Intuitively, this is a toss-up: you have a 50-50 chance of getting more or less money if you switch. However, for problems of this sort (which include the Neck Tie paradox), the Expected Value (EV) is the standard approach. Formalized by Blaise Pascal, Pierre de Fermat, and Jakob Bernoulli in the 1600s, the EV calculation adds up the probability-weighted values of all the possibilities, which results in the value you should expect to see on average.

Applying the EV calculation to the two-envelope problem, the classic argument is that you hold an envelope with an unknown value, A. The other envelope has a 0.5 probability of either being A/2 or 2A, so the EV, the sum of the value weighted probabilities, is 0.5*A/2 + 0.5*2A = 5/4*A. Standard decision theory asserts that, statistically, if you have a choice, you are better off switching to an unknown with the higher expected value. The Monty Hall Problem is a good example of where switching from your initial pick to another is the statistically correct choice. Since you are holding an envelope containing an unknown amount A, and the other envelope has been calculated to have an expected value of 5/4*A, standard decision theory suggests that you should switch.

This implies you should always switch, which in turn implies no envelope should ever be kept—a contradiction under symmetry. Numerous formal resolutions exist, typically involving improper priors, misapplied conditionals, or a contradiction of reference in the expected value formulation.

A simple resolution of the paradox: from the information provided, the assumption that the other envelope is equally likely to contain A/2 or 2A is correct. However, what the typical construction of the classic expected value calculation for the two-envelope paradox misses is that the proposed calculation, with the values 0.5 A/2 and 2*A each having probability 0.5 is ill-formed. In the expected value calculation, your can’t incorporate non-shared values from two different realities. There are two possible sets of values, either A:2A, or A:A/2, but only one of these sets can exist at any point in time. They are mutually exclusive; an expected value calculation that uses non-shared values from each of two mutually exclusive situations is invalid.

The set of possible values a random variable can hold is called its support (the set of values it can take on with a non-zero probability), the values used in its expected value calculation must be present in the support of the random variable. The classical expected value calculation is invalid because it assumes the other envelope’s amount can be either A/2 and 2A with 0.5 probability, but the support of the random variable includes only one of these values, as the two envelopes’ amounts differ by a factor of two (e.g., B and 2B), not four.”

A correct way to determine the expected value of the paradox is to consider the actual amount of money in the envelopes, let’s call them B and 2B. When you pick an envelope, the chance of you picking the one with B is 50%, and the chance of you picking the one with 2B is also 50%. The expected value of your initial choice is:

Expected Value of initial choice = (0.5×B) + (0.5×2B) Expected Value of initial choice = 1.5B

The expected value of switching would be the same, as you’re just swapping one random variable for another. The paradox dissolves when you recognize that the expected values of both choices are equal, and therefore, there is no advantage to switching.

The Expected Value is a foundational statistical metric, along with the median and the mode, but that doesn’t guarantee it is a meaningful measure in every context. In the case of the two envelopes, our intuition is correct, and the ill-posed “expected value” is a mirage.

• Current Status: Frank’s method moved to 100% long SPHB 11-April-2025
• 50% short from 28-Feb-2025 to 11-April-2025
• 100% long from 13-Sept-2024 to 28-Feb-2025

Introduction

In 2013, I wrote The Modified Ned Davis Method, an article posted on Six Figure Investing, which described a greater than 50-year out-of-sample test using primarily the Russell 2000 index. The test continued until recently, posting trade updates on Six Figure Investing and X within a few days after each trade was made. Even though results were computed only using price movements, not counting dividends or interest income on cash, the method significantly outperformed a buy and hold strategy with both the Russell 2000 and the S&P 500 index.

This method began from the 4% Rule designed by Ned Davis and described in Marty Zweig’s Winning On Wall Street, first published around 1986. Davis’s method used only weekly closes of the Value Line Geometric Index. A buy signal was given by any 4% or greater gain, and a sell and short signal was given by any 4% or greater decline. The Value Line, and later Russell 2000, had betas of about 1.2. Compared to the S&P 500, they would typically rise by a greater percentage in a bull market, and decline by a greater percentage in a bear market. I wanted to create a method that would profit in both bull and bear markets, so these were good candidates.

The Modified Model

But the 4% Rule needed improvements. It tended to sell and go short too soon, and most shorts ended with a loss. I improved things by:

• Constructing a rising trend line from the low before the beginning of a buy signal. Price would need to drop below both the sell threshold and the trend line to signal a sell.
• Enabling a short only when the daily NYSE breadth advance-decline line was making a bearish divergence from S&P 500 price.

The latter is an idea I got from Stan Weinstein’s Secrets For Profiting in Bull and Bear Markets, first published around 1983. At a sell, if there was no divergence, the method would sell only 50%, returning to 100% long at the next buy signal. If there was divergence, the method would exit 100% and sell short, but only at the 50% level. (Bear markets are too volatile for my taste.)

I also tried buy & sell thresholds other than 4% and discovered that 4% usually was not the most optimal value. Finally, I allowed the buy and sell thresholds to be different from each other.

My method thus was controlled by 3 parameters: buy threshold, sell threshold, and trend line slope. I then realized that I could run a forward, or out-of-sample, test using a composite index that began with Value Line at its inception, and switched to Russell 2000 when it began in 1979. It would work as follows: It would make the first trade using the “default” values of 4% for both thresholds and 45% for the trend line slope. When that trade ended, that result would be recorded. Then a back test would be run from the index starting date to the end of that trade, using all possible (and reasonable) values of the 3 parameters, selecting the triple that performed the best. Then the next trade going forward would be made using that triple, and that result recorded. This procedure was repeated for all available data.

The method did fairly well through the 2022 bear market, but began to degrade in 2023 as small caps began to go out of favor. About this time I happened to learn of the SPHB ETF, which is based on the S&P 500 High Beta Index. Its beta is about 1.5. I tried my method on this ETF via a back test beginning in 2012, using my original program that selects the parameter values that provided the best performance through 2022. Results were encouraging, but I was reluctant to publish anything because there was not enough data available to make an extensive out-of-sample test. All I had was an 11-year optimized back test.

But I kept rerunning the back test as more data became available, and the same parameter values were always chosen. I was fully invested in a number of ETFs, including SPHB, when the market began to decline early in 2025. The SPHB method gave a 100% sell signal on Feb. 28, so I sold just about all of my holdings at that point, deciding to purchase mostly SPHB at its next buy signal, which came on April 11. The method is still using the same parameter values that it selected more than 2 years ago. SPHB has gained nearly 40% since the April 11, 2025 buy. Unless stated otherwise, my positions will match the current status indicated at the beginning of the post.

The chart below shows the weekly results of this method, plus buy and hold of both SPHB and SPX, from 2012 on. The 2012 through 2022 results are an optimized back test, so take that for what it may be worth, but the results since 2023 are out-of-sample, using the parameters that the back test selected at the end of 2022.

Since the holdings of the S&P 500 High Beta Index are adjusted over time to include only those with the highest beta, I believe that this method is likely to continue to do well, not being dependent on the continuing superior performance of any one index or sector. As I did with the original Value-Line/Russell 2000 method, I will post all future trades of this method here as soon as possible after they occur.

The full sequences of moves is shown below:

This paper is not about making money, but rather about a fundamental theoretical error regarding stock market forecasts that has been present for many years.  The flawed methodology generates overoptimistic multi-period forecasts.  Practitioners and academics have developed workarounds for the problems created by the error, but they never identified the root problem. This paper does that and presents an alternate approach that resolves anomalies that financial analysts, Blume, Jacquier, and Blitz identified years ago. The reason this error has gone undetected for so long is likely because there was more than one error. One error masked the other.

