In the aftermath of the October 15th, 2014 Treasury flash crash that was much fake "confusion" among the punditry about what caused the dramatic 20-sigma move in the 10 Year treasury. For us, however, there was no confusion, it was all due to a vicious case of HFT algo quote stuffing - a key component of algos trying to establish whether there are credible size orders to be frontrun - gone horribly wrong.

Several months later, in July, the Joint-Staff Report released by the Treasury, Fed, SEC and CFTC confirmed as much, and even if they didn't explicitly single out HFTs as the culprit for the flash crash (that would mean having to redo the topography of the market, in the process gutting and redoing the entire market structure after tacitly admitting the market is broken), they did very clearly note that it was "self-trading", or quote stuffing, that was responsible for the unprecedented move.

Here are the only two charts that mattered from that report:

As we thoroughly documented back in July, this is what the staff report said: "Given the finite capacity of any matching engine to simultaneously process messages and execute matches between buyers and sellers, extremely high message rates appeared to cause trading platform latency to temporarily jump higher", or as we explained it " a massive burst of quote stuffing (seen with absolute clarity on Figure 3.29 above) in the form of a surge in messages, resulted in a burst of accumulated order latency, which in turn was the catalyst to send the price soaring from 129 to over 130 in the span of 5 minutes, and then sliding back down again once the quote stuffing effect was eliminated."

Which brings us to our conclusion then:

... what is surprising is that unlike the SEC's Flash Crash report which was a travesty and blamed the crash on Waddell and Reed, to be followed by another travesty of a report, one which has sent an innocent trader behind bars, this time HFT is explicitly, if not deliberately, singled out.

Which in our opinion sets the stage. The stage for what? Why blaming the upcoming market crash on HFTs, of course.  As Bloomberg commented, these findings "will probably add to regulatory scrutiny of the industry."

The reality is that regulators know very well what is really going on in the markets, and now that HFTs have been exposed as the catalyst for the bond market crash, when the inevitable stock market crash - a crash that will be the result far more of the ruinous decisions of central planners around the globe - it will be the HFTs, pardon, PTFs that will be the first to blame, while the central bankers do their best to quietly slip out to a non-extradition country.

Just look at China: the government is so terrified of losing control over its own stock market bubble and the potential for violent, social conflict that would result, that it will throw everything at the market to support it. In the US, the regulators are already one step ahead: they know a crash is inevitable, and the only thing they need is the scapegoat to blame it on when it all comes crashing down.

Nameless, faceless algos would be just the perfect scapegoat.

Today we are one stop closer to that inevitable moment when the enabled systemic parasite, high frequency trading,which exists solely as a result of the market overhaul allowed in the aftermath of Reg NMS, is rooted out.

In a report authored by NY Fed economists Dobrislav Dobrev and Ernst Schaumburg, we get one step closer to the regulators admitting what we have said since day one: HFT does not provide liquidity (although it does provide a whole lot of liquidity-rebate generating volume), it provides a "liquidity mirage."

In the note the authors roundly crush the biggest, and frequently only, benefit of HFTs as stipulated ad nauseam by its advocates, namely an increase in "market efficiency and pricing developments." The authors note:

"that the (price) efficiency gain comes at the cost of making the real-time assessment of market liquidity across multiple venues more difficult.

* * *

... it has arguably become more challenging for large investors to accurately assess available liquidity based on displayed market depth across venues.

* * *

This situation, which we term the liquidity mirage, arises because market participants respond not only to news about fundamentals but also market activity itself. This can lead to order placement and execution in one market affecting liquidity provision across related markets almost instantly. The modern market structure therefore implicitly involves a trade-off between increased price efficiency and heightened uncertainty about the overall available liquidity in the market."

* * *

The striking cross-market patterns in trading and order book changes
suggest that quote modifications/cancellations by high-frequency market
makers rather than preemptive aggressive trading are an important
contributing factor to the liquidity mirage phenomenon.

Goodbye to "fat fingers" being blamed for flash crashes, and welcome to the Heisenberg uncertainty market:you can have your 1 cent bid/ask spreads... but you can't have any real market depth at the same time.

And the moment you try to buy or sell a big chunk in Treasurys (or any other asset class), that tight bid/ask spread explodes as HFTs yanks opposing offers (or bids), and all the telegraphed market depth evaporates in an instant, leading to events like October 15, 2014.

