Tuesday, September 21, 2010

An ETF Lesson - Part I

I really don't want to write a post debunking the hysteria thrust upon the quivering masses (who reply: "OH NO! HOW IS THIS LEGAL?!!")  by Andrew Bogan in this FT Alphaville piece (and now re-hyped by Clusterstock).  Since Steve Waldman addressed a number of Bogan's errors in the comments section of Bogan's own post (if you're interested in this topic, you must read swaldman's comments before continuing!), I figured now would be a good time to explain a little bit about how ETFs work to the masses, who remain terminally confused.

One of the great things about ETFs is that they can be created and redeemed.   This means that "authorized participants" (read: big broker dealers) can take a basket containing the underlying stocks of the ETF, in specific weights, deliver them to the ETF trust and receive the ETF shares - that's called creating.  They can also do the opposite:  deliver the ETF itself to the trust, and receive the underlying basket of individual stocks - that's called redeeming.   

A great source of confusion that many folks have regarding ETFs is the concept that when you buy an ETF, like, say SPY - the S&P 500 ETF - the trust itself does not take your money and go out and buy the underlying S&P 500 stocks with it.  Similarly, when you sell SPY, the trust does not go out and sell stocks to raise money to give you.   The trust has nothing to do with it.   When you buy SPY (just like when you buy IBM) , you buy it from another participant in the market  - not from the trust (not from the company itself).  Contrast this to mutual funds: when you buy an S&P 500 mutual fund, that mutual fund does take your money and go out and buy their portfolio of stocks with it, and vice versa when you sell.

So, back to our SPY ETF - the reason this works is precisely because of the creation/redemption mechanism.  If "the market" is lacking SPY sellers, and the price of SPY rises so that it is in excess of it's NAV (net asset value)  there are arbitrageurs standing by ready to short you shares of SPY while they simultaneously buy the underlying basket of SPY components as a hedge.  Since they are selling SPY "rich" to it's fair value, they will make a profit when they eventually collapse their position.  How do they collapse it?  Well, they take their long basket of SPY component stocks and "create" SPY by delivering the individual stocks to the trust, receive newly created SPY, and use that to cover their SPY short position.  Voila - they're now flat, and the SPY's assets have increased, as have the shares outstanding. 

Of course, if everyone wants to sell SPY, the opposite happens - the arbitrageurs, if SPY is trading "cheap" to it's NAV, will buy SPY while simultaneously shorting a basket of the underlying components.  Then, they'll take their SPY shares, deliver them to the trust, redeeming them for the underlying components which they use to close out their short underlying stock positions.  In this case, the assets held by the trust decrease, as do the shares outstanding (after all, the arbs have taken SPY shares out of circulation, and given them back to the trust, effectively "retiring" the shares temporarily.)

The next level of this discussion gets somewhat complicated pretty quickly.  Bogan is worried about "naked short sellers" but he demonstrates a thorough lack of understanding of what a naked short seller is.  Bogan writes:
"Take the SPDR S&P Retail ETF (NYSE: XRT) as an example. The number of shares short was nearly 95 million at the end of June, while the shares outstanding of the ETF were just 17 million. The ETF was over 500% net short! Or to look at it from another perspective, the ETF’s operator, State Street Global Advisors, believed that there were 17 million shares of the SPDR S&P Retail ETF in existence and owned shares in the S&P Retail Index portfolio to underlie those 17 million ETF shares. But, in the marketplace there were another 95 million shares of the ETF owned by investors who had purchased them (unknowingly) from short sellers. 78 million of those ETF shares were naked short–the short seller had promised their prime broker to create those non-existent shares if necessary to cover their short in the future. In both cases the share buyer, however, is completely unaware his ETF shares were purchased from a short-seller and no doubt assumes the underlying assets in the index are being held by the ETF operator on his behalf, but no such underlying stock is actually held by anyone. Clearly this creates a serious counterparty risk and quite possibly the potential for a run on an ETF—where the assets held by the fund operator could become insufficient to meet redemptions."

First, it's perfectly possible to have 95mm shares short with only 17mm shares of XRT outstanding, although this is a concept that many people hate and are confused about.  How?  It's just like fractional reserve banking.  First, remember that in order to short shares of stock, you first need to borrow those shares from somebody (so that you can deliver them to the buyer on the other side of your short sale).  If you don't borrow the shares, that's called naked shorting, which is different from what Bogan describes.  These ownership chains get hard to follow pretty quickly, but let's pretend that I, Kid Dynamite, bought shares of XRT in the initial public offering.  I own the shares.  I take my shares and lend them to Steve, who goes and shorts them.  Now, I no longer have shares to lend - I've already lent them out   Steve has borrowed them and shorted them, and the buyer of that stock is Dave.  Dave then takes the stock which Steve delivers to him, and lends it out again, to Mike, etc...  Thus, multiple people end up short the "same" share of stock, everyone has delivered the stock they promised to, no one failed to deliver, and no one is naked short.  (if you haven't read Swaldman's comment explanations, now would be a good time to do so.  first, second).   Like fractional reserve banking, there's collateral in each step of this chain - if I want my shares back and Steve can't get them for me, I'll take his collateral instead.  SWaldman describes it thusly:

"Lending of securities creates synthetic supply of shares, just as lending by [banks] creates a synthetic supply of money. With shares and with money, the lending arrangements are sufficiently well guaranteed that most holders consider the synthetic to be a perfect substitute for the original. Most people don't fret over the distinction between bank deposits and cash-in-hand, just as most stock traders don't fret over the distinction between owning a share and owning a promise to deliver a share on demand by their broker."

Now, Bogan fears a "run" on the ETF.  That's not going to happen (at least not for the reasons he fears), for a few reasons.  Let's start with the more complicated one, using Bogan's own numbers:  as Bogan notes, now 95mm people think they "own" the XRT.  Well, we know that only 17mm of them actually have possession of the XRT, since the other 78mm have lent their shares out to others.  You can't redeem something you don't have (remember, you've lent it out), so it's not possible for 95mm shares to be redeemed.  If some of the people who have lent their XRT out (maybe they even lent it out unknowingly) decide they want their shares back, it's not the end of the world!  The short sellers of XRT will have a few choices to return the XRT that they "owe."  They might go out in the marketplace and simply cover their short positions, or they might buy the underlying basket of XRT stocks, deliver them in to the trust, and "create" new XRT shares to deliver to the person they borrowed them from.   As Swaldman puts it, "The missing supply of ETFs would create itself."

Now that we've explained the concept of possession of stock, let's go to Bogan's worry of how the implosion might happen:  He writes:

"Redemptions occur when more owners wish to sell out of their holding in the ETF than there are new buyers for the existing shares, so unwanted blocks of 50,000 ETF shares each are redeemed through the authorized participants with the ETF operator for cash, or more typically for in-kind shares in the ETF’s underlying index’s stocks."