The errors:

1. The Expected Value (EV), the probability-weighted sum of the values of a random variable, is a fundamental measure in statistical analysis, but it is not perfect. It suffers from the same issue as the arithmetic average, that for positive only, fat-tailed distributions like lotteries and stock markets returns, large outlier values bias the measure away from the typical central tendency value. In other words, in the case of the stock market, as an investor you shouldn’t expect the Expected Value.
2. When researchers, long ago, evaluated the forecasts based on the incorrect EV measure, their validation passed. What they didn’t realize is that both the incorrect measure (EV) and the correct measure give the same result for a one-period test. Discrepancies become apparent only when examining more than one period (see chart from the paper below)

This research aims to shed light on these overlooked errors and provide a more robust framework for market forecasts. By addressing the root issue, it offers a pathway for more accurate and reliable financial analyses.

Before we get into the specifics of how ETHU works, I’m disclosing that I hold a small, minority share in Volatility Shares LLC, the issuer of ETHU. Volatility Shares has also issued BITX, ZIVB, SVIX, and UVIX ETFs. I do not hold management, executive, or operational roles in Volatility Shares LLC, nor do I give recommendations or investment advice to their clients.  My analysis of these funds and their associated indexes is my own, is not investment advice, and is based on publicly-available information.

ETHU is a 2X Leveraged ETF

Architecturally, ETHU is structured as a 2X leveraged, daily-resetting ETF. Volatility Shares, the issuer of ETHU, seeks for it to deliver twice the daily percentage movements of the ether cryptocurrency. Typically, a 2X leveraged ETF uses margin borrowing to purchase enough of the underlying security to achieve the 2X leverage. In contrast, ETHU uses Ether futures that trade on the CME exchange rather than the ether cryptocurrency to achieve the desired leverage. I believe this was a wise choice, both for pragmatic reasons and for long-term stability and security. From a practical standpoint, the decision to use Ether futures instead holding ether directly may have been a key feature in Volatility Shares’ proposal for ETHU, which ultimately led to its approval by the U.S. Securities and Exchange Commission (SEC). For quite some time, the SEC had rejected all applications for Exchange Traded Funds that proposed to directly hold ether.

Cryptocurrencies like ether are traded on geographically diverse exchanges, most of which are not based in the U.S. and therefore not subject to U.S. securities laws or the SEC. While the Coinbase exchange, where most USA institutional trading of cryptocurrencies occurs, is based in the U.S., it is relatively new and lacks many of the safeguards that have been time-tested in U.S.-based futures exchanges. The Ether futures held by Volatility Shares for ETHU are traded on the CME Futures exchange, a U.S.-based exchange that has been operating for over a century. The CME runs a time-tested process that includes a clearinghouse intermediary, which protects the interests of those holding Ether futures, even when counterparties fail to meet their obligations.

In my opinion, using a regulated futures exchange with established clearinghouse processes is a significantly safer approach than holding ether directly and trading on exchanges with much shorter track records, minimal insurance, and relatively untested processes for managing inevitable stresses like counterparty bankruptcies.

Many people equate futures with speculative, widely varying price levels—which is true in some cases, but not with ether. Ether futures don’t have significant seasonality, storage, or supply/demand issues, so their prices tend to closely track the current “spot” price of ether. The main difference between the two is interest rate expenses and the cost of periodically rolling futures to longer-dated ones before they expire.

Unlike their other ETFs, Volatility Shares does not specify an index to determine the mix of Ether futures that should be held at any given time. ETHU is actively managed, meaning the specific mix of Ether futures held by the fund at any given time will be determined by its managers. This active management approach reduces the amount of information available to external traders (minimizing front-running opportunities) and gives the fund’s managers flexibility in selecting which assets to trade. For example, if longer-term futures are favorably priced, the fund’s managers can increase their allocation of those futures. Unlike exchange traded notes (such as Barclays’ VXX), where the net asset value (NAV) is determined by the value of an index, the NAV of an ETF is determined by the value of its underlying assets.

Dividends

Since its inception, ETHU has been paying a small monthly dividend. I’ve not been able to determine the algorithm used to establish the dividend amounts, but the dividends seem to average around 0.08% of the share price right before the dividend is declared. At current levels, the dividends are not economically significant. Perhaps these small dividends are a placeholder in case Volatility Shares decides to increase them later. For, example, Volatility Shares’ 2X Bitcoin ETF, BITX is distributing about 2% of its share price each month.

Since Inception

ETHU gained assets rapidly after inception and is currently (as of Oct 2024) holding around $100 million in assets under management. Its average volume is a very respectable 2.5 million a day, and its bid/ask spread is a penny. Options are available on ETHU. Clearly, a successful launch.

A ETHU Simulation

The long-term performance characteristics of an ETF seeking to deliver 2X the percentage moves of the notoriously volatile ether cryptocurrency is not something easy to predict.  We can improve our intuition on this by simulating how ETHU would have performed in the past.  Of course, there is no guarantee that ETHU will behave similarly going forward. My backtest uses the end of standard NYSE trading hours closing value of ether, starting November 17th, 2017, to drive the simulation.

Before looking at the results of the ETHU simulation, it’s worthwhile to look at ether’s historic moves, the chart below covers its price moves from its 11-November 2017 value of 300 through its 23-September 2024 value of 2648.

Ether’s run up from 360 to 2648 (7.3X) over these 7 years has had three major peaks, one in 2018, one in 2020/2021, and one in 2024. To compare ether’s performance against simulated ETHU, I calculate two portfolios, both starting with $100, one holding ether and the other holding ETHU. The range of values is so large I use logarithmic scaling for the vertical portfolio value axis.

The simulated ETHU portfolio spikes up 10X to over $1K in 2018, but then dramatically erodes during the subsequent downswing in ether in 2018 and 2019. After 7 years, the ETHU-based portfolio is at $10, down 10X from its $100 starting point. A dreary performance, but objectively a significantly less bad performance than what the leveraged long volatility ETFs delivered over that same period. See How Does UVIX Work? for more on this subject.

It’s well known that leveraged ETPs are rarely good buy-and-hold style investments, but what if a skilled and/or lucky trader bought ETHU a month before the start of ether’s massive 2020/2021 rally? The chart below compares the performance of $100 invested in ether vs ETHU starting on 15-Apr-2020.

The results in this case are eyepopping, with a 21X run-up in ether, and a 280X (!) increase in ETHU. How is this even possible with a fund advertised as being a 2X leveraged fund? The answer is that in trending markets, without significant pullbacks, a leveraged fund can significantly outperform its leverage factor. For example, if ether had 3 consecutive days of +20% moves, after 3 days it would have increased by a factor of 1.2^3 =1.728, a 72% increase. ETHU on the other hand, would have ideally gone up 40% each day, so 1.4^3 = 2.744, a 174% gain rather than the 2 x 72 = 144% you might have expected from a 2X fund.

Of course, trends like this are rare. When the price action is more random, with more up and down, a dark side of leveraged ETF emerges—it’s called volatility drag. For ETHU, volatility drag is a significant factor that will often erode performance as will be discussed later in this post.

Summarizing, the simulation shows that for the past performance of ether, a long term position in ETHU will likely lose most of its value, but if you catch a rally you can do very well.