As Bloomberg summarizes the note, which adds nothing new to what we have said over the past 6 years, "after examining trading across the most active platforms, including futures through CME Group Inc. and cash Treasuries on ICAP Plc’s BrokerTec and Nasdaq OMX Group Inc.’s eSpeed, the researchers found evidence that high-frequency traders create an illusion of liquidity in the Treasuries market."

With their stealth technology, high-frequency traders are able to detect competing investor orders on one of the trading venues, and with a five millisecond delay -- the shortest possible transmission time between the CME and BrokerTec -- they’re able to pre-empt the order that’s likely to appear on the other venue.

Once again: what HFTs do in a normal state is not trading; it's frontrunning and trying to evaluate just how many of the other concurrent "orders" in the market are just as fake; needless to say, the one with the fastest server and most expensive colo box wins, even if they never actually provide liquidity.

Investors often submit orders to buy or sell to all three venues in an effort to get the most competitive prices. The order is likely to reach one of the trading venues first, which gives the high-frequency traders the opportunity to profit from the time lapse.

The moment a real order does enter the marketplace, the quasi equilibrium represented by the order book disappears in an insant leading to the liquidity mirage the NY Fed has exposed.

How did the two authors reach their conclusion, one which has been known to our readers for years?

The researchers pointed out a trade that may be completed by an investor on BrokerTec.

"As soon as the BrokerTec transaction is observed in the market data feed, co-located low-latency market participants may immediately seek to cancel top-of-book offers on eSpeed and CME or submit competing buy orders to eSpeed and CME," researchers Dobrislav Dobrev and Ernst Schaumburg wrote on the N.Y. Fed’s blog. Top-of-book orders reflect the highest buy and the lowest sell prices. Low-latency is another term for the high-speed trading technology.

"The striking cross-market patterns in trading and order-book changes suggest that quote modifications/cancellations by high-frequency market makers, rather than preemptive aggressive trading, are an important contributing factor to the liquidity mirage phenomenon," the researchers wrote.

Worse, the researchers "did not find any evidence that the liquidity mirage was more pronounced on Oct. 15 compared with our control days." In other words, courtesy of HFTs, multiple-sigma events like October 15are always just around the corner, and always threaten to unleash market chaos the moment some unexpected "shock variable" disturbs the artificial equilibrium created by countless HFT algos to give the impression of an deep, orderly market.

In the aftermath of this report, one can be sure that the days of current market structure are numbered, and that the scene is now set to throw the book at the HFTs. The only thing that is missing is the appropriate catalyst. And what is better than an orchestrated, or ad hoc, market crash, one which exonerates the real culprit for the stock market bubble - the Federal Reserve - and unleashes populist anger by millions of investors who lose their net worth in an HFT instant, aimed squarely at the HFTs, and the 20-year-old math PhDs behind them?

The core reason that velocity has continued to decline.

The core reason is the excessive amount of debt in the system, an amount that has continued to grow since the financial crisis in 2008 as highlighted in a McKinsey report which attracted a lot of publicity when published earlier this year (see McKinsey Global Institute report “Debt and (not much) deleveraging”, February 2015). For the record, McKinsey estimated that aggregate global debt has grown by US$57tn from US$142tn in 2007 to US$199tn at the end of 2Q14, raising the ratio of global debt to GDP by 17 percentage points to 286% (see Figure 15).

One month ago, when looking at the dramatic change in the market landscape when the first cracks in the central planning facade became evident and it appeared that central banks are in the process of rapidly losing credibility, and the faith of an entire generation of traders whose only trading strategy is to "BTFD", we presented a critical report by Citigroup's Matt King, who asked "has the world reached its credit limit" summarized the two biggest financial issues facing the world at this stage.

The first is that even as central banks have continued pumping record amount of liquidity in the market, the market's response has been increasingly shaky (in no small part due to the surge in the dollar and the resulting Emerging Market debt crisis), and in the case of Junk bonds, a downright disaster. As King summarized it "models linking QE to markets seem to have broken down."

