Volatility, your name (probably not) leveraged ETF

Markets have been more choppy lately, and again, some people think leveraged ETFs are exacerbating that. After all, even the Bank of England thinks they might be a little dangerous.

Robin may not blame him, but he’s not a fan either. And he believes the Microstrategy leveraged ETF launched last month could be the industry’s shark-jumping moment. But do leveraged ETFs indeed do they have a higher level impact on stock market volatility?

This is not a new craze. Corroborating Franklin R. Edwards’ old quip that bouts of extreme market volatility tend to be blamed on “whatever happens to be new at the time,” the first concerns about leveraged funds emerged about about 16 years, shortly after they first appeared.

Very much in vogue today, the products were already so popular among professional investors by 2008 that some people apparently assumed they must have exacerbated some of the stock market volatility.

That wasn’t it complete crazy thinking. Two decades earlier, portfolio insurance or “dynamic hedging” involving index futures contracts at the time had only just now “accelerated the pace” of the black moon crash.

But the idea that daily rebalancing of leveraged ETFs, rather than concerns about the longevity of capitalism itself, explained why post-2008 markets traded a little choppy just didn’t hold water, Professor William Trainor argued . “Trading associated with leveraged ETFs does not appear to have any substantial effect” on an index as large as the S&P 500, he concluded in a 2010 paper.

Others disagreed, including Wall Street veteran Douglas Kass: Leveraged ETFs are “new weapons of mass destruction,” he said at the time, and “have turned the market into a casino steroids”.

The leveraged and inverse ETF industry has grown rapidly since then, although it hasn’t necessarily matured.

Assets under management have grown from $30 billion to $140 billion over the past 14 years (about four-fifths of leveraged ETFs are long equity; the rest are inverse or short) with the emergence of products such as long-leverage ETFs for MicroStrategy. , 2x Super Micro Computer ETP and 3x ARK Innovation Long of Leverage Shares.

Buy-and-hold products, they certainly are not: the nature of the beast means that over the course of weeks and months, the performance of leveraged and inverse ETFs often differs significantly from the products they track, particularly on more volatile markets.

So it makes sense that, following the “volume shock” of early August, analysts are once again wondering whether the daily rebalancing trades of these high-risk, high-fee investment vehicles could fuel broader stock market turmoil .

JPMorgan entered the debate this week, starting with some first principles:

The vast majority of leveraged/inverse ETFs. . . must reset or rebalance its exposure daily to bring its leverage back to target levels. They usually place orders near the end of each trading session. This daily reset, which involves the leveraged ETF mechanically buying more of their underlying stock index or stock exposure on bull days or selling on sell days, basically amplifies stock market moves towards the end each trading day.

So how big are these rebalancing flows today and what impact do they have on broader price movements?

Well, JPMorgan analysts Nikolaos Panigirtzoglou, Mika Inkinen, Mayur Yeole and Krutik P Mehta estimate that each percentage change in equity indices generates a rebalancing flow of about $6 billion by leveraged ETFs globally.

In other words, pretty insignificant in the grand scheme of things, but JPMorgan analysts say the impact could be bigger when markets are more choppy.

This $6 billion rebalancing flow corresponds to a one percentage point change in the underlying equity indices, and its amplifying impact would be much greater on days when equity market swings are more violent and liquidity conditions are usually affected.

On the other hand, they admit (citing a 2014 Federal Reserve article titled Are concerns about leveraged ETFs overblown?) that “contrarian” capital flows mean that the size of the moves associated with leveraged ETF rebalancing “is not always as large as its own calculations suggest.”

The Fed document notes that capital flows can either increase or decrease an ETF’s rebalancing demand “because flows change the size of an ETF, which in turn affects the amount of additional leverage the ETF requires to -maintains its target leverage ratio”.

Imagine for a second that you bought an Alphaville 5x BryceCorp ETF. . .

• On Monday, the price of BryceCorp’s underlying stock (generously priced at $100) is up 2%, so the ETF is up 10%.

• BryceCorp ends the session at $102 and the ETF ends the day at $110 (with an exposure of $510)

• If the fund manager did not rebalance at the end of the day, the ETF’s leverage would fall to approximately 4.6x (510/110). So he buys another $40 worth of exposure to fulfill his 5x mandate

• But if the investors participate in the profit by taking money out of the ETF on the same day (for example, a negative capital flow of -$5), this reduces the total purchase amount that the ETF provider must also make, which is $40 — $5 = $35. of rebalancing

• Thus, a (usually) contrarian flow of capital into leveraged ETFs — which i am more used as a short-term trading tool — would somewhat reduce the leverage effect from buying (or selling) at the end of the day.

Under a mean-reversion trading strategy, for example, “investors would increase (decrease) their exposure to leveraged ETFs following negative (positive) index returns,” according to the Fed document. Momentum traders would do the opposite. In other words, some of the movements induced by the ETF at the end of each day should (theoretically) be arbitrated.

The authors thus concluded that capital flows “significantly reduce the demand for ETF rebalancing and therefore mitigate the potential for ETFs to amplify volatility.”

JPMorgan only partially agrees (emphasis ours):

Such headwinds occurred in the most recent stock market correction, as shown in Figure 18. For example, on September 3rd, during an intense stock correction, we estimate that leveraged ETFs and Inverse ETFs had to sell $14.5 billion of underlying equity exposures, but investors injected $1.5 billion. bn during that day, thus offsetting part of the rebalancing flow.

However, on days when the stock market rebounded, i.e. on 22 July, capital flows acted more as an amplifier than a dampener, i.e. capital inflows meant that on that day the ETFs with of leverage had to buy $3.4 billion more than the rebalancing flows of $8.5 billion.

In general, on days when stocks are significantly lower, capital inflows tend to dampen equity selling of leveraged ETF rebalancing, while on days when stocks are significantly higher, capital inflows tend to increase buying of stocks linked to rebalancing leveraged ETFs.

However, Asym 500’s Rocky Fishman has “some concerns” with JPMorgan’s logic. He points out that capital flows occur throughout the day, rather than around market close.

As a result, they will not have the same point-in-time amplification effect as rebalancing flows. I think for some products capital flows could only be allowed at the close, but in this case an order resulting in a creation/redemption will be covered early in the day, so economically it is early, regardless.

Capital flows in/out of leveraged ETFs are only one type of offset against rebalancing flows; in reality, all ETF and mutual fund flows should offset rebalancing flows.

Fishman also points out that blaming end-of-day rebalancing for early August volatility overlooks the fact that most of the action a month ago “occurred outside US trading hours”:

“On a day-to-day basis, (the impact of leveraged ETFs) is not that visible,” he adds. “But they could become a factor if there is a sudden move at the end of the day in an illiquid market. If an event occurs at 15.58 on a trading day, say, these products could have an impact.”

In conclusion, leveraged ETFs appear to contribute to S&P 500 volatility at the margin, with smaller indices more vulnerable to their end-of-day flows and macro factors such as rates, CPI, employment numbers, etc. probably deserving most of the blame. Just like in 2008.