Spoofing Lessons From Andy Hall – The Oil AND Silver King
Posted by Vince Lanci on July 11, 2017 4:14 pm
Tags: Algorithmic trading, Andrew Hall, artificial intelligence, Backwardation, Bear Market, Business, Capital Markets, China, Commodity markets, Contango, Crude, Crude Oil, Day trading, Finance, Financial economics, Front Running, Futures contract, Futures markets, goldman sachs, Hedge, HFT, India, Iraq, Market Manipulation, MONEY, New York Mercantile Exchange, NYMEX, OECD, OPEC, Organization of Petroleum-Exporting Countries, Phibro, PIRA, Precious Metals, Reality, recovery, Reuters, Saudi Arabia, Testimony, Trading Systems, Transparency, Tyler Durden, Volatility
Categories: Algorithmic trading Andrew Hall Artificial Intelligence Backwardation Bear Market Business Capital Markets China Commodity markets Contango Crude Crude Oil Day trading Economy Finance Financial economics Front Running Futures contract Futures markets goldman sachs Hedge HFT India Iraq Market Manipulation money New York Mercantile Exchange NYMEX OECD OPEC Organization of Petroleum-Exporting Countries Phibro PIRA Precious Metals Reality recovery Reuters Saudi Arabia Testimony Trading Systems Transparency Tyler Durden Volatility
A Silver Legend Throws in the Towel on Oil
In 2010 I wrote an anonymous article for Zerohedge on Silver manipulation. Anonymous because the article was in part an indictment of the market structure on COMEX at the time. I was afraid of backlash against me and my nascent family, having already having seen the underside of a bus in 2003 via my own actions and the need for a conflicted, and unqualified compliance officer / bureaucrat Tom LaSala in danger of losing his job after a horrendous failure to protect the NYMEX electricity contract. This, according to NYMEX / CFTC sources then. But I digress.
About the author: Vince Lanci has 27 years’ experience trading Commodity Derivatives. Retired from active trading in 2008, Vince now manages personal investments through his Echobay entity. He advises natural resource firms on market risk. Over the years, his expertise and testimony have been requested in energy, precious metals, and derivative fraud cases. Lanci is known for his passion in identifying unfairness in market structure and uneven playing fields. He is a frequent contributor to Zerohedge and Marketslant on such topics. Vince contributes to Bloomberg and Reuters finance articles as well. He continues to lead the Soren K. Group of writers on Marketslant.
Hall WasThe Uncrowned King of Silver
For me, PhiBro was a mentor in how to not be a victim and to try to divine my opponents’ intentions just by watching their trades. An exercise in applied empathy if you will. To study PhiBro is to study Andy Hall.
Andrew Hall is a legend in the trading community. He was instrumental in execution of the 1994 Silver squeeze and the 1997 Buffet Silver buy. He was a manipulator of Silver to the upside. But metals were just his HOBBY. And when Hall had orderflow, he maximized returns for clients and the prop desk at PhiBro. The man knew how to front run! But without order flow..
Maybe the seat makes the money, not the man in it? But we aren’t here to kick him. Rather to describe what the man is good at. and to describe what we observed from him.
Andy Hall, Oil Perma-Bull
His baby was oil. We remember being on the wrong end of many mini oil plays by his desk at PhiBro right before a refinery fire became public. Here is one play his desk loved to do at least once a month :
- Hall Gets Long Oil > Broker buys Calls for PhiBro hard and sloppily > Oil rallies > Refinery fire news hits > Oil rallies more > Hall gets out of longs > Cue the crying options traders.
This was brilliant because he played the most liquid market against news in a much less liquid one. And a refinery fire is not necessarily bullish oil. It destroys oil demand as the refinery shuts down.
But when the option broker telegraphs who is buying, and the PhiBro reputation precedes him, and the option marketmakers rush like lemmings to buy futures to hedge their short deltas…. you get long!.
Hall Was Immune to Buffet’s House Cleaning
We also knew traders on his desk that got fired by Solly, PhiBro’s parent, after Buffet came in 1994 to rescue them. Hall did not. Because Hall was in energy, and while he had his own position in Silver on both occasions, he was not the poor sole who took the fall in 1994.