I hope that Bogan is oversimplifying (and I think he is), because I explained above that owners of XRT wishing to sell their shares do not directly cause redemptions.  I have a feeling that Bogan knows this, and that he's basically explaining what will happen after the arb situation I described above takes place:  when there are insufficient buyers of XRT (ie, too many natural sellers), arbs will step in as buyers, while shorting the underlying basket of stocks, and then redeem their XRT to the trust.  Now, as long as the sellers deliver their shares, the arbs (the XRT buyers) can take the XRT and redeem them with the trust.  If sellers don't deliver the shares they owe, we are back to the scenario a few paragraphs above - they will have to make good by buying back the shares in the market, or by creating more XRT by buying the underlying basket and delivering it to the trust.

Bogan worries about the rapid contraction in XRT shares outstanding in July/August:

"The SDPR S&P Retail ETF was one of the fastest contracting ETFs in July due to redemptions and as of July 31, it had just 7 million shares outstanding. However, the short interest was little changed—still over 80 million shares short. Suddenly, 11 times the number of shares outstanding was short, which is even more worrisome than 5 times back in June. By late August, the shares outstanding in XRT had dipped briefly below 5 million shares with 80 million shares still short (16 times the shares outstanding). Mercifully, net buying interest has rebounded somewhat for the SDPR S&P Retail ETF with the improving outlook for retailers and shares outstanding in XRT had rebounded to 12 million by mid-September. But if the rate of contraction last month had continued, the ETF was just days away from running out of underlying shares altogether."

No - the ETF was probably not days away from running out of shares. This phenomenon is self-healing.  This is the important part, but it's not as complicated as it sounds.  As we've established,  shares can only be redeemed when they are actually delivered into the trust.  If everyone wants to redeem their shares, then the shorts need to cover their positions - we have also already discussed this above.  The shorts can either borrow the XRT from someone else (impossible if everyone wants to redeem), cover in the marketplace (again, if everyone wanted to redeem, that would also be impossible, but we can pretend that I overpay for XRT to the extent that one of the redeemers is happy to not redeem and instead sell it to me for a premium to NAV), or buy the underlying basket and deliver it to the trust and create more shares (and more underlying assets for the trust to distribute). 

Thus, the most likely result if everyone wants to redeem XRT when there is 500% short interest is actually a massive short squeeze, or a rally in the instrument, as longs scramble to re-take delivery of shares that they have lent out.  But what happens if the actual physical holders of XRT do all want to redeem, and the "self-healing" I described doesn't happen in time?  Is the trust then "days away from running out of assets altogether"?  Read on.

Bogan's final few paragraphs are some confused hysteria asking who is left holding the bag if all the holders of the actual outstanding shares manage to redeem their shares. 

"Who gets left holding the bag? Is it the retail account holders who own defunct shares in a closed ETF? The prime brokers that were counterparties to all those short sellers? The hedge funds that sold non-existent shares in an ETF assuming they could always be created another day? The ETF operator? Or the Federal Reserve?"

First of all, it is indeed entirely possible to imagine a scenario where all of the actual holders want to redeem their shares, which results in the trust "running out" of assets.  In our example, 17mm people could redeem shares and receive the assets of the trust, and that would leave 78mm longs and 78mm shorts outstanding.   As commenter "Crookery" notes, the results are not apocalyptic.  The trusts have a mechanism built in to liquidate when their assets get below a certain point, and the outstanding shorts simply owe the outstanding "synthetic" longs the cash value of the trust.  There is no "bag" to be left holding - short sellers owe long holders.  

Furthermore, I think that if there is a greater supply of shorts outstanding, it actually makes it less likely that this happens, and more likely that the "self healing" scenario occurs, since there are more shares to get "bought in" and cause a short squeeze.  Said differently, the trust runs out of assets when the actual physical XRT holders all redeem their shares - not when the "synthetic" XRT holders want to redeem their shares.  "Synthetic" holders wanting to redeem causes the self healing short squeeze phenomenon, and there are more "synthetic" holders if there is more short interest.

Now, let's answer Bogan's headline question: "Can an ETF collapse?"  Sure - of course it can, but not for the reasons Bogan fears, and not with the subsequent effects he ponders either.

At this point, I feel like we've gotten into some confusing topics, so I'll close with another Swaldman comment:
"Shareholders end up with synthetic long positions by their own consent, when they agree to lend shares from their account to potential short-sellers. It is true that more people consent to this than understand they are consenting to it: retail investors with margin (as opposed to cash) accounts may fail to read the boilerplate that gives their broker the right to lend their shares. But those retail investors' credit exposure is to their broker, not to the speculative short-seller that eventually borrows and resells her stock. In general, then, retail investors in margin accounts may have credit exposure to their broker by virtue of short-selling that they do not understand. And if brokers are incautious in their securities borrowing and lending, they could blow-up and force losses on clients holding margin accounts. In theory, that is a real concern, but it has nothing to do with ETFs. We should be concerned that brokers who lend the shares of customers are regulated to prevent credit losses that impact their solvency, whether those losses result from share-lending or any other practice. This is the LTCM issue, not something specific to short-selling or ETFs."



Taylor said...

KD - I realize this complicates things, (and is slightly veering off topic) but if Mom and Pop owned 175 shares of XXX ETF in their account and weren't selling, then the idea that XXX could collapse would never be tested, correct? In these examples, they were only taking into account that every outstanding share was for sell, right?

Also, any time some idiot congressmen/senator starts railing about the evils of shorting, Steve Waldman needs to head on up and explain to them what it really is, complete with money lending examples.

Kid Dynamite said...

Taylor - as long as Mom and Pop owned 175 XXX and didn't lend it out, then those 175 shares could never be redeemed, and the trust would always have at least enough assets to cover their 175 shares. If Mom and Pop's shares were lent out, then the person who they were eventually delivered to could redeem them, in theory.

Nemo said...

First of all, it is indeed entirely possible to imagine a scenario where all of the actual holders want to redeem their shares, which results in the trust "running out" of assets.

I believe this is the crux of the issue and everything else is noise. So let me see if I can rephrase the Bogans' concern as I understand it.

When I buy a share XRT, I assume I am buying something that will track the retail index reliably. I make that assumption because, even though I personally cannot afford 50,000 shares, there are many entities who can... And they can convert any large block of XRT to and from baskets of underlying stock. So it is easy for me to believe that the market will ensure my 1 share of XRT tracks the basket.

But what happens if a "run" on redemptions forces the ETF to shut down? Overnight, my XRT share -- which has, unknown to me, been lent out -- suddenly becomes an obligation for some other party --- whom I do not know -- to pay me cash.

Most people would see a difference, I think, between an actual basket of stocks and an instrument that could, during times of distress, instantly morph into a cash obligation from an unknown third party. Especially if one can easily imagine the transformation happening simultaneously on the scale of tens of billions of dollars. Looked at this way, I must say the instrument has a very different "feel" to it.

Of course, something similar could happen with an ordinary stock. Any company could say, "We are liquidating tomorrow and distributing the proceeds to our shareholders"... And that would convert all lent shares into cash obligations. But (a) almost no companies have a short/outstanding ratio of 5-10x; and (b) the entire retail sector is highly unlikely to decide to liquidate simultaneously.