This ETHU backtest simulation spreadsheet is available for free here.

Some Observations on the simulated ETHU Backtest

The following table summarizes the best and worst-case percentage moves over various periods for simulated ETHU starting November 2017

ETHU risks:

Not surprisingly, the ETHU backtest simulation suggests that a 2X leveraged ETF tied to the notoriously volatile ether cryptocurrency will have a lot of price volatility.  Some less obvious risk factors associated with leveraged ETFs are detailed below.

• Volatility Drag
  
  All leveraged ETFs have a loss mechanism called volatility drag, sometimes called Beta Slippage that reduces returns below what a trader might expect over the long run. Volatility drag in leveraged ETFs occurs when daily security prices follow a noisy, up-and-down path.  Leveraged ETFs must rebalance their assets at the end of every day to achieve their leverage goals.  If the day-to-day price action is noisy, that rebalancing activity generates losses.
  
  Conceptually, a leveraged ETF uses a momentum strategy, assuming that the market will continue in the direction of today’s closing value relative to the previous close. To achieve its goal of matching its leverage factor in percentage moves of the underlying, it must rebalance its assets at the end of the day to be ready for the next day’s move. However if, instead of a continued trend, the next day’s movement reverses and the price goes the other direction, the ETF will lose some value. The extent of this loss depends on the magnitude of the reversal.  For a 2X leveraged product, the average daily volatility drag will be approximately the daily volatility of the underlying index squared. The daily volatility of ether is ~5.6%, so ETHU’s volatility drag will be around (.056)^2 ~0.3% per day. This might not seem like a lot, but over a year, with approximately 250 trading days, the cumulative loss would be (1-.003)^250 -1 = -0.45, or 45% per year. See this link for more information on volatility drag.

• Big Down Days and Essentially Unrecoverable Losses
  
  Big down days are treacherous for 2X leveraged ETFs.  If the underlying index or reference security of a 2X leveraged ETF drops 50% or more within a single trading day, the ETF’s price will likely drop to zero, or near zero. It is important to note that despite the large drawdown potential for ETHU, its price cannot go below zero, ensuring that investors can’t lose more than their initial investment. This characteristic contrasts with the situation where a trader buys a security and then borrows via margin loans to buy an equal amount of the security to achieve 2X leverage. In that case, the trader can lose significantly more than their initial investment if the security drops more than 50%.
  
  If ETHU’s value were to drop to near zero intra-day, the fund’s operators would likely act to dramatically reduce the number of Ether futures it holds.  A 2X fund uses leverage to effectively hold $2 of the underlying asset for every $1 invested, so if the futures held by the ETF drop in value by 50% or more during a single day, the losses could equal or exceed the asset value of the fund, leading to an end-of-day share price at or near zero.  In this scenario, the ETF issuer is exposed to huge losses if the underlying continues to drop in price.  To protect itself, the fund would likely sell assets, and/or have hedges in place, to reduce its effective leverage. Somewhat smaller daily losses, e.g., 45% in the underlying, would probably not trigger deleveraging of the fund, but could still inflict grievous (e.g., ~90%) decreases in the 2X ETF’s price.
  
  A less obvious risk is that there’s no realistic way to recover from a dramatic full-day drawdown on a 2X ETF, even if the market recovers spectacularly the following day.  If ETHU is down 90% at market close, a 30% recovery in Ether futures the next day will not help much; it would only bring the fund’s value back to 16% of what it was two days prior.
  
  A trader positioned for 2X leverage via margin loans rather than a 2X leveraged ETF would do much better in the two-day “down then up” scenario. With the futures down 45% one day and then 30% up the next day scenario, the margin-based trader would only be down 28.5%.  However, in the “down then down more” scenario they could lose up to twice their investment, whereas losses in a 2X ETF are limited to the amount invested.
  
  The CME exchange does have general market circuit breakers which are coordinated with the stock markets and in addition, the CME’s cryptocurrency futures have their own dynamic circuit breakers.   
  
• Is the Ether Adventure Over?
  
  Ether has had an incredible run-up in price since its inception 7-August-2015. Often the past is a good predictor of the future, but there are no guarantees that ether’s long-term upward trend will continue.  If it doesn’t, that’s bad news for long-term holders of ETHU.

Ether Opportunities

• Ether’s Price Goes on Another Run-Up
  
  The ETHU backtest shows there’s potential for spectacular gains if ether goes on a bull run.  The ETHU simulation shows three periods with huge gains in the last five years, 2017 (10X), 2020/2021 (230X), and 2024 (5X).

• Multi-day Upward Trends in Ether
  
  Some of ETHU’s periodic spectacular simulated performance is due to the flip side of the volatility drag phenomenon that was described in the risks section.  As mentioned earlier, a leveraged ETF effectively implements a momentum strategy, and if the underlying does go up for multiple days in a row, then the ETF can outperform relative to its leverage factor. For more on this, see When Leveraged Funds Outperform Their Leverage.

Fees, Regulatory Status, and Taxes

ETHU’s effective annual fee is 0.94%, which is deducted daily (~0.0026%) by Volatility Shares from the fund’s assets.  ETHU is registered under the US Investment Company Act of 1940.  My understanding is that ETHU will report gains/losses via IRS 1099 forms, will not issue K-1s, and that ETHU is not subject to non-resident withholding on publicly traded partnerships (PTPs).  Please consult with your tax advisor regarding your specific tax obligations. 

Trading Notes on ETHU’s Positions in Ether futures

• Each trading day the fund will likely roll some of its soonest-to-expire futures to futures with more time until expiration,
• Futures will be bought/sold to adjust for share creations/redemptions in the fund,
• To position ETHU to deliver 2X leverage the next day. The number of futures held by the ETF will be increased or decreased depending on the current day’s ending percentage move from the previous trading day.

Trading Strategies

Although most holders of ETHU will likely be short term traders, some will try to capture a future sustained run-up of Ether prices by just buying and holding.  However, even if that strategy is successful, any windfall gains will likely fade quickly away if ETHU is held for a long time.

• In practice, identifying the peak value of an ether rally is virtually impossible, so strategies that take profits as they occur, for example selling shares to cash out ten percent of the profit every week, are attractive approaches that don’t require the trader to pick the peak.
• Those long ETHU should consider mitigating potential losses from big down days by holding options that go up in value if ETHU drops dramatically in value.  Options are available on ether, Ether futures, and on ETHU.
• Traders who believe Ether is going to drop in value can short ETHU.  If their forecast is correct, they will benefit from ETHU’s leverage, volatility drag, and fees.  Buying out-of-the-money calls to limit the maximum loss if their forecast is wrong is a best practice. Holding these calls will also likely reduce upfront margin requirements.

Conclusion

ETHU is not intended to be a buy-and-hold investment. Even if ether goes on to set new all-time highs, investors that buy ETHU and never sell will very likely end up disappointed. On the other hand, ETHU offers an aggressive trader a product that’s able to significantly compound ether gains, without the need for additional capital, or risking more than they invested.  

Resources

• How Do Bitcoin Futures Work?
• Backtest of ETHU since 2017 (free)

Before we get into the specifics of how BITX works, I’m disclosing that I hold a small, minority share in Volatility Shares LLC, the issuer of BITX. Volatility Shares has also issued ZIVB, ETHU, SVIX, and UVIX ETFs. I do not hold management, executive, or operational roles in Volatility Shares LLC, nor do I give recommendations or investment advice to their clients.  My analysis of these funds and their associated indexes is my own, is not investment advice, and is based on publicly available information.

