< Back to Research

Brutal shift in top-down signals since last week

Raphael Fiorentino
8th June 2020 - 3 min read
“People who have information about an individual case rarely feel the need to know the statistics of the class to which the case belongs”
Daniel Kahneman

Butterwire users would have noticed how the application switched emphasis last week from “recession-resilience” to “recovery” scores. As a reminder, there are four major stock scores provided at the top of each stock snapshot.

image1

The first two reflect the stock’s alpha forecast, respectively in upcycle and downcycle conditions. These forecasts are derived from the stock’s relative fundamental (and volatility) characteristics (as described in this note). The third consists of a recession-resilience score which combines a stock’s fundamental (downcycle) score with its global macro score (i.e. sensitivity to the changes in market-implied global GDP growth), and is described in this note, alongside the fourth score called recovery which gets activated when adverse macro conditions look like bottoming out. Recovery scores are attributed only to the bottom 20% performers in each region (based on trailing 12 month returns).

image2

With market-derived global GDP growth expectations (iGDP) skyrocketing even faster that they’ve been plummeting over March and April, Butterwire’s “top-down” signals have been pointing to a textbook recovery trade since the start of last week across the most hit regions (in emerging markets, e.g. LatAm) and sectors (e.g. energy and materials).

Likewise, the “top-down” asset allocator has brought back equities and real estate into the mix, reflecting the de-risking associated with a switch from defensive to pro-cyclical assets (note that such de-risking points to a barbell approach, i.e. selective but aggressive new positions in stocks that are high controversy, high value, high trailing underperformance).

image3

Looking at the past week’s global equity returns, stocks with a recovery score have significantly outperformed the benchmark, with stocks deemed “potential shorts” distant seconds and Butterwire’s long candidates as underperformers.

image4

That long candidates underperform (slightly) when recovery stocks catch a bid is to be expected. Recovery scores exist to alert Butterwire users of (rare but typically substantial) opportunities to capture outsized returns from selective de-risking/“bottom-fishing”. In turn, it flags the (relative) threats posed by such rallies to fundamental factor returns (ex-value) and defensive positions since recovery stocks have, by design, unattractive overall fundamentals (with the exception of valuation metrics) that often get amplified by high controversy levels.

Recovery stocks: illustration of underlying fundamental profile by region (May-2020)

image5

Back to last week’s outperformance of recovery stocks, this has been both very significant (e.g. in EU, outperformance averaged 40% of the stocks’ underperformance in the previous 6 months) and indiscriminate (e.g. 92% of outperformers and similar levels across recovery score quartiles in EU).

EU stocks with a recovery score:
Last week’s relative returns (y-axis)
vs. Previous 6 months relative returns
Last week’s relative return
as a function of recovery score

image6

Even when considering sector-relative returns, recovery stocks have done significantly better than peers without recovery scores, with the insurance sector the only exception in Europe.

EU stocks with a recovery score: Last week’s industry relative returns image7

What makes this fledgling recovery period unlike any other, from that of early 2009 to that of early 2016, is the speed with which everything prior and during has been unfolding. Even typically lagging official indicators like US jobless claims have already turned. In 2009, we’d had 4 months to digest the Lehman bankruptcy and another two months to reflect on the subsequent recovery signal (that at the time we only computed once a month) and develop a corresponding (in our case bull commodity equities) investment thesis. In 2020, with a similarly dramatic downcycle all seemingly packed into two months, being able to analyse both the global macro nowcast and stock alpha forecasts in real-time has become a necessity, one that vindicates Butterwire’s approach to bring 21st Century research tooling to buy-side professionals in a way that delivers a time and cost-efficient boost to one’s analytical and decision-making capacity.