So yeah, I think the Bogans may be identifying a sort of systemic risk that most people would find surprising. I certainly had not considered it until now.

wcw said...

Well, not necessarily. Sometimes Ma and Pa Kettle are net long, but the APs (generally around index rebalancing time) do not want to own shares issued by the trust. APs may and do redeem without needing to borrow shares, since they are allowed to be short for a week or so. Occasionally, this process goes to extremes large enough to worry ETF sponsors. Ask your friends who work for Ishares. They may be able to tell a story or two. Or just look at a longer-term chart of IWM.SO and look for its couple forays down near zero.

This is deeply inside baseball, of course, and in no way threatens the installed base of assets or the attendant ETF. But it can get pretty weird if you're running the thing. Actually hitting zero assets and shares is something you would prefer to avoid.

Kid Dynamite said...

Nemo - yes, absolutely - I like your reasonable restatement. but this has nothing to do with short interest (as Bogan claimed), or "naked shorts" (as Bogan claimed) and I even hypothesized that large short interest may result in a lower chance that this happens (this may not be correct, it's kinda theoretical).

as for what your lent out XRT represents: again, as Waldman noted, your counterparty risk isn't to some unknown spurious short seller, it's to your broker (this is another core point, contrary to Bogan's piece). And the amount you receive isn't some random cash quantity, it's directly related to the value of the basket (recently, at least). So basically, since you already noted that you aren't buying XRT to receive the basket of stocks (since you don't have 50k shares), you're buying it to receive the return on a basket of stocks. that is still likely to track pretty well, even in times of implosion - you'll get (realized) cash performance instead of (paper) stock performance though.

note: I owned SKF before and during the blowup in 2008. There came a time when the ETF basically said that they couldn't say that their performance would be in line with their stated goals, since they were unable to execute short sales in bank stocks! (this is kinda a different topic, but anyway, etf's certainly can and do break down, just not for Bogan's reasons, and not with the apocalyptic effects he fears (WILL THE FEDERAL RESERVE HAVE TO PICK UP THE TAB?!?!?!)

Kid Dynamite said...

WCW- one of my guys told me, yesterday, "it is also true the fund provider won't release the ETF shares until they receive the pieces" - ie, you can't create until you deliver the pieces. I would assume that redemptions are the same way - that you can't redeem unless you deliver the ETF. This seems to conflict with your comment.

Nemo said...

I think it does have to do with short interest, in that the magnitude of the "overnight conversion" (globally) depends precisely on that short interest...

One more thought. If someone delivers a basket of underlying stocks to State Street in exchange for 50,000 XRT shares, will State Street loan those underlying shares out to other people?

My XRT shares represent a claim on a basket of stocks. In essence, they are that basket of stocks. Now, suppose State Street lends those underlying stocks to short sellers. And I also lend my XRT shares to short sellers. For all practical purposes, haven't those underlying shares now been lent to two different people? If I personally lent the same shares to two different people, would that not fit the definition of "naked short selling"? How is this different?

Kid Dynamite said...

Nemo - I'm not sure what you mean in your first sentence, but let's talk about the rest:

It seems highly likely that State Street will indeed lend out the underlying XRT basket of stocks. (I'm not aware of anything legally preventing them from doing so - let's just pretend that they do!)

And you may lend out your XRT also. As Waldman noted, the whole process is collateralized - now, what is your concern? People who raise hell about naked shorting usually make some reference to "counterfeit" shares. Waldman did a good job pointing out that all short sales result in "synthetic" (better term than counterfeit) shares - I would expand that by saying that the difference is that naked shorts result in synthetic shares without consent.

But when you, or the person who eventually receives your XRT, wants to redeem it, State Street will recall the underlying shares and deliver them.

It seems likely that State Street, wanting to me sure they don't have to sweat this, would take measures to be sure that they won't have to worry about meeting redemptions - that may mean that they don't lend out 100% of the underlying stock, or it might mean that they have tight stock loan agreements with their counterparties.

I guess you loaning out your XRT and SS loaning out the underlying basket just seems like a more mutiplicative form of fractional reserve banking/lending to me. I'm not worried, but if you are, explain further.

UrbanAnalyst said...

I'm glad you touched on this topic Kid. I saw the article yesterday, and Steve's valid commentary, but didn't have the time to reiterate or elaborate on any of his concepts introduced.

If we're rolling the the 5 sigma scenario of every holder wanting redemption, then the other fat tail confluent event to address might be whether liquidity in the derivative stocks dries up. I'm with you on the theoretical short squeeze, but in the flash crash market lock-up scenario you'll likely find unique downside outcomes. While the lender's exposure to their broker still stands, when an ETF is unable to deliver the underlying and the fund's assets shrink to a theoretical zero as the redemption queue spans the total shareholder list there may be a gap.

That said, I haven't read an ETF prospectus cover to cover in some time, and my memory isn't sharp to a clause, but I would imagine legalese papers over this outcome with a reasonable lag period to deliver shares, during which time presumably some sellers will arrive.

Nemo said...

This will be brief because I am late to work. Some of us have jobs, you know. :-)

Let me simplify. Forget about me, and suppose State Street did this all internally. So they go out and buy a basket of underlying stocks. They deliver it to their own ETF in exchange for XRT. Then the ETF lends the underlying shares to some other State Street group who sells them short. Finally they sell the new XRT shares into the market.

In aggregate, their net position in underlying shares is now zero (the ETF holds the physical shares, while another internal group sold synthetic shares). But their net position is XRT is negative!

Put another way, they took N underlying shares out of the market, sold them back (in synthetic form), and then sold and equivalent amount of XRT without borrowing it from anybody. From the point of view of the world outside of State Street, a whole basket of shares simply got created out of nothing and sold. That is naked shorting.

If you object to the details of this hypothetical, just add more parties to the picture until your objection vanishes. You will find that those parties, in aggregate, are effectively creating shares and selling them without borrowing them.

I am not arguing about whether naked shorting is a bad thing. All I am arguing is that this is exactly the same thing.

Nemo said...

In aggregate, their net position in underlying shares is now zero (the ETF holds the physical shares, while another internal group sold synthetic shares).

Of course I meant the ETF holds synthetic shares (since they lent them to the internal group) while the internal group sold the physical shares. Rest of description stands.

scharfy said...

Good thread...

Seems like the functionality of the creation/redemption process is the vital design component that keeps the ETF's NAV behaving as it should.

So, from a frictionless theoretical vantage point, the system might seem self-correcting.

I wonder how seamless creation and redemption is in practice, and what the potential weak points in theETF architecture are in times of stress.

One day time lags on redemptions in times of panic could cause major disjoints.

As Taleb would ask, is this robust?

Maybe not. But probably not worse than any other product where synthetic shares are created.... (futures, CDS home equity lines)

Enjoyed the commentary...

wcw said...

KD, of course the sponsor doesn't let you transact for free. You have to provide shares to redeem. However, as a bonafide market maker you can also short shares all day into the market. If all the APs want to be out of the shares come rebalancing time, they can redeem real shares while shorting into the market until they are at zero.

