Why ZSL Did Not Malfunction Last Week

| About: ProShares UltraShort (ZSL)

A lot of ZSL holders were unpleasantly surprised and outraged by the ETF's seeming failure to achieve the double inverse of the daily return on silver on May 12th and 13th. On these dates the daily returns on SLV were -3.11% and 3.21%, respectively, but the corresponding returns on ZSL were 0.46% and -9.43%, and not the expected 6.22% and -6.42%. ZSL discussion forums were full of angry investors, who were seeing their positions in the red on the day when silver was also down. Many of them were even calling for SEC investigation of the fund.

In order to make sense of what happened, one has to take into account that ZSL is rebalanced at 7am (ET) when its NAV is calculated based on the London Silver Fix. When it begins trading at 9:30 am on the NYSE, the proper baseline is not the previous day's closing price, but the 7 am NAV. This means that the fund has taken short positions on silver (forward and futures contracts) based on the London Silver Fix price, and it will be losing if the silver price goes above the London Fix after 9:30 and gaining if it is below. Since ZSL is double-leveraged, if the price at the time of rebalancing (and releveraging) was sufficiently lower than it is at the open of trading in NY, it is possible for ZSL to have lower NAV and price per share despite the price of silver being down relative to the previous day's close.

This is exactly what happened on Thursday, May 12th. The day before, ZSL closed at $21.70 and SLV closed at $34.39, however, their NAVs were calculated based on that day's London Silver Fix of $39.18 and were $18.11 and $38.22, respectively. Overnight, silver prices fell during European trading hours, and the London Silver Fix on May 12th was $32.50, which represents a 17.05% drop. Thus, after rebalancing at 7 am on May 12th, the NAV of ZSL was $21.70*(1+2*0.1705) = $24.29, and the NAV of SLV was $38.22*(1-2*0.1705) = $31.70. Note that ZSL's NAV is substantially higher than its closing price of $21.70 (see Table 1).

Table 1

Lond. Fix

%∆ Lond. Fix

ZSL NAV

%∆ ZSL

NAV

SLV

NAV

%∆ SLV NAV

5/11

39.18

18.11

38.22

5/12

32.5

-17.05%

24.29

34.1%

31.70

-17.06%

5/13

36.2

11.38%

18.76

-22.8%

35.31

11.39%

Click to enlarge

What happened next was that the silver price went up between 7 am and 9:30 am (ET), and although at the open it was still below the prior close, it was substantially higher than its 7 am low—SLV opened at $32.96, 3.97% higher than its NAV. Because ZSL had been rebalanced at the lower London Fix, it now had to go down by 2x3.97%, which it did pretty closely by opening at $22.28, which was 7.98% below its 7 am NAV, but still above the prior close.

To judge whether or not ZSL matched its objective, we have to look at its performance relative to its 7 am NAV and not the prior closing price. As you can see in Table 2, returns calculated from the time of rebalancing and to the close are correct. On May 11th, SLV closed up 5.11% from its NAV (although down 3.1% from the previous close), and ZSL closed down 10.24% from its NAV (up 0.46% from its prior close). Similarly, returns on May 13th are also pretty close when calculated relative to the 7 am NAV.

Table 2

SLV NAV

SLV close

R (daily)

R (from NAV)

ZSL NAV

ZSL close

R (daily)

R (from NAV)

5/11

38.22

34.39

18.11

21.7

5/12

31.70

33.32

-3.1%

5.11%

24.29

21.8

0.46%

-10.24%

5/13

35.31

34.39

3.2%

-2.61%

18.76

19.7

-9.63%

5.01%

Click to enlarge

So the bottom line is that ZSL did not malfunction and there is no fraud involved. But this incident does illustrate the fact that ZSL is not meant to be held overnight. And the reasons in this case go beyond the normal releveraging at the end (or start) of the day. Large price swings during European trading hours when you can't easily trade ZSL can have a large effect on its price when it begins trading on NYSE. For example, if on May 12th the London Silver Fix had for some reason crashed to $25/oz (36.19% drop), ZSL would have been rebalanced at this price with the NAV of $18.11x(1+2x0.3619)=$31.23, but when SLV in NY trading had gone back up to $33.32 (37% increase relative to its 7am NAV), ZSL would have gone down to $8.34 (=$31.23x(1-2x0.37)), a 61.6% loss relative to its May 11th close.

This also brings up another important issue - discounts and premiums for these silver ETFs, based on the London Silver Fix. I see a lot of comments and blogs about the seemingly large premiums/discounts of SLV/ZSL/AGQ. You have to realize that the NAV used in these calculations is very stale (based on the 7 am price) and the market price is more current. So a premium or a discount doesn't indicate whether these funds are overbought or oversold, but merely shows whether silver went up or down since 7 am (ET).

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.