Fresh off releasing a 28-page report on the low vol. regime called “The Upside Of Boring,” Goldman is out on Wednesday asking an important question: “Are Vol. Selling Strategies Crowded?”
Obviously that’s one of the most pressing questions for markets these days, as it’s become abundantly clear that the vol. sellers are perpetuating carry trades and amplifying the feedback loop that’s made vol. selling the strategy du jour in the first place. All of this with central banks’ implicit blessing via the ever-present liquidity flow backstop.
More than a few strategists have warned that this is becoming exceptionally precarious. The likes of Marko Kolanovic, for instance, have warned that due to the low starting point, some of these strats will face “catastrophic losses” on a nominally/optically small VIX spike.
Deutsche Bank has delivered similarly ominous warnings, noting that inverse and levered ETPs would need to buy VIX futs into a spike, potentially exacerbating things at the worst possible time.
In Wednesday’s note, Goldman begins by explaining just why it is that everyone’s asking if they should jump on the vol. selling bandwagon (spoiler alert: they’re asking because the returns are massive):
Recent volatility is low and options selling returns are strong. Volatility has been low and volatility selling strategies have produced strong risk-adjusted returns over the past several years, accelerating in the past year. For example, selling 1-month VIX futures has yielded a total return of 197% over the past year (see Exhibit 2). The average 1-month realized volatility of the SPX has been 9.1% while the options market has priced in an expected (implied) volatility of 13.5%. A variety of options sellers have benefitted from lower volatility than was priced in.
With that out of the way, John Marshall and Katherine Fogertey (who are always looking out for your best interests) get right to it, asking “should investors sell vol.?”
Here is their assessment…
Via Goldman
We see an assessment of flows as only one part of assessing the risk-reward of an investment. The strength of the fundamentals behind the investment is of primary importance followed by valuation and crowding.
1. Current fundamentals support low volatility environment. Our analysis of the correlation of volatility with major macro variables suggest that US GDP, ISM, Employment growth are consistent with low levels of volatility.
2. Equity Valuations have risen, but cash flow remains high. We find Free Cash Flow yield is the metric most closely tied to downside volatility risk for equities, whether used as a time series or cross-sectional signal. The FCF yield of the S&P 500 (exFinancials) of 4.1% is near 30 year median levels, suggesting there is not an unusually large probability of a large drawdown. Specifically, using our GS-EQMOVE model, that incorporates FCF yield and other macro variables, we estimate a 9% probability of a 1- month 5% down-move in the SPX in the current fundamental environment.
3. Volatility is a mean-reverting asset, but mind the gap (between implied and realized). It is true that volatility is near the bottom of its multi-decade range and will increase at some point; however, investors need a large increase to offset the significant carry cost associated with buying options. We believe option buying should be done selectively ahead of events that have the potential to drive volatility.
The answer to whether you should sell vol. then, is “maybe.” Or perhaps “probably.”
But one thing’s for sure, if you’re going to be the guy that bets on a vol. spike, you better get the fucking timing right, because if you don’t, you’re probably just setting money on fire.
Relatedly, Goldman notes that according to the bank’s assessment of ETP positioning, investors are actually buying VIX:
Open interest in VIX products has been rising, suggesting investors are increasingly using VIX. While some of this is a secular trend, part may be cyclical. VIX futures open interest is up 23% over the past year and VIX options open interest is up 20% over the past year. However, net positions in ETPs show that the end investor is actually more “Longvolatility” than “Short-volatility”.
It seems investors are opportunistically using the decline in implied volatility to initiate long positions in the VIX. We estimate that investors are net Long $60 million Vega (volatility exposure) after considering the positions in all major VIX ETPs. For every 1 point increase in the VIX, the aggregate ETP positions would be expected to GAIN $60m rather than lose. Admittedly this is less long than normal; this is down from $155m a year ago, and below the 3 year average of $100m.
Finally – and perhaps most importantly – Goldman asks and attempts to answer the question a whole lot of folks have been asking since the start of the year.
Namely this: “Is delta hedging by market makers causing lower volatility?”
Here’s their answer:
If investors are selling options, then we would likely see the long options positions held at market makers increase (every contract has a buyer for every seller).
Investment firms and Market Makers are required to file their holdings each quarter in their 13F filings. This includes long positions in options as well as stocks. Mutual funds are required to file shortoptions positions (not shown), but others (such as Hedge Funds) are not. By monitoring the long-options positions of investors that act as Market Makers, we can assess whether there is a growing imbalance on the other side of the trade.
We include 19 firms that are explicitly brokers or typically operate like Options Market Makers. We find that the total notional value of positions at Options Market Makers has increased modestly over the past two quarters in terms of absolute level and relative to other options market participants. We estimate Market Makers had $690 billion in long options positions at the end of 1Q2017 and were 73% of the reported long options positions in the market. This is up from $500 billion in 1Q2016 and 71% of reported long options positions.
While we do not put much stock in the absolute numbers as they do not account for the entire market, we note the increase year-over-year in both of these numbers. Our data shows that it is likely that there is a more delta-hedging in the market that would have the effect of dampening volatility.