I take it you didn't punch up IWM.SO on the Bloomberg box. Try it.

Kid Dynamite said...

Nemo -
to simplify:

T = 0 positions:

Market: long underlying stock
SS desk 1: flat
SS desk 2: flat
SS Trust: flat

net: long 1 basket of underlying

1) Desk 1 buys underlying stock in the market, and creates XRT by delivering the stock to the Trust and receiving XRT (position update: desk 1 is long XRT, market is flat, Trust is long underlying, which they have to be against net outstanding XRT)

2)The trust is now long underlying stock which they lend out to Desk 2.

3) Desk 2 shorts the underlying stock to someone in the market, who buys them. (position update: Desk 1 is long XRT, market is long underlying, trust is long synthetic underlying, desk 2 is short underlying)

4) Desk 1 sells their XRT to someone in the market, who buys them.

t = 1 positions:
Market: long XRT, long underlying
Desk 1: flat
Trust: long a basket of underlying (synthetic)
Desk 2: short a basket of underlying

the market's long XRT is tracked by the Trust's (synthetic) long underlying. Desk two's short underlying cancels out the market's long underlying.

i see no issue.

if you're still worried, Desk two buys back their underlying basket in the market, and delivers the shares to the Trust.

T=2 positions:
Market: long XRT,
Desk 1: flat
Trust: long a basket of underlying
Desk 2: flat

all is kosher: the trust is long the underlying against the XRT that were created.

then, if you want one more step: the market redeems their XRT.

t=3 positons:

Market: long underlying
SS1: flat
SS2: flat
SS Trust: flat

right back where we started. nothing odd, right?

Kid Dynamite said...

WCW - i don't have bberg, come on! my blog revenues are like $2 a month!

"they can redeem real shares while shorting into the market until they are at zero."

until WHAT is at zero? shorting into the market, even naked shorting (And that's what you're talking about, a naked shorting exemption) doesn't drive shares outstanding to zero. I am guessing that you mean that they can naked short IWM until the price goes to zero? See my next point:

fyi, by the way, ETF desks can't naked short IWM under a market maker exemption - at least to my knowledge. we never could. If anything, they might "naked short" and buy the cash underlying basket, then promptly create new shares to deliver on their short. And any exemption only applies to market making activity. ie, they might be able to naked short IWM against client demand - in which case the price certainly wouldn't go to zero, they'd be reacting to demand - (and create by buying the cash basket) - but they can't just go spray the market with unlimited naked IWM short orders until the price goes to zero!

and I don't think the problem is that guys want to be out of IWM by rebalance time (although the effect is the same, regardless of the motivation) - it's that they want to own the underlying stock, so that they can make money rebalancing it! that's why they redeem - because the rebalance is a nice trade to have (in fact, in the good old days, it used to be a huge portion of our yearly PnL - now it's become a huge poker game of mental theory, since it's totally overplayed)

wcw said...

Bloomberg gives away some things for free. Punch up the five-year view and see 2007. Shares outstanding went near zero twice. You can, I suppose, conclude that Ma and Pa Kettle collectively decided a couple of times in 2007 to sell out of nearly all their long holdings in by far the most popular small-cap ETF vehicle extant. Alternately, you can conclude that something funny happened, and look for explanations.

I do not know what was going on at BGI at the time, but I have heard ex-BGI ETF folks mutter darkly about the events. My hypothesis is that something in the AP/MM community triggered it. I didn't say the APs and MMs were unhedged, but I do know they can be and sometimes are naked short.

Kid Dynamite said...

WCW - punch it up where? on i couldn't find it - sorry. I find bberg's website less than useless for individual stock data - maybe i'm doing it wrong. anyway - i'm not doubting you - I believe you - i wouldn't be surprised if it happened every year, I already explained why.

Didn't we already address the Ma and Pa kettle thing? they don't all need to redeem - their shares are likely being lent out without them even knowing it.

wcw said...

I linked to the URL you need under the word 'Bloomberg,' so a click should suffice. Otherwise, cut and paste

Best I can tell after asking someone who heard this recounted as a war story, BGI saw almost every share redeemed for securities on a single day around rebal time. These securities flowed right back in as creates the next day. This congrues pretty well with your experience that 'they want to own the underlying stock' -- redemption, after all, is the process of turning ETF shares into underlying securities.

There is no way Ma and Pa Kettle were involved in that, I think.

Nemo said...

KD --

Starting at t=2, your global total is net long underlying plus net long XRT. Which implies new shares got created from nowhere, which is obviously impossible.

I would argue that the trust itself is effectively short the XRT it creates. The trust is never really "long" the underlying, because its assets in shares always balances its liabilities in potential XRT redemptions. Put another way, the balance sheet of State Street as a whole does not depend on the value of the shares held by the XRT trust.

But that is a side issue. Look carefully at time t=1 in your description. Add up SS as a whole (both desks and the trust), and you get that SS is totally flat while the market is long underlying plus long XRT. So XRT got created out of nowhere, which is entirely my point.

In the case of a naked short, somebody is of course long every share that is short. So the total is always zero... But of course the total is always zero. That may be relevant to whether naked shorting is "bad", but it has nothing to do with the definition of naked shorting; i.e., selling shares you did not borrow. As I said, I am not trying to argue that naked shorting is bad; all I am saying is that this is naked shorting.

In this example, SS clearly sells shares they did not borrow. That is all I am saying. (I am not saying that the books don't balance. Of course the books balance; they always balance. Not the point.)

And whether it all happens inside State Street is not the point, either. Move the desks into other companies and you get exactly the same picture: Shares created from nowhere and with no corresponding borrow. Perhaps this is what the Bogans were getting at when mentioning naked shorting. Or perhaps not. :-)

Nemo said...

Sorry that was written so badly. I just don't have time for this today.

In a naked short, the short seller creates a synthetic long+short pair (net = zero) and then sells the long. That is the definition of a naked short.

In this example, SS is creating an long XRT + short XRT pair (the trust being effectively short XRT) and then selling the long. Or, if you prefer, they are creating XRT from nothing and selling it. Either way you look at it, it's naked shorting, period.

Kid Dynamite said...

WCW - thanks - I didn't notice the link because i'm getting all these weird blue highlight ads lately - i thought it was just another one of those. to clarify, i'm not saying ma and pa kettle had anything to do with it - i'm saying they don't have to - they aren't really long, they're "synthetic" long - they've lent their shares to the APs who do the redeeming

Kid Dynamite said...

sorry Nemo -

you've lost me. you outlined an example where one SS desk borrows stock (internally, from another) and then shorts it, and then you tell me that "SS clearly sells shares they did not borrow". I disagree. I thought we just went through it step by step.

and yes, your first three paragraphs explain each other - the trusts's longs are accompanied by some sort of XRT liability against them. Of course shares got created - that's what the creation mechanism is - but it wasn't out of thin air - stock was borrowed and delivered as it had to be.

