Posts Tagged ‘Bats’

FaceBook: The Complete Forensic Post-Mortem

Saturday, May 19th, 2012

 

While much has already been written on the topic of peak valuation, social bubbles popping, and the ethical social utility of yesterday’s historically overhyped IPO, nobody has done an analysis of the actual stock trading dynamics as in-depth as the following complete forensic post-mortem by Nanex. Because more than anything, those tense 30 minutes between the scheduled open and the actual one (which just happened to coincide with the European close), showed just how reliant any form of public capital raising is on technology and electronic trading. And to think there was a time when an IPO simply allowed a company to raise cash: sadly it has devolved to the point where a public offering is a policy statement in support of a broken capital market, which however is fully in the hands of SkyNet, as yesterday’s chain of events, so very humiliating for the Nasdaq, showed.

From a delayed opening, to 2 hour trade confirmation delays, virtually everyone was in the dark about what was really happening behind the scenes! As the analysis below shows, what happened was at times sheer chaos, where everything was hanging by a thread, because if FB had gotten the BATS treatment, it was lights out for the stock market. Well, the D-Day was avoided for now, but at what cost? And how much over the greenshoe FaceBook stock overallotment did MS have to buy to prevent it from tumbling below $30 because as Reuters reminds us, “had Morgan Stanley bought all of the shares traded around $38 in the final 20 minutes of the day, it would have spent nearly $2 billion.” What about the first defense of $38?  In other words: in order to make some $67 million for its Investment Banking unit, was MS forced to eat a several hundred million loss in its sales and trading division just to avoid looking like the world’s worst underwriter ever? We won’t know for a while, but in the meantime, here is a visual summary of the key events during yesterday’s far less than historic IPO.

May 18 – The Facebook IPO

The first warning sign, was the delay in trading. Here’s the status messages from Nasdaq for that day.

The first 4 charts are 5 second interval charts of Facebook showing the first hour and 15 minutes of quotes and trades.

Chart 1. NBBO (National Best Bid or Offer) Spread. Black: bid < ask (normal), Yellow: bid = ask (locked), Red: bid > ask (crossed)all bids and offers color coded by exchange.

Chart 2. Best bids and offers (NBBO) color coded by exchange.

Chart 3. All bids and offers color coded by exchange.

Chart 4. All trades color coded by exchange.

The next 4 images are tick charts showing quotes and trades. How to read these charts

Chart 5. The first seconds of trading.

Chart 6. The first seconds of trading, continued.

Chart 7. Suddenly, a vacuum appears and produces a record 12,285 trades in 1 second.

Chart 8. Same as above, showing just Nasdaq.

The next 2 charts (10 second interval) show how Nasdaq’s quote stopped, but trades from Nasdaq did not (direct feeds must have been fine, but not the consolidated).

Chart 9. Nasdaq Bids and Offers along with NBBO.

Chart 10. Nasdaq Trades

The next 2 charts (20 millisecond interval) show the effect when Nasdaq’s quote returned. There were two significant gaps in quotes (for all exchanges) and 1 significant gap in trades.
Note how the gap in trades is not at the same time as the gaps in quotes.

Chart 11. All bids and offers color coded by exchange.

Chart 12. All trades color coded by exchange.

The next chart (5 millisecond interval) shows the result of the blast in trades and quotes when Nasdaq’s quote returned. Trades printed at least 900 milliseconds before quotes, an impossibility if orders are being routed according to regulations. We have jokingly referred to this anomaly as fantaseconds.

Chart 13. Nasdaq bids and offers (triangles), Nasdaq trades (circles) and NBBO (gray/yellow/red shading).

The next 2 charts (500 millisecond interval) detail the HFT Tractor Beam area where coincidentally or not, Nasdaq quotes began “sputtering” right before stopping for about 2 hours.

Chart 14. NBBO Spread and quote rate from all exchanges.
Note the flat lines at the bottom. Also note how the quote rate (lower panel) surges when prices rise above the flat line, which is what we would expect. However, on Nasdaq (next chart)..

Chart 15. NBBO Spread and quote rate from just Nasdaq.
When prices rise above the flat line, quotes from Nasdaq stop, exactly opposite of expected behavior and what we see from other exchanges at that time (see chart above).

 

And finally, Nanex on the fallout:

During the FaceBook’s failed IPO opening period (11 – 11:30)  and shortly after the trading began, bad prices (spikes) began appearing in other stocks, including symbols APPL, INTU, NFLX, PDCO, QCOM, QLD, UST and ZNGA. They also occurred in Facebook during the first 15 minutes of trading (see Chart 4 on this page). There are likely other stocks that were affected. In nearly all of these cases the price spikes were executing against quotes that were far outside the NBBO. Most of these executions occurred on the CBOE, and a few on Chicago and AMEX. Fortunately, by chance, the prices were not wide enough to trigger circuit breakers in these stocks.

We think these bad price executions are related to whatever issues Nasdaq was having in facebook and probably are from errors in routing software. A similar thing happened during BATS failed IPO in AAPL and other stocks.

