Thursday, August 15th, 2019

Rhyming or Repeating

Most investors know Mark Twain’s crack about history nor repeating itself: “But it sure does rhyme.” It’s a more poetic way of saying that “past returns are not necessarily indicative of future performance”. We should always be careful about extrapolating a fund manager’s performance and equally careful about interpreting our probability charts. We make this point because our main asset allocation models look very like they did in August 2018. This doesn’t mean that we will definitely repeat last year’s 20% +/- correction in global equities, but it does mean that we need to recognise the risk.

Investors should be sceptical of bullish views based on the idea that activity data from the real economy and corporate earnings are holding up well. But they should be just as cynical about forecasts which threaten deep recession as divine retribution for the sins of Brexit or linking China’s protectionist policies at home with its demand for international free trade. Last year’s pattern will probably not be repeated, but the principal reason is that governments, central banks and investors will probably respond differently than they did a year ago. If they repeat their behaviour: if the Fed pretends that it is still data dependent, if the US and China continue their stand-off and if investors think that everything can be solved with just a couple of rate cuts, then the resulting correction may well be similar to 2018.

One of the themes we often discuss with clients is the “connectivity of risk”: the idea that the more individual risks there are to equity returns, the more likely is that two will combine to create a “super-risk”. The journalistic phrase for this is a perfect storm. We think this is increasingly likely sometime in Q4. It may even happen in October, but it almost certainly won’t be just because of Brexit. The problem with super-risks is that they are just too complex to forecast with any certainty. This includes the response functions of central banks and governments. All we can do is to describe market behaviour as accurately as possible in real time, not be distracted by other noise, and go where risk is rising more slowly than elsewhere.

At the moment, our main asset allocation models – dollar, euro and sterling – all show a sharp reduction in the recommended allocation to equities since the end of June. All of them suggest that we could fall to the maximum underweight relative to benchmark within the next 6-8 weeks. This means that geographical diversification within equities is unlikely to offer much protection. On a sector basis, we see clear signs of a shift into defensives. The biggest gainer in recent weeks has been Healthcare; prior to that it was Utilities. These sort of portfolio shifts are good practice in themselves, but they won’t provide enough downside protection in a period of super-risks.

Our bottom line is as follows: we are almost exactly where we were last August, but this year most financial market participants will have a different response function. Some (e.g. the Fed) may be more supportive of equities than they were, but some may be less so (e.g. Chinese officials who determine the central rate for the renminbi). On balance, anything which is likely to be supportive of equities, will probably also be supportive of government bonds and on a shorter-time scale, with less downside risk.

Leave a Reply

Full Blog Archive:


Other Recent Posts