China has been lighting up the news outlets lately with news of crypto crack downs and general regulatory tightening. However, another story has been brewing around China Evergrande and the potential ripple effects on the global economy from Evergrande’s impending default on portions of its $300+ billion in liabilities.
As CNN notes in a primer on Evergrande, the real estate conglomerate is not only part of one of China’s largest industries, but it’s also one of the largest companies globally with business arms that reach into a wide variety of pockets throughout the world’s economy. Evergrande has recently made investors clearly aware of the drying up of its cash flow well and amid missed debt payments over the past couple weeks, investors are fearing the worst. Given Evergrande’s influence on the global economy, could its looming default be a catalyst for a repeat of the Lehman default and subsequent financial crisis?
Over the next two weeks, we’ll look to the Axioma Worldwide 4 Linked (“Axioma WW Linked”) risk model to help answer this question.
Using a Linked Model to Measure Potential Contagion
While global risk models typically offer breadth in coverage, they are lacking in granularity. In the traditional global model construct, securities across local markets are compared to one another within the estimation universe. This can, at times, have the effect of “washing out” the differences in style and sector behavior across countries and regions. In contrast, the Axioma WW Linked model is a global risk model that estimates style and sector factors within region (e.g. Emerging Markets, Developed Markets, and the United States) in order to provide regional granularity while maintaining global coverage.
Given that Evergrande is a real estate developer, we’ll use the Real Estate Management & Development (“REMgmt”) factors to illustrate the benefit of viewing global risk through the Axioma WW Linked model lens. Below, we’ve charted the EM, DM, and US regional returns of the REMgmt factors from the linked model compared to the global return of the REMgmt factor from the Axioma Worldwide model.
In this case, using the global REMgmt factor as a proxy for the industry performance in Emerging Markets would have understated the region’s drawdown. Similarly, using the global factor to proxy the US REMgmt performance would have completely misstated the trend of this local industry over time.
Another view using the implied Axioma WW Linked model correlations between the EM, DM, and US REMgmt factors further solidifies the importance of isolating the regional differences in these factors’ behavior.
Unsurprisingly, we see that the REMgmt factor correlations across regions all have varying levels and trends. This highlights some of the obscurity caused when blending all of these regions together into a single factor, as we can see from the returns and correlations that these factors move with distinct behavior.
As demonstrated above, the model can capture the correlations and possible second-order effects from market movements in one region across the others, making the Axioma WW Linked model suitable to assess the potential spread across global markets from a debt crisis in the Chinese real estate sector.
Evergrande’s Factor Profile
Now that we’ve shown how the linked model framework can help us measure Evergrande’s impact on the global markets, let’s assess Evergrande’s factor DNA.
In the case of the linked model, Evergrande picks up on the EM style and sector factors because it’s local market, China, is in the Emerging Markets. This is significant because China makes up about 30% of exposure within the MSCI Emerging Markets index and real estate (including related industries) makes up about 30% of Chinese GDP. This means that the EM REMgmt factor is likely heavily influenced by Chinese Real Estate, and likely Evergrande, specifically. For purposes of our analysis, we can refer to the EM REMgmt factor as the “Evergrande factor”.
However, Evergrande also has exposure to style factors within the Emerging Markets.
In this case, the risk model confirms much of what we know to be true about Evergrande, especially given the company’s current situation. It’s a highly levered value stock with an over-exposure to residual volatility and beta.
With Evergrande’s factor profile in mind, we can turn to the correlations among the regional risk model factors to better understand how a collapse to a major Chinese market segment can propogate to the rest of the global markets.
In next week’s Factor Spotlight, we’ll continue our analysis into the “Evergrande factor” by dissecting the correlations of EM real estate to the US style and sector factors and creating a framework to estimate the impact to regional markets from a shock to the EM real estate factor.
In case you can’t stand the suspense of waiting until next week to see the results of our stress tests on Evergrande, please feel free to reach out to us directly to discuss applying an Evergrande stress test to your portfolio.
US Market: 09/27/21 - 10/01/21
- The broad US market factor return ended the week negative, with the S&P 500, Nasdaq, Russell 2000, and Dow Jones all ending in the same negative territory. The Nasdaq faced the most pain with a loss over 3%, with the Russell 2000 experiencing the most volatility all week with a strong Friday to recover some of the week’s losses.
- The market faces a flurry of challenges to overcome: inflationary pressures from rising energy costs and disrupted supply chains, political uncertainty around the US Debt limit and Infrastructure, and a rise in the 5 and 10 year yields haven’t provided much confidence to the markets. It remains a question if this market weakness will persist into the first full week of October, or if the dip gets bought and the market continues its churn upwards.
- Eyes are on the ADP employment numbers coming out this week to see if the labor market is showing signs of recovery or further signs of slack, as labor is clearly being watched by the Fed. Will the numbers be strong enough to give the Fed the confidence to proceed with their tapering plan, and how will the market interpret it all?
Normalized Factor Returns: Axioma US Equity Risk Model (AXUS4-MH)
Methodology for normalized factor returns
- Earnings Yield continues to be oversold and generally overall struggling for positive returns
- Profitability continues to be oversold
- Growth continues to be oversold, and is currently most at threat to cross the ‘Extremely Oversold’ threshold of 2.0
- Size started the week neutral, but by week’s end was oversold with one of the biggest drops in standardized returns
Despite the week’s overall sell-off, no single Style factor left it’s fingerprint in the returns. We also see a trivial 17 bp increase in total US market risk.
Normalized Factor Returns: Axioma Worldwide Equity Risk Model (AXWW4-MH)
Methodology for normalized factor returns
- Market Sensitivity continues to be overbought, with one of the biggest positive changes across the board.
- Value crossed the threshold into the overbought realm.
- Volatility threatens to rise into overbought space while Profitability and Growth push to do the same on the oversold side.
- Earnings Yield continues it’s march into oversold territory - more so globally than in the US.
Similar to the US markets, the global markets didn’t see any dramatic changes in Style factor performance. The predicted risk of global markets was even more muted than in the US: predicted global market risk only increased 11 bps.