I think the confusion arises from the desire to treat the XRT and the cash basket as identical.. that works for OTHER market participants, but not for State Street! you already explained why- the Trust is, at least in some metaphysical sense, short XRT (that they've issued to the market) against their assets (long underlying)

I'm not sure what to make of:"In a naked short, the short seller creates a synthetic long+short pair (net = zero) and then sells the long. That is the definition of a naked short."

in the end, this is all a sideshow - as Bogan incorrectly defined "naked shorts" as short interest in excess of shares outstanding.

Kid Dynamite said...

Nemo, expanding:

you wrote "But that is a side issue. Look carefully at time t=1 in your description. Add up SS as a whole (both desks and the trust), and you get that SS is totally flat while the market is long underlying plus long XRT. So XRT got created out of nowhere, which is entirely my point."

Again, you answered this already - we should probably not even count the Trust, since they aren't really "long" - they can't sell it, after all - it's got liabilities against it (the ETF)

would you feel better if i redid it?

t = 1 positions:
Market: long XRT, long underlying
Desk 1: flat
Trust: long a basket of underlying (synthetic) vs ETF liabilities - net no risk position.
Desk 2: short a basket of underlying

net positions for everyone: long XRT, which is equivalent to long underlying (which was our starting point). The XRT didn't get created "out of nowhere" it got created out of our starting underlying basket.

Kid Dynamite said...

Nemo, one more thought -

"In a naked short, the short seller creates a synthetic long+short pair (net = zero) and then sells the long. That is the definition of a naked short."

i'd say, instead, "In a naked short, the short seller, by failing to deliver the stock, creates an unaware or non-approving/non-agreeing "synthetic" long for whomever is on the other side of his trade."

If that synthetic long sells his non-shares (that he was never delivered - that's what makes it a naked short), he just creates another synthetic long to replace himself. At some point, someone insists on delivery, and the naked short seller has to find the real shares and remedy the situation.

but I don't understand how you want to apply that to our example, where every share was properly borrowed and delivered, with consent.

Nemo said...

OK, let me try one more time.

I guess we first have to agree on the definition of "naked short". I do not think it fundamentally has anything to do with a failure to deliver. Failure to deliver is just a symptom. I mean, suppose we changed the law so anybody could create a share of any stock out of thin air and sell it, so long as they also recorded a short position on their own books. That would eliminate all failures to deliver. Would that fix all problems with naked shorting? Should it be legal?

My definition of a "naked short" is a short sale that takes place without a corresponding borrow. Again, the failure to deliver is merely a symptom; the fundamental definition involves failing to borrow, not failing to deliver.

To run our example once again, Desk 1 buys a basket of shares in the open market, exchanges them for XRT with the trust, and sells the XRT. Desk 2 borrows the underlying shares from the trust and sells them.

What happens to State Street's balance sheet if retail stocks tank? The shares in the trust lose value, which is perfectly offset by Desk 2's short position gaining value.

But since the trust still has an obligation to XRT holders, State Street's net worth goes up when retail stocks tank. Desk 1 paid cash to the market and got shares; Desk 2 sold shares to the market and got cash; and Desk 1 sold XRT and got cash. If XRT goes to zero, State Street gets to simply pocket the cash from Desk 1's sale of XRT. Similarly, if XRT should skyrocket, State Street's net worth gets clobbered, thanks to Desk 2's short position and the trust's obligation to XRT holders (since only the latter is balanced by the trust's long position in the underlying).

If your net worth goes up when an asset goes down, then I don't care how you label it; you are short that asset. In this picture, State Street as a whole is net short XRT, period. (And the market is net long XRT.)

Now, where are the borrowed XRT shares corresponding to that short position? The answer is that they do not exist. State Street has effectively created XRT shares out of thin air, sold them, and then recorded a short position on its books. Should that be legal?

(Actually, I am not sure that it is. Do you know whether shares held by an ETF trust are available for borrowing?)

UrbanAnalyst said...

Not to keep kicking this dead horse, but just to clarify definitions, the Trust form of an ETF cannot of course directly participate in securities lending programs. It is a unit investment trust structure, which is prohibited from such activities as a principal.

Not to say that there aren't many ETFs structured (open-ended) such that they are entirely able to lend shares, just not Trusts. Only wanted to clarify.

Also note: SPDRs are unit trusts, iShares are or were at least originally all open-ended.

Kid Dynamite said...

UA - thanks for the clarification.

Nemo - let me work through that. i have company here now...

Kid Dynamite said...

ok Nemo, let me walk through this (and I know UA said that SS Trusts cannot lend shares, but let's pretend they can - because I still can't find the problem), as slowly as possible (for my benefit).

market is long underlying.
SS has no positions

1) Desk 1 buys underlying in the market

Desk 1 is long underlying, no one else has positions.

SS has net long exposure.

2) Desk 1 creates XRT by delivering underlying to the trust

Desk 1 is long XRT,
Trust is long underlying vs XRT liabilities.

SS has net long exposure.

3) Trust lends underlying to Desk 2, which shorts the shares into the market.

desk 1 is long XRT,
trust is long synthetic underlying vs XRT obligations,
Desk 2 is short underlying,
Market is long underlying.

State street is net flat exposure, Market is net long 1 unit.

4) Now Desk 1 sells its XRT to the market.

Desk 1 is flat
Trust is long synthetic underlying vs XRT liabilities
Desk 2 is short underlying
Market is net long 1 underlying and 1 XRT.

SS is net short XRT underlying, the market is net long 2 units, and our net sum of all participants is still net long 1 unit - of course, which it has to be. That one unit has been transformed from underlying to XRT.

so we can answer your question: "Now, where are the borrowed XRT shares corresponding to that short position?" (the net short held by SS)

Answer: they've been delivered to the "market" - to the person who bought when Desk 2 shorted - because Desk 2 didn't naked short - they borrowed (And then delivered) the shares. I'm talking about the underlying - of course - as no one borrowed and shorted XRT at all. Someone (Desk 2) borrowed and shorted the underlying though.

SS isn't net short XRT. they're net short underlying. (maybe we need to do another example where they end up short XRT??? I don't think anything will change)

Again, SS didn't create an XRT position out of thin air - they had to buy it from the market first.

sorry to frustrate you. and sorry about that pesky job ;-) I am enjoying this discussion, although it's mentally consuming me.

Greycap said...

Nemo, I don't think your analysis of this is quite right.

1. Forced wind-down.
Now agree that this is an issue, I just don't think it is the same issue that you think it is. The problem is that if the ETF is just one leg of my position, I could go to bed thinking I'm flat and wake up with a big 1-way bet. But can we agree that this is not really a problem for the retail investors at whom the original scare story was aimed? The wind-down will happen when there is insufficient open interest in owning the ETF - not likely on a breakout to the upside. So it is unlikely to cost retail upside participation.