Chart 1. AAPL

Chart 2. NFLX

Chart 3. QCOM

Chart 4. QLD

Chart 5. UST

Tags: , , , , , , , , , , , , , , , , , , ,
Posted in Markets | Comments Off

Homer Simpson’s Financial Markets and “Fixed Income” Products (Tchir)

Sunday, March 25th, 2012

 

by Peter Tchir, TF Market Advisors

Stop tomorrow’s problems today.

Just this week we had:

TVIX – an ETN that provides double the daily change in the vix futures.  Who is smart enough to be able to take big bets on VIX futures that doesn’t have a futures account?  Who is this designed for?

MF Global & “customer money” – months after the problem, no good explanation of where the money went, and even more scary, is that the it remains unclear whether MF did anything illegal with customer money.  Our understanding of how our money should be treated, and the legal rights we have signed away don’t necessarily match up.

CPDO – the legal battle in Australia over this disaster continues.  In the top 3 of mis-rated product of all time.  You take something that is BBB+ on average, LEVERAGE it, and get AAA.  It relied on “self-insurance” the thing partly responsible for the equity crash in 1987.

Greek CDS auction – finally a Credit Event occurred and settled this week.  Very few people still seem to understand how lucky CDS holders were that the auction on New bonds delivered a real payout.  The system didn’t fall apart as some had worried, but no reason that CDS cannot be at least 90% cleared, or better yet, traded on an exchange.

BATS – “Making Markets Better” according to their website had to pull their own IPO.  Maybe they didn’t realize algo’s don’t provide actual liquidity, all they do is take real liquidity from exchange and run around the electronic world trying to scalp a few fractional cents not available to individual investors anyways.  If they have to list on a proper exchange, people really should question the need for these other exchanges, sub-penny trading, etc.

I’m all for some complexity and innovation, but it does seem after a week like this, that the financial markets have become too complex, and some real effort should be made to simplify things and put everyone on an even playing field.

Which brings me to a story I’m just getting up to speed on.  It seems like banks and investment banks are working on ways to satisfy their customer’s demand for yield.  They should come with a  warning that “yields in hindsight may be smaller than they appear”.  I haven’t been able to confirm that this is being sold to retail or how much has been done, but I decided to poke around in some bonds listed by Citibank – mostly because somehow they seemed to have needed more support from the taxpayers than any other bank (except for BAC which I have picked on too often).

So let’s take a look at what appears to be a Citibank NA Certificate of Deposit – how dangerous could that be?

It seems a bit long for a CD – 2032 final maturity, especially since it is callable at any time.  According to this it hasn’t been issued yet, so maybe this is all a bad dream, but since I was able to find it on Bloomberg, it probably is something they are trying to sell.

So on any “fixed income” product, the big question is what is the coupon?  It pays 6%!

Ok.  I could buy Citigroup Inc 5.85% bonds with a 2034 final maturity.  They are non-call and priced around 103.5 to give a yield of 5.57%.  So stop right there.  The CD may be marginally higher in the capital structure and slightly safer, but for 20 years I would much rather have 5.57% non callable bond rather than a 6% bond callable at any time.  I would spend more time working out the value of the call and if the trade-off is even remotely fair, but there is no point, because the coupon isn’t “fixed” it resets annually.

So after 1 year, the coupon will be 5% minus 6 month LIBOR at the time.  If today was a “setting” date, the coupon would be only 4.25%.  So as short term rates rise in the future, this coupon on this Inverse Floater will go do.  If 6 month Libor is ever at 4% or above on a setting date, then this bond will have the “floored” coupon of 1%.  So if the Fed starts raising rates or LIBOR goes up because bank credit risk deteriorates, you own a low coupon bond in either a high rate environment, or weak bank credit environment.

But this “Certificate of Deposit” looks tame compared to another they seem to be marketing at the same time.  Again, I don’t know for certain that they are marketing this, but it does show up on Bloomberg under a list of Citi bonds, so I have to assume it isn’t there by accident.

So this one is a “dual range accrual”.  So it look like you have to track the number of days in a period where 3 month Libor is between 0% and 5% and the Russell 2000 is above 75 (maybe they mean 750?).  If both conditions are met for the entire period, you get a 4.25% coupon.  So a Citibank CD that is callable at any time, has a best case coupon of 4.25%, and could be 0% in either a high rate environment (libor above 5% or in a weak stock market the RTY is below the threshold).  Retail investors are selling options hand over fist with the promise of some decent yield in the first year.  I find it hard to believe they understand the options they are selling, and I find it impossible to believe that  they are selling the options at anything close to fair value.

Stop tomorrow’s problems today, but if you are show a “fixed income” product where the coupon is too good to be true, it is too good to be true!

 

Copyright © TF Market Advisors

Tags: , , , , , , , , , , , , , , , , , , , ,
Posted in Markets | Comments Off

May 6, 2010 – The Day That Will Change Market Structure

Friday, May 7th, 2010

This article is a guest contribution from Tyler Durden, ZeroHedge.com.