BITX is a 2X Leveraged ETF

Architecturally, BITX is structured as a 2X leveraged, daily resetting, exchange-traded fund.  Volatility Shares, the issuer of BITX, seeks to have the ETF deliver twice the daily percentage moves of a mix of short-term CME Bitcoin futures.

It’s not surprising that Volatility Shares chose to use Bitcoin futures rather than Bitcoin as the underlying security for BITX.  For a long time the US Securities Exchange Commission (SEC) rejected all the applications of proposed USA-based Exchange Traded Funds that directly hold Bitcoin, reportedly at least partially due to concerns about the use of crypto exchanges like Coinbase. The Bitcoin futures used by BITX use the CME exchange, a USA based exchange that has been in operation for over a hundred years that includes a counterparty arrangement that further protects the interests of those holding Bitcoin futures.

I will get into some details later in this post, but unlike some other futures types, the historical daily percentage moves of Bitcoin futures have closely tracked Bitcoin’s moves and this close correspondence is likely to continue.   

Specifically, BITX seeks to deliver 2X the daily percentage moves of  S&P Global’s SPBTFDUE index.  This index documents a process for rolling Bitcoin futures from sooner-to-expire futures to longer-to-expire futures.  While this index provides guidance to Volatility Shares for managing BITX’s assets, Volatility Shares does not guarantee that BITX’s mix of futures or performance will exactly track the index.  The fund’s official Net Asset Value (NAV) will ultimately be decided by the value of the assets it holds, not by SPBTFDUE’s value.

A BITX Simulation

The long-term performance characteristics of a 2X leveraged ETF holding futures on the notoriously volatile Bitcoin challenges our intuition.  One way to inform our intuition is to simulate how this product would have performed in the past.  Of course, there is no guarantee that BITX will behave similarly going forward. 

S&P Global has computed values for BITX’s tracking index, SPBTFDUE, starting from late 2017, when CME Bitcoin futures started trading.  Simulating BITX’s performance from that point forward is straightforward.  I extended the simulation back to September 2014 (when a Bitcoin was trading at around $450) by assuming that Bitcoin futures prices, had they existed at that point, would drop at the same average rate relative to spot Bitcoin (~4% annually) as in the 2017 to 2023 period.  

The chart below shows the results of the BITX simulation.  The price per share is scaled such that the 26-Jun-2023 price aligns with BITX’s $15 inception price.

The log-scaled chart below provides visibility into the earlier years and adds a comparison to historical Bitcoin prices.

The chart below compares the performance of $1000 invested in Bitcoin vs BITX starting on September 17, 2014.

This BITX backtest simulation spreadsheet is available for free here.

Some Observations on the BITX Backtest

• Despite Bitcoin’s huge price gains from 2014 to 2023 (on average 60%/year) a 2X leverage product would have delivered much less than a 120%/year gain.  My simulation suggests around a 37% annual gain.  Still impressive, but the associated ups and downs along the way would have made a buy-and-hold strategy an adrenaline-soaked experience with multiple episodes of riches to rags. 
• A bad time to buy BITX during the backtest period, was December 18^th, 2017.  One thousand dollars invested then would, as of June 27, 2023, be worth about $100; an average loss of around 35% per year.
• Unlike some leveraged long ETFs, the backtest suggests BITX won’t be on an inevitable ride to zero. Certainly, the future performance of Bitcoin will be the primary driver in BITX’s performance, but the relatively low carry costs associated with Bitcoin futures suggest that BITX’s cumulative contango losses, which decimate some ETFs funds (e.g., the VIX futures-based VXX, UVIX & UVXY) will be relatively low.
• During Bitcoin upswings the BITX crescendos are very short-lived, a trader long BITX would have to either be lucky or have a very good sense of timing to exit near the peak values.
• The following table summarizes the best and worst-case percentage moves over various periods for Bitcoin (BTC) and BITX starting in January 2019

The table above shows some of the benefits of BITX’s daily resetting ETF structure.  For multi-day periods, BITX’s effective leverage is lower than 2X for down-trending periods, and better than 2X, sometimes dramatically, for up-trending periods.  The reason for this is discussed in the next sections.  

For single-day moves, BITX’s simulated best-case gain is 2X of the BTC move, as you would expect, but on the worst loss side, the simulated BITX only moved -47% compared to the 37% move of Bitcoin.  The worst-case day was March, 12^th, 2020, during the worst of the Covid crash.  The CME’s bitcoin futures drop was less than the full BTC drop that day, probably because either the USA markets were closed when the drop happened, or the drop was very short-lived.

• The scatter chart below shows the per day BTC percentage moves vs BITX’s underlying index moves.  The basic correlation is a one-to-two mapping in percentage moves, but it’s a fairly noisy relationship.  The limited trading hours of the Bitcoin futures/BITX are probably the biggest cause for the differences.  Some of the largest deviations from the 1:2 mapping are labeled with the dates they occurred.

• While BITX only trades during typical USA trading hours, Bitcoin trades continuously.  This raises the question of how large the typical Monday moves of BITX will be.  At open, Monday mornings, BTC will have been trading for approximately 64 hours (~2.66 days) since the US markets were last open.  Standard volatility theory asserts that volatility increases with the square root of time, so it predicts Monday’s volatility will be 1.63 times the normal daily volatility. The chart below shows the simulated close-to-close day-of-week sensitivities.
  

Monday’s daily volatility is 11.6%, a bit higher than the 9% daily average, but well below the predicted 14.7%.  Perhaps this shortfall is due to lower volatility during the weekends.  

BITX risks:

Not surprisingly, the BITX backtest simulation suggests that a 2X leveraged ETF tied to the notoriously volatile Bitcoin cyber-currency will have a lot of price volatility.  Some less obvious risk factors associated with Leveraged ETFs are detailed below.

• Volatility Drag
  
  All leveraged ETFs have a loss mechanism called volatility drag, which reduces returns below what a trader might expect.  Volatility drag in leveraged ETFs occurs when daily security prices follow a noisy, up-and-down path.  Leveraged ETFs must rebalance their assets at the end of every day to achieve their leverage goals.  If the day-to-day price action is noisy, that rebalancing activity ends up generating losses.
  
  Conceptually, a leveraged ETF executes a momentum strategy that assumes that the market will continue in the direction that it moved the current day.  If instead, the next day’s movement reverses and the price goes the other direction, the ETF will lose some value.  The magnitude of this loss depends on the magnitude of the reversal.  For a 2X leveraged product the average daily volatility drag will be approximately the daily volatility of the underlying index squared.  The daily volatility of BITX’s underlying index, SPBTFDUE, averages ~4%, so BITX’s volatility drag will be around (.04)^2 = 0.16%.  This might not seem like a lot, but over a year, with approximately 250 trading days, the cumulative loss would be (1-.0016)^250 -1 = -0.33, or 33% per year.  See this link for more information on volatility drag.

• Big Down Days; Essentially Unrecoverable Losses
  
  Big down days can be treacherous for 2X leveraged ETFs.  If the underlying index of a 2X leveraged ETF drops 50% or more in a single day, the ETF’s price will likely drop to zero.  
  
