Sometimes, after highlighting a futures market trade opportunity, I get an email from a reader asking if they can put that trade on using levered ETFs.  My answer is always the same - levered ETFs are fine for short-term trading, but you never want to hold them for a long period of time!

Levered ETFs often get a bad rap.  They shouldn't.  They do what they were designed to do remarkably well.  The real confusion lies in investors' understanding of how the leverage is achieved.

Let's go through an example of a typical margin or futures account trade to understand how "traditional" leverage is accomplished and then compare it to the methodology used in a levered ETF.

For a "traditional" leveraged trade, an investor buys a security at a price of $100 and puts up $35 of margin with the broker lending the balance of $65.  We will use a 35% margin rate for this example, but that can differ depending on the instrument.  The investor must maintain an "equity" value of 35% the value of the security at all times.  If the security rises in value, the investor can actually withdraw the extra amount.  However, if the security's price falls, the investor must meet the margin call or the position will be liquidated.  Pretty simple.

On the other hand, let's examine a levered ETF.  For this example we will use a three times ETF in which the underlying assets of ETF trust are $100 million.  The ETF administrator must then borrow $200 million in some fashion to allow the investor to achieve a 3 times return.    So far, it's similar to the "traditional" margin buying in the previous example.  But what happens if the price of the asset falls?  If all of a sudden the $300 million portfolio loses $60 million, there would only be $40 million of equity versus $240 million of assets.  The ETF cannot issue a "margin call" on its holders, so have to find some way to protect themselves.

But how do they do that?  A slight change in language of the type of leverage offered by the ETF makes all the difference in the world.  Look closely at what the ETF says when it comes to the payoff profile of the leveraged ETF.   Here are the specifics of the NUGT and DUST 3X bull and bear Daily Gold Miners leveraged ETFs:

Look closely at the first line of the second paragraph:

"These leveraged ETFs seek a return that is 300% or -300% of the return of their benchmark for a single day."

In our previously discussed traditional levered case, if an investor meets margin calls, then a 3 times return (minus borrow costs) is assured regardless of the time period.

However, the levered ETF is not offering that payoff profile.  The levered ETF's return will be a series of 3 times returns for each day.  I know that doesn't sound that different, but it's night and day.

To best illustrate this difference, let's engage in a mental exercise of pretending we are the ETF administrator in charge of offering this 3 times daily return.  Bear with me as we do a little math.  At the end of this process, I think it will be more clear what is happening with the levered ETF.

Let's use our original ETF example of $100 million of equity controlling $300 million of assets.  Sometimes the best way to understand something is to use dramatic price changes and walk through what happens.

If the underlying asset price moves up 20% in a single day, the total value of the assets is now $360 million.  Assuming the ETF starts at a price of $100 per share, it's now worth $160 ($360 MM - $200 MM borrowed = $160 MM).

However, what happens if it moves up another 20% the next day?

If the ETF administrator makes no changes, then the $360 million of assets rises to $432 million.  That means the ETF is worth $232 ($432 MM - $200 MM borrowed = $232 MM).  

But let's remember back to what the ETF was promising.  It's supposed to be 20% of the daily return each day.

In our example, if you bought it at the end of the first day ($160 close) and sold it at the end of the second ($232 close), would you achieve this 20% x 3 daily return?

[ $232 - $160 ] / $160 = 45%

Whoa!

That makes no sense, right?  Well, I have made the moves unrealistic so that you can see what's going on, but this happens on every single levered ETF.

The ETF provider needs to ensure a levered return on the asset for tomorrow's move, not for the original investment.  So at the end of each day, they need to readjust.  

In our example, the ETF was priced at $160 with $360 million in assets.  But to get a three times levered return the next day, the ETF needs to have $160 * 3 = $480 MM in assets.  

Therefore the ETF needs to BUY $120 MM of assets on the close of the first day (which it borrows, making total borrowing $200MM + $120MM = $320 MM).

Now, let's go through the numbers and see what happens from there.

The ETF has $480MM of assets which rise another 20% the next day.

After that rally, the numbers are as follows: $480 MM * 1.2 = $576 MM.  The price of the ETF would be $576MM - $320MM = $256 MM = 256 per share.

Let's review; 160 with a 20% rise x 3 = 160 * 1.6 = 256 per share.

So what does this mean?  The ETF administrator needs to BUY MORE EXPOSURE EVERY NIGHT AT THE CLOSE WHEN THE ASSET IS UP IN PRICE, and SELL MORE EXPOSURE EVERY NIGHT AT THE CLOSE WHEN THE ASSET IS DOWN!

Now let's think about that.  The ETF buys as it goes higher, and sells when it goes lower.  What happens in our example if the asset price falls the next day by 30.55%?

Well, let's just confirm what the underlying asset return would be:

Day 1: 100Day 2: after 20% rally = 120Day 3: after another 20% rally = 144Day 4: after 30.55% sell off = 100

As far as the ETF, last we checked, it had $576MM of assets with $320MM in borrowing.

After a decline of 30.55%

$576 MM - $176MM = $400 MM with $320 MM borrowing = $80 MM = 80 per share.

Wait, we started with $100 and the underlying asset is back to its original price, but we have less than we started.  

That's because we bought higher and sold lower.  

Leveraged ETFs do exactly what they were designed to do - provide a constant levered daily return of an asset.  However, that payoff return is dramatically different than the traditional interpretation of a leveraged return over an extended period of time.

I love to day trade levered ETFs.  And in rare circumstances, I will hold them for a few days.  But I do so understanding my changing exposure based on the math above.

However, there is nothing I like better than using this decay to my advantage, by shorting levered ETFs as a way to express a view.  But beware - in a trending market, this is a worse position you could have.

Thanks for reading,  
Kevin Muir  
the MacroTourist