Stock market volatility and central bank tail risk are about to both increase, a more reserved Marko Kolanovic write in a September 7 research report. The highly watched JPMorgan quantitative derivatives analyst noted that over the summer stocks were exiting a “dead zone” and that leverage and resulting equity risk exposure is at high levels among risk parity, volatility targeting and even CTA strategies. This leverage could turn around and send the market lower with the slightest of nudging from negative stock market price trends. Two of the potential catalysts involve the central bank and the increasingly close US election.
Marko Kolanovic - With mechanical strategies fully leveraged, any sudden moves in stocks could be exacerbated by an unwinding
The stock market needs to move only 1% to 2% lower for volatility to dramatically increase to the downside, as highly leveraged strategies could engage in mechanical selling.
“Given the low levels of volatility, leverage in systematic strategies such as Volatility Targeting and Risk Parity is now near alltime highs,” Kolanovic wrote. “The same is true for CTA funds who run near-record levels of equity exposure.”
When markets are pushed to extremes, the snap-back to normalcy can be hard. Kolanovic notes that record leverage in these strategies could push the market lower and volatility higher. The market might not even need a catalyst to increase volatility, seasonality in September and October could do the trick. When the systematic strategies start to deleverage nearly $100 billion in assets could be pulled from the stock market, Kolanovic projected.
While the stock market doesn’t need a catalyst, there are numerous potentials lurking.
Marko Kolanovic - Central bank tail risk a concern
Looking at high cross asset class correlations – sometimes a sign of stress when stocks “correlate to one” – and here Kolanovic notices a nuanced world:
First, most cross-asset correlation measures incorporate bond-equity correlation with a negative sign (equivalently, rate-equity correlation has a positive sign, i.e. correlation spikes in risk-off events when bonds and equities move in opposite direction). A potential tail event driven by central banks would happen if bonds and equities drop together. Also, cross-asset correlation measures are backward looking – the current near-record level of cross-asset correlation can in part be explained by a sharp move of risk assets (and bond rally) during Brexit. Indeed, over the past few weeks, cross-asset correlations have started declining.
It is not just that correlations have risen, but what is more important is the rate of change. The fact market correlations have been changing – notably occurring in Brexit – makes market structure analysis more difficult.
High levels of correlation between interest rates and stocks assists strategies like risk parity and volatility targeting, but negative correlation is harmful. Considering the correlations, Kolanovic notes correlation of FX to equities (e.g. EUR/USD vs. S&P 500) spiked to +75% on Brexit and then quickly dropped to -60%. Such a change in correlation is typically linked to a fundamental economic event of significant magnitude, but as major banks revise their Brexit projections to a more positive stance, the quick swings in correlation appears out of place.
The average stock correlation spiked to +70% on Brexit and then declined to only 10% in August. “These record swings in the levels of cross-asset correlation point to a high level of macro uncertainty which makes asset allocation difficult,” Kolanovic noted.
Other analysts have noted different pulls and tugs in the market. “The massive stimulus since 2008 by monetary authorities around the world has had less and less impact on global growth and leaves these organizations with very little dry powder to prevent a deeper and more prolonged recession than experienced in 2008,” Donald Steinbrugge, founder and CEO of Agecroft Partners, wrote in a report Wednesday. “In addition, these dynamics increase the probability of a major market sell-off as central banks continue to artificially prop up markets.”
Marko Kolanovic - Central banks might be able to withdraw stimulus without a market crash, just zero growth for the coming years
Kolanovic, who had previously led the charge regarding to central banks propping up markets, was more muted in his comments when he noted the concern that central bank monetary gymnastics. The Fed withdrawing market stimulus from the economic system would be “negative for a broad range of financial assets” just as it was positive when first injecting artificial buyers into the markets.
Kolanovic explains the historic period that is now being unwound:
Most investors intuitively understand the concept of ‘a rising tide that lifts all boats’. Another concept in analyzing the impact of central banks is the law of communicating vessels. In the analogy, asset classes represent vessels and when funds poured (by central banks) into bonds reach a certain level, they spill over into other asset classes. Flows between the assets are determined by relative yields, often currency hedged yields or risk-adjusted yields. Here is an example of this asset flow: as the BOJ buys bonds (currently over 200% of issuance) and yields turn negative, local investors compare local yields to risk-adjusted foreign yields.
This would result in buying US Treasuries or other assets with higher hedged yields. As all bond valuations turn expensive, bond investors start buying bond-like equities such as low volatility stocks, etc. (e.g. see here). The asset flows can be modeled by calculating correlations between central bank balance sheets and individual assets, as well as correlation between different assets (both on a contemporaneous and lagged basis).
How the central bank withdraws for their stimulative adventures could greatly impact the potential for a significant market value re-adjustment:
if central banks normalize policy very gradually over 3 years and the economy doesn’t stall, one could see near-zero returns for equities over that time period. The rationale is that the average historical return on equities of ~7% would be erased by the withdrawal of CB liquidity (~20% over 3 years). On the other hand, bonds (that don’t have expected price return beyond yield), would likely have negative total return (as ~10% price decline would not be offset by higher yields in the short run). The price adjustment could also happen more rapidly, if CBs move aggressively or market participants position in anticipation.
US political analysis is biased, Marko Kolanovic notes likely stock impulses with Clinton or Trump victory
In addition to central banks being the potential cause of tail risk, US politics are likely to be a cause for market concern.
Kolanovic notes the presidential election is essentially a dead heat with the candidates having very different economic outlooks. Clinton will likely lead to higher taxes on income and capital with increased reliance on public-private partnerships, while Trump advocates a larger fiscal stimulus alongside lower taxes on income, capital and corporations along with reducing regulations.
Marko Kolanovic isn’t buying into the common meme that a Trump presidency will automatically lead to economic ruin, particularly with regards to fiscal stimulus. “We do want to point out that analyses that we came across have been potentially biased (e.g. by ideology, affinity to a certain candidate, etc.),” he writes. “For instance, one such analysis predicts a 2 year recession in case the Republican candidate wins – primarily as a result of an increased debt to GDP ratio and interest expense.”
Kolanovic sees a more nuanced economic result.
If the Republican candidate wins, investors may not engage in year-end selling, expecting capital gains taxes to be reduced under a Trump administration. Year-end selling under Clinton might increase as investors expect higher taxes in 2016. Different stock categories could also benefit or contract depending on the election outcome. Momentum stocks would be winners under Trump but sold under Clinton while value stocks would be treated oppositely, he wrote.