2. Double shorting
Suppose that A has a long cash position in a stock. B wants to short it, so he borrows the shares from A and sells them to C. Now A decides he no longer likes the stock and sells it. But he doesn't have it any more! So his broker borrows it from C to deliver to D. You can see that this is just the ordinary chain of synthetic long/short positions described by Waldman and KD, except that the chain has been looped so that one link plays a double role. It is not true that a short sale occurred without a corresponding borrow. But A is in exactly the same position as State Street in your thought experiment, having "shorted the same shares twice."

You stated that it didn't matter that a single agent was playing multiple roles, but I think that is the entire substance of your example.

Nemo said...

KD --

Focusing on your T=4 state...

SS is net short XRT underlying

No. Desk 2 is short the underlying, but the trust is long the synthetic, so SS's net position in the underlying is zero.

SS is only "net short" in the sense that they have an obligation to XRT holders. That is why I said they are "net short XRT".

so we can answer your question: "Now, where are the borrowed XRT shares corresponding to that short position?" (the net short held by SS)
Answer: they've been delivered to the "market"

No. I am not talking about the (gross) short position held by Desk 2, which was properly borrowed from the trust.

I am talking about the XRT liability which is, for all practical purposes, a short position in XRT. (SS's net worth goes up and down exactly as XRT goes down and up. That is a de facto net short position in XRT.) My question is, from whom was that XRT borrowed?

In none of these examples am I talking about naked shorting of the underlying stocks. I am talking about naked shorting of XRT itself. In this example, SS's net exposure is short XRT, and there is nobody from whom that XRT was ever borrowed. (Indeed, they never borrowed anything from the rest of the market at all.)

Put another way, the global net state at your T=4 is identical to T=0, except that the market is now net long XRT while SS is net short XRT. That's it. And that XRT was never borrowed from anybody. In principle SS could do this ad infinitum, because SS as a whole never has to borrow anything from anybody to execute this hypothetical transaction.

Greycap --

I am not talking about naked shorting of the underlying. And I do not see what difference it makes whether a single company does this or several do so in (deliberate or accidental) collusion.

UrbanAnalyst --

Thanks to for explaining some the differences between unit trusts and open-ended ETFs. I think it matters a great deal whether the fund can loan out its shares... Well, if you believe naked shorting is a problem, anyway.

Kid Dynamite said...


this gets back to the question you already answered about what it means for the Trust to be "long" underlying - they don't really have long exposure from that - they have flat exposure, always - by definition. Maybe that's an important point: the trust is ALWAYS net flat exposure.

SS's short exposure doesn't come from the Trust, it comes from Desk 2. That's an important distinction. It's definitely not in XRT, it's in XRT underlying - that matters too.

The trust isn't really short XRT anymore than IBM is short its own stock when it issues stock in exchange for cash. It's just something for us to think about in terms of liability matching.

In other words, the T=4 state is really:

Desk 1 is flat
Trust is long synthetic underlying vs XRT liabilities
Desk 2 is short underlying
Market is net long 1 underlying and 1 XRT.

Now, you want to simplify (canceling out the underlying SS positions) that to:

Desk 1: flat
Trust: XRT liabilities
Desk 2: flat:
Market: net long underlying, and net long XRT (which is what trust's liabilities are against)
SS net: XRT liabilities (which you are calling "short").

this is kinda a bastardization, as the Trust's assets/liabilities don't work like that. But anyway, we can see, of course, that the net total position of everyone is still "long underlying" - after we cancel out the SS short XRT from the market's long XRT.

SS has thus "borrowed" their short position from the market, in a convoluted way, by borrowing underlying stock before the creation.

but maybe that's the whole point of confusion: we ARE actually creating shares here! That's the mechanism, as designed.

perhaps you need to think of SS as two entities: 1) the Trust, which is always flat exposure, no matter what and 2) all the other desks combined - even pretending that they borrow from each other and the trust.

Greycap said...

"And I do not see what difference it makes whether a single company does this or several do so"

So what? Dude, that's not what I'm talking about. "Do what" is the question. You have claimed that in your setup, there is not a borrow for every short. But you are wrong, as you could see if you took the trouble to work it through for yourself.

Kid Dynamite said...

Wait - Nemo - I may have had a Nemotic epiphany - shame on you for not forcing me to take these logical states to the next level earlier.


Desk 1 is flat
Trust is long synthetic underlying vs XRT liabilities
Desk 2 is short underlying
Market is net long 1 underlying and 1 XRT.
checksum: total is net long 1 XRT equivalent (just as we started)

you're saying that we do it again... In a closed environment:
5) Desk 1 buys underlying in the market (but they don't buy a new underlying, they buy the one that the market is net long,

T = 5:
Desk 1 is long 1 underlying
Trust is long synthetic underlying vs XRT liabilities
Desk 2 is short underlying
Market is net long 1 XRT.
total checksum exposure: net long 1 XRT equivalent

6) Desk 1 creates XRT from their underlying, delivering underlying to the Trust,

t = 6:

Desk 1 is long 1 XRT
Trust is long 1 synthetic underlying and 1 real underlying vs 2 XRT liabilities
Desk 2 is short underlying
Market is net long 1 XRT.
checksum: long 1 XRT equiv

7) Desk 2 borrows the underlying stock from the Trust, shorts it in the market,

T = 7

Desk 1 is long 1 XRT
Trust is long 2 synthetic underlying vs 2 XRT liabilities
Desk 2 is short 2 underlying
Market is net long 1 XRT, 1 underlying
checksum: net long 1 XRT equiv

8) Desk 1 sells their XRT to the market

Desk 1 is flat
Trust is long 2 synthetic underlying vs 2 XRT liabilities
Desk 2 is short 2 underlying
Market is net long 2 XRT, 1 underlying
checksum: net long 1 XRT equiv

now, the market wants to redeem their 2 XRT. so they tell the trust. the trust says "oh crap, we don't have 2 underlying, we only have 2 "synthetic" underlying, so they recall the borrow from Desk 2. Desk to goes out in the market and goes to buy 2 underlying - but there aren't two underlying... there's only 1 underlying - there only ever WAS one underlying... BINGO!


9) Desk 2 buys 1 underlying from the market and return their borrowed shares to the Trust:

t = 9

Desk 1 is flat
Trust is long 1 synthetic underlying and 1 real underlying vs 2 XRT liabilities
Desk 2 is short 1 underlying
Market is net long 2 XRT
checksum: net long 1 XRT equiv

10) so the Market redeems one of their underlying with the trust:

t = 10

Desk 1 is flat
Trust is long 1 synthetic underlying vs 1 XRT liabilities
Desk 2 is short 1 underlying
Market is net long 1 XRT, and 1 underlying

checksum: net long 1 XRT equiv


11) desk 2 can go out and cover their other short by buying the 1 underlying unit back from the market and returning the shorts they borrowed from the trust...

t = 11

Desk 1 is flat
Trust is long 1 real underlying vs 1 XRT liabilities
Desk 2 is flat
Market is net long 1 XRT
checksum: net long 1 XRT equivalent

12) now the market can redeem their final XRT vs the trust!

t = 12:
desk 1 is flat
trust is flat
desk 2 is flat
market is net long 1 underlying - right back where we started...

so, my epiphany failed - sorry - I still don't get you. the process looks like it worked fine to me.


