Friday, November 12th, 2021

The Roundabout Accelerates

This note is inspired by one chart, which comes from our Global Equities vs the World report, part of the regular suite of reports we summarize in Synopsis. So, before you read much further, we recommend that you turn the page and have a look.

Each dot represents a country index. For the time being, the identity of the country is not important; what matters is the regression line, which is sloping downwards from left to right. Normally, it slopes upward from bottom left to top right at an angle which fluctuates between 20-50 degrees. The slope has only been downwards in 61 weeks out of the last 26 years – less than 5% of all observations. These are typically clustered together in groups of three or four and there have only been 24 of these episodes in the history of this model. The reason they matter is that they often indicate that global equity markets are the verge of a significant rotation, where countries move up and down the rankings much faster than usual. This rotation typically lasts about four weeks and typically starts within four weeks of the slope first turning negative, as it has done this week. So, exactly what are we plotting and why?

The chart plots each country’s future against its recent past. On the x-axis, we plot the change in each country’s probability score (PRATER) over the last four weeks, while the y-axis shows its lead indicator – our estimate of its likely future direction. PRATER stands for the Probability of Risk-Adjusted Total Excess Return, which is our estimate of the probability that each country is going beat the return of the world index on a risk-adjusted basis. It is also a measure of how much capital we are prepared to invest in each of them.

The lead indicator looks at a series of different sample periods. If the shorter-dated samples (13-20 weeks) have a higher PRATER score than the longer-dated samples (21-28 weeks), it means that we are likely to become more positive about the country in future. Most of the time, countries which been performing well in recent weeks also have a good (high) lead indicator. The chart is important because it suggests that this relationship has temporarily broken down. A majority of countries which have done well in recent weeks now have a low lead indicator, and a majority of those which have done badly now have a high one.

It is now time to explain why there is a circle in the middle of the chart. Countries which have a very high or very low PRATER tend not to have experienced a significant change in this score over the last four weeks. For example, it normally takes longer to go from a score of 90% to 100%, than it does to go from 80% to 90%, or from 40% down to 30%. To get close to an overall score of 100%, most of the individual sample periods (long and short) must also score close to 100%, so the lead indicator – the difference between these scores – will also be close to zero.

The list of countries in the middle currently includes leaders, such as India, Russia, Czechia and the US, and laggards, such as China, Chile and Korea. The slope of regression line is not significantly influenced by anything inside the circle. So, this chart does NOT suggest that the USA or India will suddenly start to falter, or that China or Chile will rally. It is really about the “wannabes” and the unfairly stigmatised on the outside of the circle.

Most of the time, most countries on the edge of the chart progress in a clockwise direction. Countries which underperformed develop an improving lead indicator, which then translates into a higher PRATER. As the PRATER goes above 50%, the lead indicator starts to decay and it eventually goes negative as the country gets to its peak PRATER (which is frequently nowhere near 100%). From then on, the deteriorating lead indicator leads to a lower PRATER, until the country ends up in the bottom-left quadrant and the whole cycle starts again. This is, of course, a highly stylised account and there are always zig-zags and retracements along the way, but it remains true that countries in the bottom-left quadrant move into the top-left far more often than they move into the bottom-right.

The fact that the regression line is inverted means that this rotation is likely to happen faster than usual and affect more countries. For any two weeks, we can calculate the average gross change in each country’s PRATER score and repeat this to create a time series covering the whole 26-year period. If we exclude the top and bottom five, because they are normally inside the circle, we get a measure of the typical weekly change in the score of countries in the middle of the table.

Looking at the 24 episodes where the regression line was inverted, we find that in 16 of them the rate of change jumped into the fourth quartile (i.e., above the median) within four weeks of the first inversion, and normally two or three. There were only three episodes where it dropped into the first quartile and five where it stayed close to the median. By its very nature, this is a small sample size, but a score of 16 vs 3 gives us good reason to believe that a significant rotation between underperformers and outperformers is imminent.

Our second chart identifies all the countries outside the circle and we would be particularly cautious about anything in the bottom-right quadrant: Germany, the Netherlands, Belgium, Sweden, Denmark, Ireland and Switzerland. By contrast, we would look for opportunities to invest in the top-left quadrant, including Hong Kong, South Africa, the UK, Malaysia and Thailand. We might even be prepared to look at Mexico.

And finally, we ought to mention the wild-cards, countries which are inside the circle, but just passing through and not ranked in the top or bottom five. The two most important are Japan and Australia. They are just as likely to be affected by the rotation as the countries outside the circle, but we have much less visibility about their direction of travel. If we had to nominate two datasets or events which might trigger this rotation, we would look at inflation data in Europe and the possibility of an easier policy mix in China, after the ferocious squeeze on the housing sector.

Leave a Reply

Full Blog Archive:

Other Recent Posts