"QE's Are... Cake" - The Full Walkthru How Bond Traders Manipulate Daily POMO
It has gotten to the point where even we are amused how every "conspiracy theory" we suggest becomes proven fact in the span of a few years, usually involving criminality on behalf of at least one (usually more) perpetrator.
In today's most recent instance of "conspiracy theory" becomes "non-conspiracy" fact, we have the not so curious case of one Mark Stevenson, a former bond trader at Credit Suisse Group, who was fined 662,700 pounds ($1.1 million) and banned from working in the finance industry for "deliberately" manipulating a U.K. government bond price. Specifically, what Stevenson did, was to take advantage of the BOE's POMO auctions in 2011 to sell some 1.2 billion-pounds of existing gilt holdings which he had accumulated previously with the express intention of selling back to the BOE during a POMO operation at an artificially high price. All this was revealed as fact earlier today by the U.K. Financial Conduct Authority, the markets regulator.
How shocking. Who could have possibly predicted that banks would abuse POMO - that noble operation meant simply to provide banks with a means to sell paper without making abnormal profits on the back of the taxpayer (because all that debt that is being issued and monetized simply means that either there will be far less growth in the future, or inflation is set to surge - no other options) - to simply enrich themselves and a few select criminal traders.
Zero Hedge, that's who.
It was on January 11, 2011, months before Stevenson was engaged in this particular crime (in which he certainly is not alone in and where every single govvy and MBS trading desk has engaged in comparable price manipulation courtesy of POMO, with an intent and a means to sell at the highest possible price to Fed or any other central bank), when we explained precisely this:
While commenting on yesterday's NYT joke of a profile of the New York Fed POMO group, we openly mocked the claim by one Mr. Frost who said that when monetizing debt "We are looking to get the best price we can for the taxpayer.” We politely suggested that this is a blatant, tendentious lie, and that in fact the New York Fed merely cares to gift the Primary Dealers with any price it can for their bonds just so it stays on their good side (think Primary Dealer Auction take down over 50%), and after all - it is only money that according to Steve Liesman appears out of thin air.
And... three years later we were shocked to learn we were right. Because the Stevenson case, described in clear, informative language, explains precisely how it is that Dealers took advantage of the BOE (and by extension continue to take advantage of the Fed's) POMO, where the central bank buys back bonds from the dealer community, with zero regard for its cost-basis during the POMO reverse auction segment, to generate abnormal profits for these same dealers.
What follows is a somewhat technical narrative released earlier today by the UK's Financial Conduct Authority, which explains precisely this. We recommend readers read it to understand precisely that the purpose of POMO is, in addition to loading up banks with reserves which can and in the case of JPM were used to corner risk markets as initial margin collatereal, is to provide bank traders with riskless profits. We certainly hope US regulators (yes, we are laughing too) read it as well, to understand just how US traders abuse the Fed's own POMO on a day to day basis. We know they won't because their rightful owners, the banks, will never allow this generous daily piggy bank spigot to ever be shut down.
Events from June 2011 to 9 October 2011
Mr Stevenson’s position in the Bond was first established on 14 June 2011. This was a purchase of £50 million from a sovereign wealth fund, an existing client of CSSEL. This was a typical client trade by CSSEL and one which was consistent with its obligation as a GEMM to provide market liquidity through client trading. Mr Stevenson dealt with the trade for CSSEL.
Two further acquisitions of the Bond (totalling £100 million) were made by CSSEL on 16 June 2011. These purchases were made by the GEMM side of the desk and were transferred to Mr Stevenson’s proprietary book shortly after the acquisition.
1 July – 5 October 2011
Between 1 July and 5 October 2011, Mr Stevenson acquired further amounts of the Bond. Mr Stevenson had formed a view that the Bond was undervalued relative to other gilts and that it would be a good trading strategy to accumulate the Bond. Mr Stevenson believed there was considerable potential for the Bond to appreciate in value should further QE be introduced by the BOE. The first phase of QE had been halted in 2010 but Mr Stevenson agreed with some market commentators who felt that further QE was more likely than not.
On 29 September 2011, Mr Stevenson had a telephone conversation with Trader A, an acquaintance and employee of another GEMM. During the conversation, the following exchange took place in the context of QE:
STEVENSON: I mean I’ll give you my sort of run down. So we’re sort of thinking where the cusp could come, so obviously it could be 3 years and out but we’re sort of fancy longs because 5 years and out is possible.