Why Hall Was Great
The reality is, it is not so easy to be right without client flow behind you. Hall’s early success in oil on his own may have been in part a function of a secular bull market in commodities to begin with. Personally, our experience has been that Hall was not a directional expert, but he knew when a market was lopsided and knew how to catalyze the exit problem for everyone else. He also was expert at creating exit strategies for his own massive positions.
Once I saw a broker in Silver futures laughing (in relief) and shaking his head after executing and seemingly butchering a 5,000 sell order. I knew this broker well and asked him: “Was the client upset?” His answer was ” NO, He was laughing!” To which i prodded him: “Was this the big player you have?”. His response was: NO, THIS WAS THE CLIENTS ENERGY DESK TRADER. And HE WAS LAUGHING AS I FILLED HIM 20 cents LOWER THAN I SHOULD HAVE.
Apparently this was Hall. And he had been long from $3.00 lower in Silver. To top that, a silver options broker had bought 1,000 calls loudly 30 minutes prior, driving uo the futures price about 20 cents. This was the same energy tactic Hall used so often. And a lesson was learned.
Taking on the Banks
Possibly least understood was his acumen in playing the oil futures term structure. We saw and were informed of his massive plays where calendar relationships were out of whack and he stepped in to fix them. He would be the buyer of 2 year December futures after a bank was done laying off producer hedges for his “back-to-back” vig. Then he’d sell another month in which he thought the price was out of whack on the high side. He exploited distortions created by organic order flow. Then he waited to be right. And sometimes, he nudged himself in being right as these were illiquid contracts. He would test the resilience of the sell side (maybe the bank actually held onto some of the hedge given them by the client?) by buying in thin hours to see if they pushed back. This is spoofing by the way. But he wasn’t necessarily fishing for stops like a slow motion algo. He was looking for sellers as he bought. And if they didn’t buy, he’d keep pushing.
I labelled this to all who would listen as an inverted pyramid style. It was the antithesis of investing. It went something like this:
- Already be long (when wasn’t he?)
- Buy 1 contract
- Buy 2 contracts if the fill on the first one was poor
- Buy 10 contracts even worse
- Buy 100 contracts even worse, then bid for 1000 at that price
- Buy some calls which will create option related futures buying
- Sit back and see what happens.
- If the market takes off, sell as many futures as you were long
- Use the calls as either a tail or convert them to synthetic puts
Bidding to Sell
I knew a precious metals floor broker who actually lost his business because of executing for PhiBro in this fashion. The broker was bidding and showing some ridiculous volume for one part of the PhiBro desk. And in between his announcing his bid he was selling to locals 10 cents under his bid. He could not cross the trades as they were for the same firm, but from different desks / clients. The broker was accused of facilitating market manipulation.
Crime scene depicted, but pales by comparison to HFT/ Algo crimes which are floor tactics on streroids with no counterparty transparency.
What really happened was he was selling for a Phibro trader or client long, while simultaneously bidding for another Phibro person. That broker was handing money to locals who ran from him, scared it was a trick. I saw this happen. And it was hilariously scary. I’m sure some of the details are not right here, and I have an alternate explanation of what may have been going on at the PhiBro desk, but it changes nothing. Phibro was bidding above where they were selling and neither side could get filled.
A Product of PhiBro Culture
To begin to try to understand Hall a bit one must understand the culture of PhiBro. That firm started in the least liquid products imaginable: iridium and such. To be a marketmaker in assets like these, one must have a brilliant tactical mind for creating your exit liquidity. Hall has that skill. One must also recognize the right time to corner a market. Hall did this. And one must know how to disguise one’s intentions in a small market where participants are easily identified. The PhiBro trained Hall did this.
These survival skills lent themselves greatly to manipulations of Silver on more than one occasion (crushing overhedged producers), front running refinery fires on many occasions, and in recognizing distorted futures curves from undigested order flow (and subsequently taking the banks who were order-flow monkeys on)
Read on and keep the above in mind when reading Hall’s letter.
Andy Hall’s Letter to Investors
as published in ZH
[emphasis by Tyler Durden]
July 3, 2017
The oil rout continued in June with prices entering bear market territory. Not only did sentiment plumb new depths but fundamentals appear to have materially worsened. Demand growth seems to be somewhat less than anticipated while supply keeps surprising to the upside. The expected acceleration in inventory drawdowns has not materialized – at least as evidenced by available high frequency data. Several weeks of lackluster inventory data from the EIA, along with reports of increasing amounts of oil in transit and in floating storage, disappointed expectations for accelerating stock draws following the arrival of peak seasonal demand.
