Stress testing COVID-19: Macro explains market moves SPONSORED POST

COVID-19 has been a major macroeconomic shock for the world. As a result, we have seen very large moves in inflation expectations, real GDP growth, credit spreads and energy prices across regions. These shifts in “macro factors” have deeply impacted all asset classes.

In late January/early February, as it became clear that the COVID-19 situation was becoming more serious, global investors started to think about how to effectively manage their exposures, initially to stress in China and then to a broader global deflation shock. What they needed was a way of understanding which stocks, regions, sectors, indices, FX pairs, rates and commodities were most vulnerable to a global growth/inflation slowdown so that they could adjust their portfolios and risk profiles accordingly.

What’s required is a comprehensive, robust, quantitative and multi-asset macro factor investing framework. Ideally one that leverages cloud computing, enormous data availability, a rigorous scientific approach and technology that provides effortless delivery of this key solution for global investors.

Fortunately such an approach exists and its model answers in early February 2020 played out with an accuracy of 65-95% over the next six weeks. Assets moved broadly in line with model predictions, given the moves in key macro variables. This provided proof of effectiveness as an objective macro playbook.

Going forward, the key issue for many is how to position for the global monetary and fiscal policy response, and this is also where a macro factor framework can be so useful.

In this exercise, we focus on 10 benchmark assets covering major equity, interest rate, currency and commodity markets across the globe. Given their factor sensitivities on February 17th, the last week of complacency before the virus hit Western economies, and the subsequent shifts in those factors, how accurate was Quant Insight in modelling subsequent price action?

For example, the S&P500 was 3,380 on February 17th. Given its macro factor sensitivities on that day, and the factor shifts between then and March 16th, Qi's implied model value was 2,623. In fact SPX fell an additional 237 point to 2,386. That implies a 76% accuracy rate from Qi in modelling forward price action in US equities over one of the most volatile months ever endured.

The Stoxx 600 was 432 on February 17th. By Monday of last week, the Qi model price, given factor sensitivities and how those factors moved, was 293. At that point spot, SXXP had fallen to 285. A 95% accuracy rate in modelling European equities.

The results are equally impressive for Asian securities. From 6.987 on Feb 17th, spot USDCNH rallied to 7.159 on Mar 26th. Qi factor sensitivity put model value at 7.166 a 96% accuracy score!

JPXN, the iShares Nikkei 400 ETF, recorded a 91% accuracy rate between mid-February and late March. Irrespective of asset class, macro is critically important and Qi captures macro in an empirical, easy-to-use way.

Moreover, Qi can decompose the move in asset prices; attributing price changes to key macro factors. In the chart below we show the dominant drivers over this period in six of our global benchmarks. The COVID-19 narrative a deflationary shock and a shortage of USD liquidity that prompted a spike in risk aversion and fears of a credit crunch is neatly caught and broken down asset by asset.

The % values show the accuracy of Qi’s predictive models.

An experienced macro investor, Mahmood started researching quantitative methods in 2011. He spent several years as a macro PM at BlueCrest, Millennium and CCA. He started his career at Morgan Stanley as a derivatives structurer. Mahmood also worked in his first technology start up from 1998 to 2000, and holds a B.Sc & M.Sc from the LSE.

Quant Insight provides quantitative macro data across multiple asset classes to a wide array of different investors. Whether discretionary or systematic, whether equity long/short or absolute return, Qi brings a single, comprehensive and robust solution to their clients. In an increasingly complex world, Qi’s curated macro factor data set brings signal not more noise. Our core belief is that an empirical approach to analysing financial markets can be combined with human judgement to enhance performance. For more information, please visit