  If BITX’s value was to drop to near zero intra-day, the fund’s operators will likely act to dramatically reduce the number of Bitcoin futures it holds.  A 2X fund uses leverage to effectively hold $2 of the underlying asset for every $1 invested, so if the futures held by the ETF drop in value by 50% or more during a single day, the losses could equal or exceed the asset value of the fund, leading to an end of day share price at or near zero.  In this scenario, the ETF issuer is exposed to huge losses if the underlying continued to drop.  To protect itself, the fund would likely sell assets, and/or have hedges in place, to reduce its effective leverage. Somewhat smaller daily losses, e.g., 45% in the underlying, would probably not trigger deleveraging of the fund, but could still inflict grievous (e.g., ~90%) decreases in the 2X ETF’s price per share.
  
  A less obvious risk is that there’s no realistic way to recover from a dramatic full-day drawdown on a 2X ETF, even if the market recovers spectacularly the next day.  If BITX is down 90% at market close, a 30% recovery in Bitcoin futures the next day will not help much; only bringing the fund’s value back to 16% of what it was two days before.
  
  A trader positioned for 2X leverage via margin loans rather than a 2X leveraged ETF would do much better in the two-day “down then up” scenario. With the futures down 45% one day and then 30% up the next day scenario, the margin-based trader would only be down 28.5%.  However, in the “down then down more” scenario they could lose up to twice their investment, whereas losses in a 2X ETF are limited to the amount invested.
  
  The CME exchange does have general market circuit breakers which are coordinated with the stock markets and in addition, the CME’s cyber currencies futures have their own dynamic circuit breakers.   
  
  From a risk management standpoint, it’s prudent to assume that BITX could drop more than 2X Bitcoin’s daily percentage move.  So, it’s possible that BITX’s share price could drop to zero, or near zero, even if Bitcoin’s single-day drop was less than 50%.  I discuss some ways to mitigate that risk in the trading strategies section below.
  
• Is the Bitcoin Adventure Over?
  
  Bitcoin has had an incredible run-up in price since its inception in 2009. Often the past is a good predictor of the future, but there are no guarantees that Bitcoin’s long-term upward trend will continue.  If it doesn’t, that’s bad news for long-term holders of BITX.

BITX Opportunities

• Bitcoin’s Price Goes on Another Run-Up
  
  The BITX backtest shows there’s potential for spectacular gains if Bitcoin goes on a bull run.  The BITX simulation shows three periods with huge gains in the last five years, 2017 (100X+), 2019 (14X), and 2020 (22X).

• Multi-day Upward Trends in Bitcoin
  
  Some of BITX’s periodic spectacular simulated performance is due to the flip side of the volatility drag phenomenon that was described in the risks section.  As mentioned earlier, a leveraged ETF effectively implements a momentum strategy, and if the underlying does go up for multiple days in a row, then the ETF can outperform relative to its leverage factor.  For example, if Bitcoin were to trend up 10% a day for three consecutive days, then BITX’s performance would likely be around 72%, rather than the 60% gain you might expect.  For more on this see When Leveraged Funds Outperform Their Leverage.

Fees, Regulatory Status, and Taxes

BITX’s annual fee is 1.85%, which is deducted daily (~0.005%) by Volatility Shares from the fund’s assets.  BITX is registered under the US Investment Company Act of 1940.  My understanding is that BITX will report gains/losses via IRS 1099 forms, will not issue K-1s, and that BITX is not subject to non-resident withholding on publicly traded partnerships (PTPs).  Consult with your tax advisor regarding your specific tax obligations. 

Bitcoin futures

When first introduced in late 2017, there was speculation that Bitcoin futures would not closely track Bitcoin’s daily (spot) price, but Bitcoin futures have proved to be a good proxy for Bitcoin.  This performance is not surprising given how futures really work.  The popular notion of futures is that they are vehicles for wild speculation, subject to huge unpredictable swings.  While some futures do have those characteristics, in the five years that bitcoin futures have been trading, they’ve been quite predictable, mostly driven by mundane things like interest rates.  This historic performance strongly suggests that futures market makers can cost-effectively hedge their positions with Bitcoin.

 Since their inception in 2017, Bitcoin futures with more time until expiration have had higher prices than futures nearer to expiration.  This is a term structure condition called contango. Historically, holding Bitcoin futures and rolling them as needed costs around 4% per year relative to just holding Bitcoin directly.

Volatility Shares may change the quantity and mix of Bitcoin futures held by BITX daily.  There are at least three drivers for those changes:

• Each trading day the fund rolls some of its soonest-to-expire futures to futures with more time until expiration,
• The flow of cash in and out of the fund,
• To position BITX to deliver 2X leverage the next day. The number of futures held by the ETF will be increased or decreased depending on the current day’s ending percentage move from the previous trading day.

Trading Strategies

Although most holders of BITX will likely be short term traders, some will try to capture a future sustained run-up of Bitcoin prices by just buying and holding.  However, even if that strategy is successful, any windfall gains will likely fade quickly away if BITX is held for a long time.

• In practice, identifying the peak value of a Bitcoin rally is virtually impossible, so strategies that take profits as they occur, e.g., selling shares to cash out ten percent of the profit every week, are attractive approaches that don’t require the trader to pick the peak.
• Those long BITX should consider mitigating potential losses from big down days by holding options that go up in value if Bitcoin or BITX drop dramatically in value.  Options are available on Bitcoin, Bitcoin futures, and BITX.
• Traders believing that Bitcoin is going to drop in value can short BITX.  If their forecast is correct, they will benefit from BITX’s leverage, volatility drag, and fees.  However, as shown earlier in the best- and worst-case percentage moves tables, BITX’s leverage in tending multiday declines will likely not achieve 2X leverage.

Conclusion

BITX is not intended to be a buy-and-hold investment.  Even if Bitcoin goes on to set new all-time highs, investors that buy BITX and never sell will very likely end up disappointed.  On the other hand, BITX offers an aggressive trader a product that’s able to significantly compound Bitcoin gains, without the need for additional capital, or risking more than they invested.  

Resources

• How Do Bitcoin Futures Work?
• SPBTFDUE methodology and Data (BITX’s underlying Index}
• Backtest of BITX since 2014 (free)

The rate at which VIX Futures converge to spot VIX varies, and their tracking to the VIX depends on how much time the VIX future has before expiration. Near expiration, a VIX future will converge rapidly to the VIX, and will track changes to the VIX at pretty much a 1:1 ratio.  A VIX future with 3 months to expiration is much less correlated with the VIX value.

Complicating this whole situation is the fact that the ETP’s change their mix of VIX futures on a trading daily basis to shift out of the next-to-expire future into one with more time until expiration. Contrary to most people’s intuition, this roll process is economically neutral, it’s like changing two nickels for a dime; there’s no net loss (see The Cost of Contango)

With leveraged ETPs like UVIX and SVIX, things are further complicated by their need to rebalance their assets daily so that their next-day performance will match their leverage goal. Over time this rebalancing process results in volatility drag that significantly erodes the price of the leveraged ETPs, but in a strongly trending market results in leveraged ETPs outperforming their leverage factor.

Measures of the VIX term structure, e.g., VIX3M/VIX or VIX90/VIX30 give a qualitative feel for how the ETP prices are going to trend, but as you can see, they are fairly distant from what is really going on. 