Kid Dynamite said...

actually, Nemo, maybe my last comment WAS your point... it worked EVENTUALLY, but it had to be done piecemeal - the market couldn't redeem all of their XRT at once - only one at a time to unwind the circle and repeat the unwind process in reverse...

was that your point?

anyway, since the Trust cannot loan out shares, this is all a moot point, just a fun exercise, and not something to really worry about.

wcw said...

Aren't these things just 40-Act funds? They can do securities lending.

Nemo said...

was that your point?

Essentially yes. My point is only that this is a kind of naked shorting. Whether it is a problem depends on whether you believe naked shorting is a problem.

When someone naked short-sells you a share of ordinary stock, it results in a "failure to deliver". If you go and pound on their door and say "give me my share, dammit", the only way they can comply is to purchase it on the open market (thus closing their short) and then deliver it to you.

But this is exactly the situation SS is in in our hypothetical. Their net position in the underlying is zero, but their net short position in XRT is theoretically limitless. When someone demands to redeem some XRT, SS has no shares to deliver; they must purchase them on the open market, exactly like someone who naked-short-sold you ordinary shares. The only difference is that there is no visible failure to deliver up front. I don't think that changes the fundamental picture.

Now, are failures to deliver -- or difficulties delivering -- necessarily so bad? I mean, it is not very different from having the shares delivered to you and then immediately loaning them out again to parties unknown... Either way, you wind up holding a promise for a share, not a physical share. It is just a number in a computer, and if you net all of the long+short positions in the world (including the naked shorts) you always get the correct number of shares outstanding.

So I am not really trying to argue about what "the problem" is. I am only arguing that our hypothetical arrangement is functionally equivalent to naked shorting.

Of course, this whole argument hinges on whether the trust can loan out its shares. There seems to be some disagreement about that... I would be curious to read an authoritative answer.

bbg said...

Nemo - If you substitute "ADR" for "ETF" does it make any difference to how you think about the issue? In both cases we're talking about a tradeable wrapper, and as far as I'm aware the structural issues are quite similar if not identical.

Kid Dynamite said...

Nemo -

i don't want to make this a semantic debate, but after going through the exercise and convincing myself that everything still works even in our unusual rules example, I am maintaining my insistence that you're using the wrong term in "naked shorting." I think that you seem to have some issue with the concept of "fractional reserve lending."

I think you can prove it to yourself by doing the same exercise with no short sales - just borrow and create.

Ie, Market starts out long underlying.

SS Desk 1 borrows the market's stock (but doesn't sell it short!), and delivers it to the Trust to create XRT.

T = 1 positions:

Market: synthetic long underlying, with an IOU from Desk 1
Desk 1: IOU for underlying to Market, long XRT
Trust: long underlying vs XRT liabilities.

now, Desk 1 can repeat a cycle of borrowing the Trust's long underlying, accruing IOUs (to deliver stock back) and creating XRT indefinitely.

there are no short sales, and hence, certainly no naked shorts.

When the entire scenario needs to be "fixed" - it can be - it just needs to be done one cycle at a time, just like it was created. Desk1 can redeem 1 XRT,

ps - this claim of yours is false:
"When someone naked short-sells you a share of ordinary stock, it results in a "failure to deliver". If you go and pound on their door and say "give me my share, dammit", the only way they can comply is to purchase it on the open market (thus closing their short) and then deliver it to you."

they can comply by simply borrowing the stock too (with consent, creating a consenting synthetic long), which is pretty much what SS did in our example.

Nemo said...

bbg --

An ADR is exchangeable for a foreign share (correct?), but those foreign shares are not sitting in an account where they can be loaned out. That is the critical difference.

KD --

No, the difference in your new example is that SS is actually borrowing the underlying from the market. In my example, they are getting short XRT without borrowing anything from the market. That is the definition of naked shorting.

Let me try yet again. Simplify the picture by imagining that Wal-Mart were the only constituent of the XRT (hey, give it time). Suppose that WMT were impossible to borrow. Finally, suppose SS could simply create XRT out of thin air and sell it at will, recording nothing on their books other than an obligation to deliver WMT in exchange.

The only reason this does not cause a "failure to deliver" is that whoever bought the XRT is not bothering (or is unable) to demand a redemption.

And no, the naked short seller cannot simply borrow the shares and deliver them; if they could, they would have done that in the first place. The whole reason for naked shorting is that the underlying is difficult or impossible to borrow. In fact, that is another perfectly good definition of naked shorting: If you manage to get short an impossible-to-borrow stock, then you are naked shorting.

Nemo said...

One last thought. If an instrument has 5x or 10x of its float sold short, wouldn't you expect that instrument to be hard to borrow?

Are ETFs like XRT ever hard to borrow? If not, why not?

Kid Dynamite said...

Nemo - feel free to diagram your XRT-WMT example as I did the other one above to try to illustrate your point - I will work on it on my end to see if i can get your point. (although I still doubt it, since I still disagree with your claims like "In my example, they are getting short XRT without borrowing anything from the market." SS bought the stock from the market in step 1 - and then repeatedly borrowed/lent it back to various internal SS desks. They did take delivery of stock from the market in step 1 - they had to, obviously.)

to your other questions: yes - I would expect that a higher short interest relative to float would generally result in tougher borrow - because there's a higher probability that the borrowed and shorted shares will eventually end up with someone who won't lend them out. This doesn't mean that it's impossible to borrow, and it doesn't mean that shorts are naked shorts.

and yes - of course XRT gets hard to borrow - I tried to short it several times between April and July and was not able to via Etrade. I don't know what it's like right now.

as to the question about the fact of trusts loaning out their shares, this is what my people told me this morning:

1) "I know that MER Holders cannot lend out the longs." - he's talking about SMH, BBH, etc. the Holders Trusts

2) "i believe that UIT's are restricted but most ETF's are allowed under prospectus to loan out their holdings"

3) the following trusts do not loan out their shares: SPY MDY QQQQ DIA

Kid Dynamite said...

Nemo - I don't get your WMT example. If WMT is impossible to borrow, how will SS (or anyone else) create XRT out of thin air? they have to deliver WMT to the Trust (even if it's borrowed WMT - which you've just declared impossible in your assumptions) each time they create XRT.

If WMT can't be borrowed, XRT can't be created - except by people who already have WMT, of course, but in our example we assume that the T=0 state is:

Market: long 1 WMT
SS desk 1: no positions
SS desk 2: no positions
SS trust: no positions.

Desk1 buys WMT from the market, and delivers it to the trust to create XRT. Now positions are

t = 1

Market: flat
SS desk 1: long XRT
Trust: long WMT vs XRT obligations.
SS desk 2: flat

you've declared WMT impossible to borrow, so the game is over. there's no more borrowing WMT to create XRT.

Nemo said...

I still disagree with your claims like "In my example, they are getting short XRT without borrowing anything from the market." SS bought the stock from the market in step 1

Right, they bought it, they did not borrow it. Enormous difference. Kind of my whole point, actually.