Meanwhile, U.S. shale operators have continued to add rigs at a surprisingly fast rate thus raising the odds for significant oversupply in 2018, even if OPEC maintains its production cuts beyond Q1. Over the past month, the market has in effect priced in two negatives, one long-term, the other short-term.
The longer-term negative is that it is becoming increasingly evident that, under most reasonable scenarios, U.S. shale oil will be the marginal source of supply, at least until 2020. There are enough non+OPEC, non-shale production projects already in the pipeline (and which were sanctioned when prices were much higher than today) that incremental U.S. shale oil production alone can balance the market for the next two or three years. Moreover, and more importantly, it is now becoming apparent that the cost of this oil is significantly lower than was believed to be the case even a few months ago. That means the long-term price anchor for oil has moved lower. At the start of the year, the anchor was thought to be about $60 (Brent) and rising over time. Today, it appears to be closer to $50 (and possibly still falling). Prices for long-dated futures have therefore moved down to reflect this new perception.
The short-term negative is an apparent deterioration in the supply and demand balances for 2017. Until recently it looked like demand would exceed supply by as much as 1.5 million bpd if OPEC maintained its production cuts through 2017. This would have eliminated the global inventory surplus sometime in Q3 and resulted in a backwardated market. It now seems, however, that the supply deficit will be considerably smaller than originally expected – probably only around 0.5 million bpd. There will therefore still be sizeable excess stocks at the end of the year. This realization has resulted in the market moving into a steeper and uninterrupted contango with spot prices falling relative to deferred prices which, as just noted, have themselves ratcheted lower.
We discuss both these developments in more detail below. However, absent some geopolitically induced supply curtailment or a further cut by OPEC, oil prices are likely to be range bound around a level that limits the growth in shale oil production. That would mean the forward WTI strip ought to be somewhere below $50.
Shale is now the marginal barrel
Technological advances have continued to drive down well breakevens as well as expand the shale oil resource base in the U.S. In a recent report, PIRA estimated that there are now 80 billion barrels, or half of the recoverable U.S. shale oil resource base, that is economic at $50 Brent (say $48 WTI) or less. This represents some 215,000 well locations. Each of these on average can produce around 300 bpd in its first year on stream. The current horizontal oil rig count is 650 and has been growing at a rate that would bring the count to close to 800 by the end of the year. 800 rigs can drill about 15,000 wells per annum which means potentially 4.5 million bpd of gross new production. After deducting legacy decline this would translate into net production growth of more than a million bpd per annum, which exceeds the expected “call on shale” (demand growth less non-shale crude supply growth from non-OPEC, OPEC crude and other non-crude liquids). Today’s rig count or lower would be necessary to constrain shale oil growth to the 0.7- 0.8 million bpd of year/year growth in shale oil production that is probably required to balance supply and demand.
The market is therefore trying to find a price level that curtails rig additions (and/or well completion activity) to a level commensurate with the call on shale. Exactly what that price is can be debated and the truth is no one really knows. It depends on current and future rig productivity, how drilling and completion costs respond to rising oil field activity levels, the willingness of shale operators to outspend their cash flows and the availability and cost of capital to the industry.
Notwithstanding uncertainties surrounding all these variables, it does seem that the price needed for a given rate of growth in shale oil production has been falling over time. Well breakevens have dropped because of steep rig productivity gains and cyclical cost declines. They could fall further if continued secular gains in rig productivity outstrip the cyclical cost increases now resulting from higher oil field activity.
Over the past two years, average rig productivity in the U.S. Lower 48 states has grown by more than 20 percent per annum. In the Permian basin productivity grew by around 30 percent last year. These gains have been achieved through reduced drilling times from the use of pad drilling and increased well productivity from longer laterals, more intense fracking and higher proppant loadings.
Whilst the rate of rig productivity growth appears to now be moderating, it is unlikely to stop altogether. A recent Goldman Sachs analysis posits continued productivity growth for years ahead. This will be driven by higher rates of recovery of initial oil in place through the application of artificial intelligence and big data analytics. Goldman argues that this could eventually reduce breakevens to $45 and below. The best operators in the Permian like EOG already have well breakevens at, or even below, $40 WTI. As the rest of the pack catches up with the leaders, average breakevens are likely to fall further if Goldman is correct.