Needless to say this was bad news for everyone hoping that just a little more QE is all that is needed to return to all time S&P500 highs. And while this concern has faded somewhat in the past few weeks as the most violent short squeeze in history has lifted the market almost back to record highs even as Q3 earnings season is turning out just as bad, if not worse, as most had predicted, nothing has fundamentally changed and the fears over EM reserve drawdown will shortly re-emerge, once the punditry reads between the latest Chinese money creation and capital outflow lines.

The second, and far greater problem, facing the world is precisely what the Fed and its central bank peers have been fighting all along: too much global debt accumulating an ever faster pace, while global growth is stagnant and in fact declining.

King's take: "there has been plenty of credit, just not much growth."

Our take: we have - long ago - crossed the Rubicon where incremental debt results in incremental growth, and are currently in an unprecedented place where economic textbooks no longer work, and where incremental debt leads to a drop in global growth. Much more than ZIRP, NIRP, QE, or Helicopter money, this is the true singularity,because absent wholesale debt destruction - either through default or hyperinflation - the world is doomed to, first, a recession and then a depression the likes of which have never been seen. By buying assets and by keeping the VIX suppressed (for a phenomenal read on this topic we recommend Artemis Capital's "Volatility and the Allegory of the Prisoner’s Dilemma"), central banks are only delaying the inevitable.

The bottom line is clear: at the macro level, the world is now tapped out, and there are virtually no pockets for credit creation left at the consolidated level, between household, corporate, financial and government debt.

What about at the micro level, because while the world has clearly hit its debt-saturation point, corporations - at least the highly rated ones - seem to have no problems with accessing debt markets and raising capital, even if the biggest use of proceeds is stock buybacks, thereby creating a vicious, Munchausenesque close loop scheme, in which the rising stock prices courtesy of more debt, is giving debt investors the impression that the company is far healthier than it actually is precisely because it has more, not less, debt!

The reality, as we first showed in January of 2014, is that for all the talk of "fortress" balance sheets, and record cash buffers, the debt build up among US corporations has more than surpassed the increase in cash. In fact, as of early 2014, total debt was 35% higher than its prior peak, as was net debt.

Therefore, to us, the answer whether debt markets are once again approaching (or have crossed into) full capacity was clear; just look at what happened to IBM when, as we predicted, it bought back so much stock its investment grade rating was put in jeopardy and the company has seen its stock languish ever since unable to lever up any more just to repurchase its own stock.

Others, of the "more serious people" variety, have finally caught up, and as UBS' Matt Mish asks in a note late last week, "Releveraging: are debt markets approaching full capacity?"

His take:

In our latest strategy piece we concluded that, even in a stressed scenario, US and European high grade issuance could decline from peak levels yet overall activity should remain quite resilient. Well, that thesis could be tested in the coming months following a rash of (large) M&A announcements, including AB InBev's $106bn proposed acquisition of SABMiller, Dell's planned takeover of EMC, Sandisk's reported attempts to find a suitor and Analog Devices indicated to be in talks with Maxim. The phenomenon is straightforward, and one we have been touting for some time: firms are increasingly releveraging balance sheets as earnings languish. Wal-Mart is perhaps the latest example, issuing disappointing profit guidance as it seeks to spend significant sums on labor and the internet in an effort to reignite sales growth (and authorizing $20bn in share buybacks to boot).

It should be clear to most what this means, but since "most" haven't seen a rate hike in their Wall Street careers, here is UBS' summary "This is textbook later stages of the credit cycle."

* * *

Having seen the light, Mish asks why what is now so obvious to him, is so confusing to everyone else:

What we find interesting is that most issuers and equity investors do not consider the prospect that debt markets could be reaching a point of full capacity – at least not in the near term. There are two root causes of this belief, in our view. First, neither has a strong appreciation of the divergences between debt and equity market universes. First, equity investors typically focus on large cap benchmarks (e.g., S&P 500) – of which most of the market capitalization lies in the top 100 firms – and generally see strong balance sheets with low net leverage, many of which are rated single and double A. However, that is not what credit investors view in their own universe. By definition, while equity indices weighted by market capitalization have been biased towards higher quality companies which have low debt and high cash balances, debt indices weighted by debt outstanding have been skewed towards those issuers raising more debt and generally levering up (Figure 1).

