Bond yields: to infinity and beyond?
We think the market is on a voyage to discover what all that stimulus means for growth and inflationary pressure. We think this story is best split into three ‘acts’. It’s a dynamic process, though, so we wouldn’t want to put too much emphasis on any fixed forecast.
Act one: The market narrows the confidence interval around the path for growth
We think we’re about two-thirds to three-quarters of the way through this first act. We still think that risks are skewed towards positive surprises in consumption and employment, and would point to the performance of the Antipodean economies as the best guide for what consumption and labour demand look like after re-opening in nations that have conducted enormous fiscal stimulus.
But markets won’t be able to pull forward rate hikes much further without being able to price in yet more inflationary pressure. And because data about growth dynamics become available well before data about inflation dynamics, it will be hard for markets to price inflation dynamics that are radically different from those experienced during the last cycle. Until data on inflation dynamics become available, we think this translates to 10-year US Treasury yields struggling to rise above 1.80%.
Act two: Stronger narratives develop about the inflationary pressure of higher growth
Before COVID-19, strong and well evidenced narratives had formed about the death of inflation. Trained economists are eager to re-assert the value of their models, but need data to do so. We’re focused on the data that will enable them to make or break narratives about inflation’s demise. In the US, we’re watching the inflation sub-indices very closely for signs that key relationships have shifted and, perhaps more importantly, we’re monitoring the inflation dynamics in the rest of the world. Remember that extraordinary stimulus is not a phenomenon seen in large parts of the world, but many prices are set in a global marketplace.
We don’t believe that anybody – ourselves included – has a perfect model for inflation dynamics, as evidenced by the consistent one-directional forecast errors.
But as data about the inflation dynamics of this cycle become available, we believe we’re very well prepared to make a judgement about what can be priced.
Act three: If we realise inflationary pressure, how high can rates go?
This is really interesting. None of us have seen money supply growth on the current scale, the only precedent being during World War Two. Half of the increase in broad money supply sits directly in household accounts: cash as a share of financial assets for non-financial corporates is at its highest levels since 1969. Markets are still in the process of figuring out what all the monetary plus fiscal stimulus means for growth, but we believe that a significant amount of this cash will be spent, boosting growth, corporate profitability and possibly inflation.
A world where inflation is high enough to constrain monetary policy is still a way off, and isn’t our base case yet. But if we get there (act two), central banks in developed markets might be surprised by how much they have to raise rates to reduce inflationary pressure.
So what are we doing?
We’re very aware of the negative skew of outcomes from interest rates that cannot fall very far but can rise a lot. We’re combatting this in two key ways:
1) By diversifying our duration exposure among interest rates that are high enough that they can fall a longer way, e.g. the long end in Australia, to which we have very recently started adding; and
2) By diversifying into duration proxies: things that we have reason to believe should be negatively correlated with risk assets (and preferably a history of doing so, too) but that should have a better risk/reward balance, in our view, than outright duration. Here we have been doing interesting things with curve and inflation exposure.