• VIX1D quotes, download data for the previous 3 months, chart data back to May 2022 are available on Cboe’s Global Index site. Beware, the open/high/low data is the same for all days until the beginning of April. It appears it will be a while before we can get a statistical feel for things like day of week close to open moves.
• The VIX1D is a relatively straightforward extension of the current VIX methodology down to one day, the details are given in this VIX1D whitepaper.
• The real-time value of the VIX1D will represent a dynamic mix of the SPX options expiring today and of the ones expiring the next trading day. By the end of the trading day, the VIX1D will be totally driven by the price of the SPX options expiring the next trading day, which could be multiple calendar days hence (e.g. weekends + holidays).
• VIX1D will be very sensitive to Known-Unknown event risk, where the timing of an event (e.g., Fed report, Election date) is known but the outcome is not. When that event has the potential to move the market, SPX option prices, in particular options expiring soonest after the announcement, reflect the magnitude of the likely move, which in turn impacts the value of the VIX1D index.
• Calendar effects will be present in the VIX1D, which are described in more detail in, No the VIX is Not Broken. Briefly, the VIX calculation needs a normalization calculation to standardize the volatility metric to be an annualized number, but things like weekends and holidays can create unreal artifacts in the index. For example, the VIX and the VIX9D typically have a significant jump on Monday mornings, which is not present in the tradeable VIX futures. With VIX1D this weekend effect will move to Fridays.
• Two important tweaks to the VIX methodology described in the VIX1D whitepaper to make it more suitable for the one-day time duration are:
  • The use of “business” time” vs the standard calendar time for the variance interpolation between today’s and tomorrow’s variance calculation during regular trading hours. This will dampen down some of the weirder effects. For a deep dive into the variance interpolation calculation, you can read Calculating the VIX-the Easy Part, where I discuss/document this process for the standard VIX index.
  • The methodology also calls for ignoring the changes in the current day’s results after 3PM ET, freezing the value at the 3PM level. This prevents last-minute option price funnies common right before market close from unduly influencing the VIX1D’s level.
• With the explosion of 0DTE option volumes, it seems likely that the Cboe would want to get VIX options into the expirations every day game, however using its current processes, VIX options have a 30-day horizon, which react slower than true ODTE options which closely track the current “spot” price of their underlying security.
• One possible goal of introducing the VIX1D would be to facilitate the introduction of everyday expiring VIX options with a one-day horizon. Current monthly and weekly VIX options have a 30-day horizon. To accomplish this the Cboe would have to create a variation on the current auction process, called the Special Opening Quotation (SOQ) to generate the settlement price, because the current one, running typically on Wednesday mornings, settles to SPX options with 30 days until expiration. Currently, VIX options typically only expire on Wednesday mornings, and have a 30-day horizon.
• When/if VIX everyday expiring options come out, they might take the opportunity to change the expiration to the market close, when virtually all other options expire. The current AM expiration is just an artifact of the initial SPX options that expired in the morning (AM). If they did create a new SOC then running it at market close would be attractive, more intuitive, aligns better with other option products, and avoid the sometimes chaotic environment at market open.
• It’s not likely that the Cboe will introduce VIX daily expiring VIX futures any time soon.
• If the CBOE is just adding a new VIX index, then they’ve created a very quirky one. One that virtually no one will really understand, that occasionally generates very high, untradeable values as clickbait.

So far, the Cboe has released 1 year of backtest data for the VIX1D (VIX1D dashboard), but even that little snippet of data confirms that this extension to the VIX family will expand its general ability to confuse people. The chart below compares 14 days of VIX1D (dotted black line) with VIX9D, VIX, and the daily percentage moves of S&P500 (scale on right).

The major two spikes, one on April 6th, and another on April 11th, immediately stand out. Not only do the other indexes not have these big spikes, the VIX1D spikes don’t align with the general S&P 500 market action. Neither of these spikes are reflecting the “true” volatility of the market that day (whatever the hell that is). The first spike is likely the weekend effect I mentioned earlier, and the second spike on April 11th is probably an event-driven spike driven by the April CPI announcement on the 12th.

Regarding the introduction of everyday expiring VIX futures, I think this is unlikely. The Cboe did create weekly expiring VIX Futures around the time they created weekly expiring VIX options, but they have never had much volume, their volumes are much lower than the monthly expiring VIX future. With the liquid monthly VIX futures, VIX options, and VIX-based Exchange Traded Products (e.g, VXX, VIXY, UVXY, UVIX) and their options I think investors and market makers don’t need/want another tenor of VIX futures. Unlike Americal style options, VIX options are European style (not early exercisable) and cash-settled, so they don’t require a physical security to assign to.

The Cboe recently (20-Mar-2023) updated their Special Opening Quotation (SOQ) process in the VIX white paper. This auction, which currently occurs most Wednesday mornings after market open is the mechanism that establishes the settlement price of VIX options and futures that expire on that day. One of the changes in the SOQ process is that the Cboe now specifies the specific SPX option put and call strikes that will be included in the auction. This removes one source of uncertainty in the SOQ auction process that was exploited at least once in the past, in an attempt to game the VIX settlement price. If the Cboe does introduce everyday VIX options with one-day horizons a modified SOC process would need to be created that occurred every trading day to establish the settlement price. A robust final value process for determining the expiration price is very important.

If the Cboe does introduce everyday expiring VIX options, one consistent confusion factor will likely be amplified. For most options, e.g., SPY, Apple, the current stock price is a reasonable proxy for the underlying security that the option is tracking. In reality, options track what’s called the forward of the underlying security, which captures things like interest rate effects. For VIX options the standard VIX indexes are essentially never a good estimate of what the options are tracking. For VIX options expiring at the same time as a monthly future, that future is a good estimate of the forward price, but for something like a VIX option expiring next Thursday there’s nothing reliable. If you just use the current VIX values you could be way off. The best estimate of the forward value for your option is the level where the put and the call of that expiration have the lowest prices. The Greeks that brokers display for these options are often calculated using the wrong underlying, so they can be way off.

In conclusion, these are all preliminary guesses, we will know more as we get more details on the new VIX1D index.

According to its prospectus, ProShares can reverse split VIXY at any time. Historically the reverse splits have been 1:4 or 1:5, with the 1:4 ratio being used during the last 6 years.

History of VIXY Reverse Splits

The first reverse split of VIXY occurred 21 months after its inception, but the second didn’t occur for 37 more months. The high contango of 2016 and 2017 forced a 3rd reverse split after only 12 months, but the up-and-down market from 2018 to 2021 avoided a 4th split until 2021.

The chart below, with both log and linear scales, shows VIXY’s sordid split-adjusted price history.

Given its horrid track record, it’s fair to ask why people keep investing in VIXY. Some are just trying to profit from volatility spikes, hoping to catch the next big crash that somebody is always predicting to happen soon. Others are trying to hedge their equity holdings with VIXY because it is one of the few securities that reliably goes up when the market is panicking.  Unfortunately, this strategy rarely works well.  Unless your timing is very good owning enough VIXY to effectively hedge your portfolio is prohibitively expensive.

Specifics of the Split

If you hold shares of VIXY there isn’t anything to worry about when it reverse splits.  The value of your investment stays the same through the reverse split process.  You just have 4X or 5X fewer shares that are worth 4X or 5X more each.   If your shareholdings are not a multiple of the split ratio, say 215 shares for a 1:4 reverse split, you will get 53 reverse adjusted shares and a cash payout for the 3 remaining pre-split shares.

If you are short VIXY, same story, no material impact.