If the marginal cost of oil for the next 3 or 4 years really is headed to the mid-$40 range then OPEC’s attempts to push prices to $60 seem futile. It is unlikely that OPEC will find the cohesion necessary to keep prices at an artificially elevated level if all it does is accommodate rampant growth in shale oil production.
That line of thinking raises the possibility of yet another reversal in OPEC policy – abandoning supply management and letting market forces balance supply and demand. This would obviously result in significantly lower prices, at least in the short term. On the other hand, with many OPEC countries needing much higher prices for fiscal sustainability (for Saudi Arabia the level is over $80), there is an inherent instability which adds to the geopolitical risks to supply. But for now, it seems likely that OPEC, led by Saudi Arabia, will stay the course with its current policy of production restraint. Oil at $45-50 is preferable to it being at $40 or below, even if the loftier target of $60 has proven elusive.
With hindsight, OPEC’s attempts to manage supply were poorly conceived. Given the short response time of shale oil to changing prices, OPEC should have acted more quickly and more decisively. The production cuts should have been deeper and implemented immediately. As it was, OPEC talked up the market ahead of the actual production cuts thus helping to unleash a fresh wave of future shale oil production as emboldened operators upped their capex budgets and raised capital on the back of the higher prices. Additionally, OPEC ramped up production in Q4 2016 ahead of its mandated cuts, thus adding to the very stock excess they were hoping to eliminate. OPEC members also then inexplicably offset the impact of their cuts by drawing down their own inventories to maintain exports during Q1 2017. This made no sense given OPEC’s stated goal of reducing OECD inventories to their five-year average.
Fundamentals have deteriorated significantly
In implementing its production cuts at the start of the year, OPEC and its allies were aiming to eliminate the inventory excess. This would have allowed spot prices to rise relative to deferred ones, pushing the market into backwardation. A backwardated market would eliminate the “subsidy” shale operators have been realizing by selling forward to hedge production. This would therefore help curtail shale oil supply growth by removing this windfall hedging profit. But it clearly hasn’t happened. The spread between Dec 2017 and Dec 2018 futures contracts moved $3, from a $1 backwardation to a $2 contango, over the past month as it became increasingly likely that there would still be substantial excess inventories at the end of the year.
There are several explanations for why the expected supply deficit has not materialized.
- Firstly, demand growth has been somewhat disappointing. Based on indicators of economic activity, demand in 2017 should be growing by around 1.7 million bpd, if not more. Actual growth, however, seems to be closer to 1.4 to 1.5 million bpd for reasons that are not yet clear.
- Secondly, non-OPEC supply growth has been exceeding initial expectations – largely because of faster shale growth in the U.S. Forecast growth in non-OPEC supply for 2017 has been revised progressively higher by 0.3 million bpd. OPEC production is also now expected to be greater than seemed the case just a month ago because of the earlier than anticipated return of shut-in production in Libya and Nigeria. This will add around 0.2 million bpd of additional supply on average in 2017.
- Finally, revisions to data for 2016 now show a small flow surplus of 0.1 million bpd whereas previously there had been a small flow deficit.
Together these changes amount to a 0.9 million bpd deterioration in the supply and demand balance for 2017 and an initially expected supply shortfall for the year of 1.4 million bpd now looks like it will be closer to 0.5 million bpd. Because of lower SPR purchases in India and China, as well as stock reductions in the OPEC countries, the drop in observed commercial inventories will be even lower – perhaps as little as 0.3 million bpd. This is much less than the rate needed to mop up the stock surplus – some 450 million bbls at the start of 2017 – and the market will almost certainly enter 2018 with a still substantial inventory overhang.
Moreover, at the rate at which oil drilling rigs have been added in the U.S., non-OPEC production has been on a path to grow by as much as 2 million bpd in 2018. With demand growth of, say, 1.5 million bpd and a 2017 flow deficit of only 0.5 million bpd, and with higher year/year production from Libya and Nigeria, that would imply an annual average stock build next year, even if the current OPEC production cuts remained unchanged for the whole of 2018, something which is by no means a given. It is this specter of renewed stock builds in 2018 adding to still inflated inventories that has panicked the market and caused the forward curve to move into contango. This reversal of the time spreads, combined with the drop in deferred prices to match a lower perceived marginal cost, has resulted in nearby prices collapsing, even though seasonal factors are becoming their most favorable.
