Italy than corporate bonds. Italy has been hampered by the rise in political uncertainty across Europe, yet credit risk assets have carried on unburdened.

So is Italy overreacting, or is credit complacent?

We think the latter.

Political risks in Europe are not currently priced in to credit markets, across investment grade and high yield. While we have a relatively sanguine view on the most likely outcomes for Europe, the Brexit referendum vote has put previously inconceivable ideas into the future political landscape – and complacency is not recommended.

Certainly a cheapening of Italian bonds is a welcome move – while the country’s December referendum caused little stir, investors are realising that elections outside its borders in 2017 could intensify its existing economic problems.

We expect this to translate into a credit market wobble at some point this year as uncertainty about the strength and implications of rising populism hurts sentiment. In our view this will be a buying opportunity – overall we expect credit to perform relatively well in 2017, thanks to an improving macroeconomic backdrop and continued policy support.