If you were holding VIXY options (long or short) when the reverse split occurred there’s theoretically no material impact, however, the option chains are going to hurt your head for a while.   This example VXX Options Clearing Corporation reverse split memo describes the adjustments for its March 2023 reverse split.   It adjusts the number of VXX shares per option on the pre-split contract from the usual 100 to 25.  The option strikes are not adjusted and the underlying symbol that the options trade against is VXX1, a new symbol—which is set at 25% of VXX’s price.  These contortions are required so that holders aren’t left with fractional contracts–something the options clearinghouse doesn’t want to deal with.

So, as an example, let’s say you hold ten pre-split call options on VIXY with an expiration date of September 15th and a strike price of $10.  Each contract was worth around 2.81 ($281) at the close on August 21st (VIXY’s closing price was 12.77)  so your overall position value is around $2891.   After the reverse split, your contracts are adjusted so each contract has 25 shares of the new reverse split VIXY as its deliverable.  The strike price, $10 stays the same, and the effective price of the underlying that the option is priced against is the current VIXY value divided by 4.    So, if on August 24th, right after the reverse split VIXY’s price is $51.08 and VIXY2’s price is 12.77.  Your options will continue to be worth about $281 each, and you still own 10 contracts so your position is still worth around $2891.  The only difference is that if you exercise all your contracts you won’t get 1000 shares of VIXY, you’ll get 250 shares.  If you do exercise your shares the price you pay per share will be the reverse split value of the strike price, in the example above you would pay $40 per share (4*10) if you exercised.

New options created after the split will be generated with VIXY as the underlying, but the old adjusted options will hang around until they expire.  Typically the spreads on the adjusted options are not good and sometimes margin calculations are disrupted by reverse splits, so if you are planning on exiting your options, rather than just letting them expire you should consider closing out your positions and re-establishing them after the split. For more on this topic see the end of my UVXY reverse split post.

For regular, forward splits things are more straightforward —the strike price of the options is divided by the split ratio, and the number of contracts is multiplied by the split ratio.  See the OCC memo on SVXY’s 2017 2:1 split for an example.  This basic approach can’t be used on reverse splits (multiply the strike price and divide the number of contacts by the split ratio) because depending on the number of contracts held some customers would end up with fractional contracts—which is a no-go.

The chart below uses my simulated data plus actuals to show VIXY’s price history since 2004

Updated June 7, 2023

For more see:

• Backtest of VIXY back to 2004
• How Does VXX work
• How to Go Long on VIX
• How to Short VXX

This post will discuss how ZIVB works, including how it trades, how its value is established, what it tracks, tax treatment, termination risk, likely liquidity, and performance simulations. 

Volatility Shares’ new -1x Short Mid-Term VIX Futures ETF (ticker: ZIVB) started trading Wednesday, April 19^th. After a gap of over two years, volatility traders have access to a -1X leveraged mid-term volatility ETF. ZIVB’s Prospectus is located on the Volatility Shares website.

Before we get into the specifics of how ZIVB works, I’m disclosing that I hold a small, minority share in Volatility Shares LLC, the issuer of UVIX. Volatility Shares has also issued BITX, UVIX, ETHU, and SVIX ETFs. I do not hold management, executive, or operational roles in Volatility Shares LLC, nor do I give recommendations or investment advice to their clients.  My analysis of these funds and their associated indexes is my own, is not investment advice, and is based on publicly-available information.

Just about anyone who’s looked at a multi-year chart for a long volatility fund like Barclays’ VXZ or ProShares’ VIXM has thought about taking the short side of that trade. VelocityShares’ ZIVB is an Exchange Traded Fund (ETF) that allows you to hold a short volatility position while avoiding some of the issues associated with a direct short position in VXZ or VIXM.  Because ZIVB is tied to VIX futures with at least 4 months until expiration its daily percentage moves are considerably smaller than the moves of funds (e.g., VXX, UVXY, UVIX) that are tied to shorter-term, more volatile VIX futures.

What’s Different About ZIVB?

The biggest difference between ZIVB and ZIV, the previous -1X leveraged short term volatility fund is that ZIVB is structured as an ETF, and is compliant with the Investment Company Act of 1940 regulations (’40 Act). As a result, ZIVB’s tax treatment likely will be the same as regular stocks, with Form 1099 reporting of gains and losses, rather than the K-1 reporting that ZIV used. I am not a tax advisor, but my understanding is that ZIVB held in non-taxable accounts such as IRAs will not require tax reporting.

Unlike ZIV, ZIVB will probably eventually have options available, something that never happened with ZIV because of its structure as a leveraged Exchange Traded Note (ETN). Volatility Shares’ ability to structure ZIVB as a ’40 act fund will make ZIVB an investment option for investors that are limited to only holding securities that are compliant with the ’40 Act.

While not likely to impact ZIVB significantly, Volatility Shares, when doing their end-of-day rebalancing, limits their trade volume to no more than 10% of the market volume in any 15-minute period of continuous market trading, including after-hours trading.  Rebalancing is an operational process (described later in this post), used by leveraged funds to achieve their target leverage.  Since ZIVB’s likely percentage moves are significantly smaller than the short term VIX volatility funds the mid-term VIX futures volume associated with its rebalancing operations will likely be small. The net effect of this rebalancing restriction is to structurally prioritize the smooth functioning of the market over the precise tracking of ZIVB/SVIX/UVIX share prices to their underlying indexes.

How is ZIVB’s Value Established?

• Ultimately ZIVB value is tied to the daily resetting inverse of an index (S&P VIX MedTerm Futures Inverse Daily Index) that specifies a hypothetical portfolio of VIX futures with 4 through 7 months until expiration.  Every day the index specifies a new mix of VIX futures in that portfolio. On any given day one-third of ZIVB’s assets are allocated to VIX futures with 5 months till expiration, another third is allocated to 6^th-month futures, and the final third is split between 4^th and 7^th-month futures. This mix of VIX futures gives ZIVB the approximate performance of a VIX future with 153 days until expiration.
• The excess return index ZIVB tracks, SPVXMPI, is maintained by the S&P Dow Jones Indices.  The net asset value of ZIVB is published every 15 seconds during market hours as the “intraday indicative” (IV) value.  Yahoo Finance publishes this quote using the ^ZIVB-IV ticker. Because ZIVB’s day-end value is set by the settlement prices of VIX futures the closing IV value of ZIV is established at the NYSE 4:00 PM ET close. To get historical data on this index, go to this overview page, select the Excess Returns option, then set the chart range to maximum and then click export to get the index back to 28-March-2013.
• One potential wrinkle in ZIVB’s precise tracking of the SPVXMPI index is a side effect of how ZIVB is structured. Since ’40 Act funds can’t directly hold futures or derivatives, all of ZIVB’s VIX futures buy/sells will be done in a wholly-owned Caymen subsidiary, specifically, -1x Short VIX Mid-Term Futures ETF Cayman Ltd. The ’40 Act rules require that holdings in subsidiaries such as these be less than 25% of ZIVB’s overall investment at the end of its fiscal quarters, so ZIVB may need to lower its investment level in its Caymen subsidiary before the end of each quarter, which might affect tracking to its index.
• Volatility Shares’ short term funds, UVIX and SVIX use the SHORTVOL and LONGVOL indexes that incorporate measures to smooth out the end-of-day rolling/rebalancing volumes in the front two VIX futures contracts. Volatility Shares did not create a SHORTVOL-style mid-term index for ZIVB, probably because the end-of-day market stresses on the mid-term VIX futures are much lower than the short term futures, and do not justify the expense of creating a new index. Operationally, Volatility Shares will include ZIVB’s rebalancing needs into their overall commitment to constitute no more than 10% of the market volume in any 15-minute period of continuous market trading, including after-hours trading. 