In the US, this first occurred in US leveraged loans (and to a lesser extent high yield) – driven primarily by financially savvy private equity owners; now it is manifesting itself in high grade as strategics lever up balance sheets to juice earnings in an environment where hiking dividends (further), buying back (more) stock, and spending on capex (particularly overseas) appears to have diminishing marginal returns. Second, this cohort perceives low rates as a key stabilizer for financing costs. As we argued last week, low Treasury yields are a key source of support for high grade bond yields. In recent months, even as IG credit spreads have widened, government bond yield declines have helped soften the overall impact on funding costs. For high yield yields, however, the major component is credit spreads, so low Treasury yields can only do so much.

* * *

And releveraging and the underlying dynamics are not occurring in a US vacuum. In our opinion, European issuers and equity investors also do not fully appreciate the divergences in fundamentals between equity and debt markets. Our analysis shows median net leverage has been rising for European IG and HY companies for several years, while trends in median leverage for Eurostoxx 50 issuers have been more stable (until 2014, Figure 2).

Late last and earlier this year European credit investors are increasingly seeing US issuers selling Euro-denominated IG debt to fund M&A as well as viewing domestic issuers releveraging balance sheets (e.g., in technology, healthcare, consumer staples and telecom, Figure 3). And the general direction appears to be similar, whether we look at high grade or high yield (Figure 4).

For those who missed our preview of all of this from April 2012 "How The Fed's Visible Hand Is Forcing Corporate Cash Mismanagement", here is UBS' far simpler summary which even 17-year-old hedge fund managers should get:

Here lies the problem. The predestined outcome is essentially a standoff between equity and debt investors where the former will continue to pressure the latter until credit spreads widen enough to cause capital market access to contract, stemming the deal flow. In high yield, the likelihood of reaching a breaking point is greater – we have seen instances where this has already occurred and equity investors could be complacent in this respect. However, in high grade, we reiterate that the markets are somewhat bulletproof. But the stakes are rising with each record deal. Near term, credit investors in aggregate will likely continue to hold their noses and absorb the releveraging until it becomes very extreme, though extracting wider spreads in the process. Unlikely, yes, in the next quarter or two; however, even in high grade we cannot envision this type of punishment lasting for a couple more years.

And that is the real countdown, because while the Fed may or may not have any credibility left, the only thing that matters is what is left of the once proud "bond vigilantes", virtually all of whom have been euthanized by the Fed's steamrolling of every last fundamental tenet of the market held dear by the bond trades and analysts of the world. According to UBS this, too, is now coming to an end, and even in IG the relentless issuance of one record debt deal after another, will soon hit a brick wall. That, coupled with the peak debt at the macro level described on top, will be the catalyst for the next phase in the evolution of centrally-planned capital "markets", whatever it may be.

Excerpted from Artemis Capital Management letter to investors,

Global central banking has artificially incentivized bets on mean reversion resulting in tremendous demand to short volatility.  The growth of short volatility exchange traded products (“ETPs”) since 2012 is nothing short of extraordinary and at the end of August, total short volatility assets exceeded long for only the second time in history. The rise of this short complex is intrinsically linked to the recent schizophrenic behavior of the VIX and adds significant shadow convexity to markets.

Velocity Shares Daily Inverse VIX (“XIV”) is the largest of these short VIX ETPs and has a cult-like following among day traders. Although the product has gained +111% since 2012, when decomposed on a risk-adjusted basis, it basically resembles a 3x levered position in the S&P 500 index with more risk. As the short and leveraged volatility complex becomes more dominant it is contributing to dangerous self-reinforcing feedback loops with unknowable consequences.

Many retail investors simply do not understand that short and leveraged volatility ETPs rebalance non-linearly (see below). To the casual observer it may appear that short and long assets counterbalance one another but this is not the case. For example if the first two VIX futures move 20% higher the short volatility ETP providers must buy an estimated 33% more volatility (vs. 25% for long) to balance that exposure. The first rule of derivatives hedging is that you never hedge a non-linear risk with a linear tool.  The mismatch means a large move in spot-volatility in either direction requires excessive buying or selling pressure whenever short volatility assets are dominant. Therein lies the problem. Falling volatility begets falling volatility and rising volatility begets rising volatility.

