by Olivier d'Assier
Axioma’s new fixed income multi-factor risk model includes an exciting innovation: high quality issuer and sector spread curves for debt from both sovereign issuers (developed and emerging markets) and corporate issuers covering a range of currencies. Issuer curves and sector curves for thinly traded or illiquid issuers/sectors are notoriously difficult to estimate, particularly as a consistent time series, due to stale or unreliable bond prices. The new curves, based on robust cross-sectional and time-series outlier detection, peer analysis and missing data estimation methods, provide significant improvements in both quality and coverage.
In this post we examine how the richness of this data enables us to visualize investors’ changing risk aversion to various risk events. We start with a company-specific risk event: the Volkswagen emission scandal. On the 18th of September 2015, VW recalled 482,000 cars in the US after news of the scandal broke. On the 2nd of December 2015, VW took out a €20bn loan to help it survive the diesel emissions scandal, while vowing to protect jobs at the company.
Below is a chart of the issuer-specific spread for a Volkswagen EUR bond and the corresponding sector spread. Investors started demanding a higher risk premium for Volkswagen over the relevant sector/rating spread as soon as the news broke, and it wasn’t until early December that they started to believe the company would survive this scandal[1].
Another company specific event that received considerable media attention was the crash of a Tesla Model X car on March 23, 2018: Walter Huang, a 38-year-old Apple, Inc. engineer, died after his Model X crashed into a highway barrier in Mountain View, California. On March 27th, Moody’s lowered its corporate family rating on Tesla to B3 from B2 and changed the outlook to negative from stable. On March 30th, Tesla posted a blog late on Friday night acknowledging its driver-assistance software, Autopilot, had been engaged at the time of the crash.
Below is a chart showing Tesla’s issuer-specific spread, as well as the sector/rating spread for the Auto sector B1 and B3 of equivalent tenure. The jump in issuer-specific spread on the date of the accident is clear. The company’s spread kept rising as Tesla’s woes were compounded by Model 3 production delays and other issues.
Next, we take a look at the Japan Tsunami disaster of 2011. On Friday, March 11, 2011, an undersea earthquake of magnitude 8.9, one of the most powerful ever recorded, struck off Japan’s northeastern coast just before 3:00 pm local time. The authorities issued a tsunami warning. On Saturday, March 12, the Japanese government ordered the evacuation of residents living close to the Fukushima No. 1 nuclear power plant (owned by Tokyo Electric Power), where the quake caused cooling systems to fail and raised fears of a meltdown. An explosion subsequently occurred in a building housing one of the plant’s reactors, followed by a second explosion at the Fukushima nuclear power station on Monday, March 14.
The chart above shows the Tokyo Electric Power (TEPCO) issuer-specific spread, as well as the spreads for the Utilities Sector investment grade (A1) and non-investment grade (B1) JPY bonds. Tepco’s issuer specific spread was negative before the tsunami, meaning investors viewed the company as ‘safer’ than the average A1-rated Utilities bond. That all changed after the tsunami. In fact, investors begun to demand a much higher risk premium than even the average JPY non-investment grade (B1) Utilities bond.
Finally, we wanted to see if Axioma’s new curves could help us visualize how investors feel about the US-China trade tensions. According to President Trump, China has a lot more to lose than the US in a trade war. One of the ways we can track how investors feel about this assumption’s validity is by looking at the spreads for investment grade (IG) and non-investment grade (SUB-IG) bonds. If the gap between the two grades starts to widen, then we can infer that investors are beginning to worry that trade tensions are putting some pressure on the economy and that the more vulnerable segment (SUB-IG) might come under pressure.
Below are the corresponding charts for the CNY (China domestic bonds) and USD All Sector rating IG and SUB-IG spreads. During 2016, as successive economic indicators confirmed a stronger US (and global) economy, the gap between IG and SUB-IG spreads narrowed in both countries. During the final stages of the US presidential elections, however, and following the surprise win by Donald Trump, the anti-China rhetoric began to weigh on investor sentiment in China and the gap between IG and Sub-IG bonds widened. Following consecutive tit-for-tat tariff announcements in 2018, the gap has continued to widen in China. The US chart, however, shows a continued convergence of IG and SUB-IG spreads. Investor’s risk aversion seems to confirm President Trump’s assertion that China is more vulnerable than the US to a full-blown trade war (if it comes to that).
The impact does not seem to be limited to China. The chart below shows the USD SUB-IG spreads for APAC and North American issuers for Q1 & Q2 2018. Here, too, we see a widening gap between Asian and North American issuers when it comes to the risk-premium investors are demanding for the former.
In conclusion, although I am just starting to explore the richness of Axioma’s curves offering, I find them to be a great canvas on which to paint a clear picture of the type of risk investors are worried about and how they react to certain events. The granularity that they offer is unmatched and I am looking forward to incorporating them in my market analyses.
[1] The holding company is called Porsche SE and does hold the majority of shares in Volkswagen Group according to their website.