Hypothesis is that WMT is impossible to borrow (in general). That does not mean it is impossible to buy and sell! So Desk 1 buys WMT from the market, swaps it for XRT with the trust, borrows WMT from the trust (which is always willing to lend), and sells it again.

Or even move Desk 1 out of the company. Suppose nobody in the world wants to loan out their WMT. Fine, so you buy WMT, swap it for XRT with the trust, borrow WMT from the trust (who is always willing to loan), sell them again, and sell the XRT. You are now net short WMT free and clear, even though it was supposedly impossible to borrow!

In essence, even if nobody in the world wants to loan out their WMT, anybody can create a willing WMT lender simply by purchasing shares and delivering them to the trust. And this can be repeated ad infinitum. (This is slightly different from my previous example. But both involve getting short an instrument none of whose current holders is willing to lend it to you.)

This effectively bypasses any sort of restrictions on naked shorting... If you can merely buy the stock and sell the stock, you can automatically short the stock just by virtue of the ETF's existence. Assuming the ETF is a willing lender, that is.

Kid Dynamite said...

correction to earlier info: SPY does lend out their stock.

Kid Dynamite said...

Nemo - I fear this is going to deteriorate quickly.

If the trust is always willing to lend WMT, it's not impossible to borrow. By definition. You're making your own rules, and then violating them a sentence later.

again, note that your claim: "both involve getting short an instrument none of whose current holders is willing to lend it to you." is not debatable - it's false. The Trust is a holder, and in your own new rule, the Trust is willing to lend WMT.

If we stick to the consistent, normal meaning of "impossible to borrow," then the trust does NOT loan WMT, and you can't magically get short it, as I illustrated already above. It's one or the other - either the Trust loans WMT, it's not impossible to borrow, and you can short it, or it's impossible to borrow, they don't loan it, and you can't short it.

Anonymous said...

Outstanding commentary

Andrew Bogan said...

Great to see some serious discussion here, thanks.

As a clarification, the nakedness of the massive short interest in some ETFs is not material to our argument. These massive short positions in excess of the shares outstanding, are certainly serial short, if not technically naked (and are quite possibly some of both). We revised our article slightly to clarify any confusion.

Nemo said...

KD --

Actually, I think we are getting close to understanding each other...

If the trust is always willing to lend WMT, it's not impossible to borrow. By definition.

Not at all! Just because one holder of WMT is willing to loan it out, that does not mean the stock is "easy to borrow", because that one holder can run out of shares to loan.

For any stock, there are always some holders who are willing to loan their shares. But once everyone who is willing to loan shares runs out of shares to loan, the stock will become "impossible" to borrow.

If the trust does not exist, then once that happens you cannot magically (and free of cost) create a new holder of physical shares who is willing to loan them out. But with the presence of the ETF, you can; you simply buy the shares, swap them for XRT, and sell the XRT. This costs you a net of zero (except transaction fees), and suddenly the stock becomes possible to borrow again because the trust will loan out the shares you just delivered.

So I stand by my claim that the existence of these ETFs makes it possible to bypass naked shorting rules. In fact I am starting to suspect this is the primary reason they exist at all...

Kid Dynamite said...


I don't want to be rude, but you're just talking nonsense now. Terms have meanings - you can't just redefine them to fit your thesis. Impossible to borrow means no one lends. Not "no one lends except the trust."

you are not ever going to encounter a situation where no one in the world will lend their shares, except some monkey at the Trust who keeps loaning them out mindlessly - but if you did, the stock wouldn't be impossible to borrow - it would be borrowable from the Trust (just as you DEFINED it to be!).

and you keep mis-using the term "naked shorting." Naked shorting is shorting without a borrow. In every one of our examples, we have a borrow.

Kid Dynamite said...

Here Nemo, let me try to rewrite what I think your point is:

If market participants are loose with their stock lending practices, there is a limit to the effect, since they can only lend their shares ONCE. But if the trust is too loose with its lending practices, the effect can be much greater, since the created ETF baskets will continually flow back into the trust so they can repeat their loan process.

How's that sound?

Of course, we already showed that this can all still be unwound (piecemeal) and also that it's not "naked shorting."

Anonymous said...

I have a whole new question reguarding the creation/ redemption process. I understand how this works for a simple case of an ETF that consists of a basket of underlying stocks, but how do the APs arb the inverse ETFs, especially the multiple inverse ETFs which use derivatives to give inverse daily percentage closes. The creation or redemption process must be far more complicated than a simple index ETF. I am assuming the APs know far more about the composition of the inverse ETFs than I do, as reading the prospectus tells me nearly nothing about how the inverse ETF tracks inversly. In any case, assuming the APs do have more information, I feel that the markets for the derivatives that compose the inverse ETFs could possibly pose far more questions about the stability of these inverse ETFs, and about the ability of the APs to efficiently stabilize the ETF to its NAV.

To summarize, my questions are 1.)Do inverse ETFs' creation and redemption process work the same as regular ETFs? 2.) If so, how do APs correctly arb the inverse ETFs with such a difficulty in assessing the underlying instruments they would have to be buying or selling to compose the ETF? 3.) Does the fact that the underlying basket is made up of dervatives pose more problems in times of panic?

Kid Dynamite said...

Anon - great question. let's talk about it:

using the SKF as an example: they are both inverse and leveraged (financial basket). They achieve their returns by using a series of daily rebalanced and daily rolled swaps and other derivatives.

You (well, APs, of course) actually can create and redeem SKF - and when you do, Profunds (the ETF manager) will likely increase or reduce the quantity of swaps they have on their books. How do they do this? well, the counterparty for their swap has short positions in the underlying basket of stocks - so Profunds would pass this through to their swap counterparty when reducing/increasing the daily swap notional. (remember, Profunds is short the performance of a financial basket on swap - so their counterparty is long the performance of a financial basket (on swap to Profunds) and short a basket of financial stocks as a hedge.

this is why the leveraged and inverse ETFs also frequently do a pretty darn good job of tracking what they are supposed to. APs just need to deliver the underlying basket - which the Trust will publish for them - they don't need to deliver "swaps" to the Trust.

now, do they pose more risks? You may have already heard about the "gamma" effect of daily rebalancing - profunds mentions it in their prospectus:

"At the close of the markets each trading day, the Fund will seek to
position its portfolio so that its exposure to its benchmark is consistent
with the Fund’s investment objective. The impact of the
Index’s movements during the day will affect whether the Fund’s
portfolio needs to be re-positioned. For example, if the Index has
risen on a given day, net assets of the Fund should fall, meaning
that the Fund’s exposure will need to be decreased. Conversely, if
the Index has fallen on a given day, net assets of the Fund should
rise,meaning the Fund’s exposure will need to be increased."

this can lead to volatility around the close.

the SKF also faced its own problems during the crisis when short selling of financial stocks was banned - they stopped creations and redemptions as a result of this. (or maybe just creations - I can't remember)

Unknown said...

Diamond hands GME GME GME