The great unknown is that this massive short volatility animal that appears tame given a regular diet of central bank liquidity may turn wild when that liquidity is removed. The wrong ‘risk-off’ event may expose a hidden liquidity gap in the short VIX complex that could unleash a monster. Artemis has attempted to quantify this theoretical liquidity gap by gauging the percentage of VIX open interest and volume required by exchange-traded products for rebalancing.

During recent market stress points such as October 2014 and August 2015 the short and leveraged volatility ETP complex required upward of 40-50% of the total liquidity of VIX futures as measured by average trading volume and open interest. Consider that the largest one day VIX move in history was the +64% jump that occurred on February 27, 2007 when the VIX went from 11.15 to 18.31. This was not even a period of high financial stress! If a similar volatility spike occurred today, given the current size of the short VIX complex, the ETPs by themselves would require an estimated 95% of the liquidity for rebalancing!

This would drive the price of the VIX futures up further exacerbating the nonlinearity. The VIX futures market may struggle to absorb the demand for long volatility. Dealers seeking to plug the liquidity gap would purchase S&P 500 options and forward variance swaps. The excess buying pressure exerted from the short-volatility complex would then push spot-VIX higher contributing to panic selling in the underlying S&P 500 index and a vicious and self-reinforcing cycle of fear followed by horror.

The recent bi-polar behavior in spot-VIX empirically supports the theory that a structural weakness now exists in this market by crowding of short volatility players. The shot across the bow for the short volatility complex came during the August 24th correction when SPX futures opened limit down and the CBOE struggled for 30 minutes to calculate the VIX. By the time the VIX level was finally calculated it opened 25 points higher at 53.29, before falling to 28 intra-day, then rebounding to 40.74 by the close, with the S&P 500 index down -3.9%.

At the time of the crash, the assets in long VIX ETPs outnumbered shorts on a two to one basis however, the complex still required an estimated 25% to 46% of market liquidity between August 21 and 24th.  Markets delivered historic volatility-of-volatility despite relatively mild historical declines in the S&P 500 index.  It is important to understand that markets have experienced much more dramatic oneday losses across history than what occurred in August 2015. For example on August 8th 2011, the market suffered a oneday decline of -6.7%. September to December 2008 experienced ten declines of more than -5%, and on Black Monday 1987, the market fell an incredible -20.5% in one day. During the Black Monday 1987 crash implied volatility in the S&P 100 index more than tripled going from 36.37 to 150.19.

If the VIX experienced any of these historic moves at current levels of short convexity the entire $2bn+ short volatility ETP complex would likely be wiped out overnight.

Short volatility sellers ridicule the fact that the prospectus for the iPath Long Volatility ST Index (VXX) clearly states that the ETF has an expected long-term return of zero. They should ask themselves, is it better to know with certainty you are going to go bankrupt slowly, or be completely ignorant of the fact you will go bankrupt suddenly. 

On the basis of the fundamental economic backdrop, Goldman Sachs sees VIX fair-value at least 19, with low-teens more consistent with ISM in the upper 50s (not the current sub-50 levels). However, credit (and FX) protection markets imply significantly more risk ahead with CDX HY stalled at 2month lows (while VIX hits 3 month lows). Given historical relationships, credit markets suggest a VIX of 25 is more consistent with reality(especially as Skew tail risk rolls down to normal risk).

Credit Risk has not normalized...

And while FX and Oil volatility has fallen, it has not fallen as much as equity volatility...

Leaving VIX a clear outlier.

As we noted previously, The U.S. economic landscape & cross asset metrics both suggest a VIX in the high-teens.

Point 1. The business cycle: In last week’s edition of The Buzz we estimated that baseline VIX levels of 18 would be justified given the current U.S. economic landscape. Our point is not that the VIX will go to 18 tomorrow. It is that trend VIX levels should now be 4-5 points higher than the average level of 14 experienced in 2013-2014 given the current state of the economy.

In the same way that we have argued that VIX levels in the high 20’s and above are consistent with recessions (and therefore the VIX should have dropped over the past couple of weeks) we also argue that VIX levels in the high teens are more consistent with the current ISM level in the low 50’s. VIX levels in the low teens are more consistent with ISM levels in the upper 50’s.

And the impact of rate hikes on volatility has historically been modestly higher

Charts: Bloomberg and Goldman Sachs