How does ZIVB trade?

• ZIVB trades like a stock. It can be bought, sold, or sold short any time the market is open, as well as pre-market and after-market periods. 
• ZIVB’s shares can be split or reverse split—but unlike long volatility funds such as VXX (with 7 splits since inception), splits of ZIVB will be an infrequent event.
• ZIVB is tradable in most IRAs/Roth IRAs, although your broker will likely require you to electronically sign a waiver that documents various risks.  Shorting of any security is not allowed in an IRA.
  

How is ZIVB’s price established?

• The bid and ask prices of ZIVB should typically be close to its Indicative Value (IV) price. The IV price is the per-share value of the fund’s underlying assets minus fees and operational costs.  During periods of heavy selling/demand, the bid/ask prices might start moving away from the IV price. If ZIVB’s prices start diverging from its IV value during regular market hours then market makers and wholesalers called “Authorized Participants” (APs) will likely intervene by buying or selling shares of ZIVB. If ZIVB is trading enough below its IV price they will start buying large blocks of ZIVB—which tends to drive the price up.  If it’s trading high, they will short ZIVB. The APs have an agreement with VolatiltyShares that allows them to do these restorative maneuvers at a profit, so they are highly motivated to keep ZIVB’s tracking in good shape. For more information on these behind-the-scenes arbitrage processes see Why Arbitrage is Essential For Exchange Traded Products.
• ZIVB will likely register gains during rising market periods but will experience drawdowns when markets are in turmoil.  The chart below shows a simulation of ZIVB from March 25, 2004, using index values. The simulation assumes ZIVB has an annual fee of 1.35% and an estimated 0.63% per year in additional fixed and operational costs. A free copy of this simulation including ZIVB’s IV values, and index similar to SPVXMPI is available here, ZIVB backtest.

• ZIVB’s potential for big drawdowns is shown in this simulation with a 30% drawdown in the February 2018 Volmageddon and 62% in March 2020 during the Covid crash.
• ZIVB seeks investment results before fees and other expenses that correspond to the daily moves of the SPVXMPI index. This index is maintained by the S&P and its values are published during market hours.  SPVXMPIT values can be found on the S&P Dow Jones Index website and Yahoo Finance as ^SPVXMPI.  The Net Asset Value/Indicative Value (NAV/IV) values for ZIVB will be published by Yahoo Finance as ^ZIVB-IV. I offer a free spreadsheet with simulated close IV values for ZIVB since March 2004.
  

What is ZIVB’s Market Strategy? 

• The term structure of VIX futures is contango about 80% of the time, with prices increasing with additional time to expiration.  Unless there is significant turmoil in the market the futures usually decay significantly in value over time when they are in this configuration.  Since ZIVB is short VIX futures, it’s positioned to benefit from that decay.
• This situation sounds like a short seller’s dream, but VIX futures occasionally go on an upward tear, punishing those that are short volatility.
• ZIVB does not behave like a static short position. It attempts to track the daily percentage moves of the SPVXMPIT index.  To accomplish this strategy ZIVB must rebalance its investments near the end of each day.   For a detailed example of what this rebalancing looks like see “How do Leveraged and Inverse ETFs Work?”
• There are some very good reasons for this rebalancing strategy, for example, a static short position can only produce at most a 100% gain and the leverage of a static short is rarely -1X (for more on this see “Ten Questions About Short Selling”).  ZIVB, on the other hand, should be able to deliver a move very close to the percentage daily move of SPVXMPIT on an ongoing basis.
• Detractors of the daily rebalancing approach correctly note that ZIVB and funds like it often suffer from volatility drag.  If a long volatility fund like VIXM moves around and then ends up in the same place ZIVB will lose value, whereas a static short of VIXM would not.  However, as I discussed in “Is Shorting UVIX, UVXY or VXX the Perfect Trade?”, static short positions have other problems. 
• While volatility drag often reduces performance, daily resetting funds like ZIVB don’t always underperform.  If VIX futures are trending down, they can deliver better than -1X cumulative performance.  For more on this phenomenon see “Sometimes Leveraged Funds Outperform.”
• In backtest simulations, ZIVB has median moves of -0.21X compared to the CBOE’s VIX index.  If the VIX moves up 10% you can expect ZIVB on average to move down 2.1%.  However, this relationship is not cast in stone.  At times ZIVB and VIX will even move in the same direction.
  

Can ZIVB Terminate?

ZIVB doesn’t have a percentage decline trigger as ZIV did, but it can terminate, as can any Exchange-Traded Product.  In its prospectus (on approximately page 71) it says:

The Trust or any series or class thereof may be terminated at any time by the Board of Trustees upon written notice to the shareholders

How Good is ZIVB’s Liquidity?

• Initially, ZIVB’s bid/ask spreads will likely be a few cents, with relatively low average daily volumes. As I discuss in Evaluating the Liquidity of Low Volume Exchange Traded Funds, the key liquidity source for ETPs is the liquidity of the underlying security, which in ZIVB’s case are the liquid fourth through seventh-month VIX futures. 
  

What are the advantages of ZIVB relative to inverse short term volatility funds like SVIX or SVXY?

• The 4th through 7th-month VIX futures that ZIVB uses are in contango more often than the short term funds. Contango tends to drive up the price of ZIVB, because when in contango the futures held tend to drop in value every day.
• Volatility drag, ever present in leveraged ETPs, will be significantly lower in ZIVB compared to the short term funds. Volatility drag in leveraged funds is proportional to the volatility squared of the underlying index. In this case, ZIVB’s volatility will be about 2% daily, instead of the 4% daily of SHORTVOL, therefore its average volatility drag will be 1/4th of SVIX.
• More investigation is needed, but it appears that the SPVXMPI index may often lag the short term indexes by a day in strongly reacting to equity downturns. If true that might offer precious time to exit a ZIVB position before significant damage is done.
  

Conclusions

• ZIVB is a product intended for sophisticated investors and should not be viewed as a “buy and hold” investment.
• ZIVB will experience drawdowns when volatility spikes.
• In my opinion, ZIVB is best used when the VIX Futures’ term structure is in contango and its investors have a plan to exit their position when the market is nervous (e.g., a VIX/VIX3M ratio greater than 0.95) or protective options strategies are in place to handle a volatility spike.

For more information:

• The Cost of Contango
• How does SVIX Work?
• How does UVIX work?
• How Does VXX Work?
• Free Simulation of ZIVB back to 2004
• ZIVB’s Index S&P GlobalX SPVXMPI Select Excess Returns option
  
  

Podcasts

• Keeping it Simple on the Road on (Simplify Asset Management)
• Trading Volatility in Volatile Markets on (nextSignals)
• Volatility Series: Finding True Value in the World of Volatility featuring Benn Eifert on (Top Traders Unplugged)
• How An Exotic Investment Product Sold In Korea Could Create Havoc In The U.S. Options Market on (Odd Lots) 2020
• Here’s What’s Happening With Those Korean Structured Notes That Bet Against Market Volatility on (Odd Lots) 2020
• Options, volatility, and the things we don’t know we don’t know on Flirting With Models, Corey Hoffstein July 2020
• Derivatives markets history, impact of Dodd Frank legislation, why some derivatives buyers are price insensitive and the opportunities that creates, three trading strategies using the VIX, dispersion trading, and European dividend futures. (Mutiny Funds, Taylor Pearson)