In short, OPEC, the market and oil bulls have run out of runway. There are just 10 weeks before fall turnarounds kick in and crude stocks in the U.S. start to build again. Excess crude inventories in the U.S. are around 80 million barrels, up sharply since the beginning of June, reversing the trajectory seen in April and May when sequential crude oil draws were rapidly eliminating excess crude oil inventories.
In the past month, however, excess crude stocks in the U.S. are back to the levels seen this time last year and there now appears to be little chance that they can be eliminated before the fall – especially if the rate of inventory change seen in the data for the past three weeks is maintained. Moreover, Q4 2017 will see an acceleration in U.S. oil production as the impact of higher rig counts is increasingly reflected in higher production.
The main culprit for the disappointing stock draws in the U.S. is a stubbornly elevated level of net imports. While imports from Saudi Arabia have finally turned lower, those from other OPEC producers (notably Iraq) have risen. Crude exports have also fallen in recent weeks, at least if the preliminary data are to be believed.
Backwardation was meant to take care of excessive shale production in 2018 and beyond by driving deferred prices to levels that would constrain its growth. But stocks have not fallen fast enough to sustain backwardation so the whole futures curve has downshifted instead.
When the facts change…
For all the above reasons, it looks increasingly like oil prices will be rangebound for some time to come. Hitherto, it had been our view that oil would trend higher as prices would need to rise to a level that would justify investment in more costly sources of supply than just the core areas of U.S. shale. However, not only has the core shale oil resource grown significantly – above all in the prolific Permian basin – but breakevens have dropped because of secular productivity gains outpacing cyclical cost increases, at least for now. Furthermore, there has been no shortage of capital to fuel the growth in shale oil production and this has allowed operators to significantly outspend their cash flows. The marginal economics of the typical shale oil producer have proven to be no impediment to the industry’s resilience. The breakevens referred to earlier are based on half-cycle economics. Full-cycle costs that cover land acquisition, infrastructure and overhead are probably almost $10 higher. But companies base their drilling decisions on half-cycle costs even if this leads them on the path to eventual bankruptcy (to which the shale oil industry is no stranger) so long as they have access to capital. It’s quite possible that shale oil production growth can only be reined in by the capital markets rationing the supply of funds as industry management seems to be more focused on growth than generating free cash flow or even paper profits [ZH: this is something we have been pounding the table on since 2014, most recently in mid-June].
It also appears that the cost of developing other supply sources, such as deep water offshore, has been falling dramatically making them competitive with shale in many cases. Because of these developments, the cost curve for oil has become much flatter. There is now an abundance of potential supply at around $50 Brent. Prices will tend to oscillate around this long-term price anchor in response to changing inventory levels as the market tries to determine the right price to satisfy the call on shale. With the current inventory surplus and what looks to be its slow dissipation, markets are also likely to stay in contango, barring some sort of supply shock.
These developments call for a more opportunistic approach to the oil market than hitherto. Whereas it once seemed positions could be held with an eye to a longer-term secular appreciation, that is no longer the case. Indeed, the evidence is now in plain sight. Over the past year, the front month WTI futures contract has moved by double digits in percentage terms 10 times within a $40 – $55 band. This volatility has been accentuated by large financial flows into and out of the market by non-traditional investors and algorithmic trading systems. Attempting to capture just a percentage of those moves makes more sense than trying to ride what has turned out to be a non-existent trend, especially when contango inflicts a negative roll return on investors. The extreme volatility within a rangebound environment also argues for a more tactical and conservative approach to portfolio management.
For now, the market is heavily oversold with a record speculative short position. Q3 should still see decent stock draws even if they are insufficient to eliminate excess stocks. Also, increases in the rig count appear to be ending and could decline in the coming weeks and months. It also remains to be seen how quickly the increased drilling activity will translate into a commensurate increase in the number of completed wells and whether cyclical cost pressures will accelerate or bottlenecks develop. Already, the number of drilled uncompleted wells (DUCs) has been rising quite rapidly. There is also a non-zero chance that OPEC might surprise the market with a further “shock and awe” production cut. Taken together these considerations mean there is a good chance for a price recovery from current levels. However, this recovery will be limited for the reasons set out earlier and because of the overhang of potential selling from oil producers who are significantly underhedged for 2018.
Andrew J. Hall
Chairman and CEO
Read more by Soren K.Group