May 6 market commentary from our friends at Themis Trading

May 6, 2010 – The day that will change market structure

Today’s action left us amazed, and we have been warning about this stuff since December 2008. Where do we even start? Yesterday afternoon and evening all the business programming focused on how the markets were in turmoil, and Greece this, and overdue correction that, and fat finger the other thing. They couldn’t even recognize the story, as even the business media doesn’t understand that the markets are a changed structure and beast. The story is not a key-punch error. The story is a failed market structure. The market failed today.

The market melted down and “liquidity providers” quickly pulled all bids. According to today’s Wall Street Journal, high frequency firm, Tradebot, closed down its computer systems completely, as did New Jersey’s own Tradeworx, who was so critical of our silly market structure comments in their SEC comment letter. By the way, if you don’t know who or what Tradebot is, it is the proprietary trading engine that used to be part of the BATS exchange. In fact the reason BATS was rolled out as an exchange to begin with was to lower costs and facilitate trades for Tradebot (Tradebot’s 1251 NW Briarcliff Pkwy Kansas City address is next door to BATS’s North Mulberry Drive address fyi). In the WSJ article Mr. Cummings said his Tradebot system was designed to stop trading when the market becomes too volatile, because he “doesn’t want to compound the problem.” Too bad he doesn’t understand that that was and is the problem. To make matters worse, while some high frequency firms shut down yesterday and pulled their bids, as we warned they would do for over a year and a half, other high frequency firms turned from being liquidity providers to liquidity demanders, as they turned around and indiscriminately hit bids like Randolph and Mortimer Duke.

We are just plain outraged, and think every investor and market participant in the USA should share this outrage. They were sold a lie. How many times over the last year have we all heard that HFT liquidity was a blessing that lowered costs and helped investors, and that it would be there in stressful markets just like the market makers and specialists they replaced were there? How many times have you read in the big media that HFT helped the markets perform brilliantly during the global meltdown in 2008 and 2009? We said it before and we say it now. Lies.

Not so long ago, if our markets experienced severe stress, and certainly a “fat finger”, human wisdom would intervene. Reasons for the stress would be ascertained, trading in affected stocks would be slowed or halted, stabilizing bids would be initiated as needed, and severe volatility would be dealt with in a calm and reasoned manner. Today, the human specialist model has been replaced by an automated market maker model. Our market structure has evolved. It has evolved, not by design,?or a well-thought and reasoned plan, but it has evolved to cater to masters of expensive technology, deployed unfettered by participants whose only concern is to squeeze out every last picosecond and fractional cent before they move on to other countries’ markets and asset classes. The for-profit exchange model at every chance sacrifices the protection of long term investor interests for the profitability of serving hyper-leveraged intraday speculators. By the way FLASH orders are still utilized at Direct Edge, but that is here nor there.
Today’s price swings in a great number of stocks highlight the inherent and systemic risk of our automated stock market, which has few checks and balances in place.  Once the market sensed stress, the bids were cancelled and market sell orders chased prices down to the lowest possible point. Investors who thought they were protecting themselves with the prudent use of stop orders were left with fills that were far away from the closing price. In some stocks like our SAM example above, this was $0.01. We warned of the potential for HFT to behave this way when we met with and showed our regulators the NY Fed study that highlighted HFT’s vanishing act around stressful news announcements in the currency markets.

We read this in a recent comment letter to the SEC about HFT and couldn’t agree more: “When markets are in equilibrium these new participants increase available liquidity and tighten spreads. When markets face liquidity demands these new participants increase spreads and price volatility and savage investor confidence.”

The EXCHANGES’s response late yesterday was to cancel trades that moved by more than 60%. Yes 60%. SO if you bought a stock at $21, put in a stop-loss market order at $20 (expecting to get filled in a market decline of somewhere less than but close to $20), and got filled at $10 (yes this happened and worse), your trade stands! And if you bought this same company’s stock (that fell from $20 to $3 before closing back at $18) at $3 and sold it at $14 thinking you made a big profit, your buy is cancelled, you are short stock at $14, you have a loss, and the futures are green this morning. Inspires investor confidence, right? With this wise remedy and redress by our exchanges, along with their other maneuvers (stay tuned for our coming Data Feed White Paper), one can’t help but be confident in playing ball on this level playing field. NOT.

Today’s severe market drop should never have happened.  The US equity market had at been hailed as the best, most liquid market in the world. ?The market action of May 6th has demonstrated that our equity market has major systemic risks built into it.  There was a time today when folks didn’t know the true price and value of a stock. The price discovery process ceased to exist. High frequency firms have always insisted that their mini-scalping activities stabilized markets and provided liquidity, and on May 6th they just shut down. They pulled the plug, as we always said they would, and they even admit it in the papers this morning. We need a new mousetrap. This is not an isolated incident, and it will happen again

Source: ZeroHedge.com, May 6, 2010 – The Day That Will Change Market Structure, May 7, 2010

Tags: , , , , , , , , , , , , , , , , , , ,
Posted in Energy & Natural Resources, Markets, Oil and Gas | Comments Off