TRADER A: Yes.
STEVENSON: In which case 8 3 17 [i.e. the Bond] I mean they look off the dial-y cheap to me and who’s … going to stop you putting them up.
TRADER A: Yeah yeah.
STEVENSON: I mean they’re the same err you know same swap thing as 8 21 or are they were until 2 days ago until the basis went a bit bid.
TRADER A: Yes.
STEVENSON: And who’s … going to stop them going 20 basis richer.
TRADER A: Yeah yeah.
STEVENSON: If [the BOE says its] buying them against the DMO’s screen.
The “cusp” refers to the maturity bucket starting point for QE reverse auction eligibility – that is, whether gilts with a residual maturity of 3 years or 5 years will be the eligibility starting point for the QE program. “Buy[ing] them against the DMO screen” refers to the process the BOE uses to buy gilts offered to it during QE – that is, ranking the value of offers relative to the DMO mid-price for the relevant gilt. “Putting them up” refers to the Bond increasing in value. “Going 20 basis points richer” means decreasing the yield of the Bond by 20 basis points with the price of the Bond increasing or becoming “richer” by an equivalent amount.
Between 28 September and 5 October 2011, Mr Stevenson did not purchase the Bond. The Bond did not trade at all in the IDB market between 30 September and 5 October 2011. There was a significant level of trading in the Comparator Bonds in the period between 30 September and 5 October 2011 (£648,230,000 of UKT 1.75% 2017, £504,850,000 of UKT 5% 2018 and £581,300,000 of UKT 4% 2016 were traded in the IDB market), illustrating the Bond’s relative illiquidity. Mr Stevenson did not buy more of the Bond because QE2 had not been announced and there was no certainty that it would be. Mr Stevenson’s described his strategy as “…a trade idea for an event which might or might not occur.” QE was therefore a key factor in his strategy to buy more of the Bond.
6 October 2011
At midday on Thursday 6 October 2011, the BOE announced that QE would be reintroduced on 10 October 2011. The market notice released by the BOE confirmed that the Bond was potentially eligible for inclusion in the QE2 operation on 10 October 2011. At 16:00 on 6 October 2011, the BOE issued a further market notice, which stated that certain gilts would be excluded where the BOE owned more than 70% of the free float (free float information is freely available to the market), consistent with the approach taken in QE1. The BOE did not own more than 70% of the Bond meaning that it was not excluded from, and was eligible for, QE2.
Mr Stevenson purchased an additional £31 million of the Bond through the IDB market on 6 October 2011 through switch trades with one of the Comparator Bonds, UKT 1.75% 2017.
7 October 2011
On Friday 7 October 2011, £29.3 million of the Bond was traded by other market participants through the IDB market. Mr Stevenson did not acquire any more of the Bond, although he did actively trade gilts on 7 October 2011.
During the day, Mr Stevenson took part in a telephone conversation with Broker A, a friend of Mr Stevenson and an agency broker. During the telephone conversation with Broker A, Mr Stevenson said “And we’ve been loading up with QE trades for months”, and “QE’s are … cake…”. The Authority has concluded that Mr Stevenson was indicating his belief that QE was an opportunity to profit from selling gilts to the BOE.
Mr Stevenson received a telephone call on 7 October 2011 from Risk Management at CSSEL, who were calling in respect of a profit and loss spike on a book jointly held by Mr Stevenson and another trader. Mr Stevenson explained on the telephone to the Risk Management officer that:
“… I mean the MPC, the Bank of England has announced you know they’re extending their asset purchase scheme yesterday, and you know we’re preparing to… well we’re buying some assets to sell to them, I mean basically that’s what’s going on.”
In other words, Mr Stevenson stated that he had been acquiring assets and his intention was to buy more, for the express purpose of selling them onto the BOE as part of QE2.
10 October 2011
Monday 10 October 2011 was the first day of QE2. Between 14:15 and 14:45 GEMMs could submit offers to sell gilts to the BOE through its B-Tender electronic platform. Buying gilts on QE days can be more difficult than non-QE days because sellers of eligible gilts are fully aware that a large buyer (that is, the BOE) will enter the market at 14:15.
At 07:34, Mr Stevenson made enquiries to try to find out if CSSEL’s client from whom the initial acquisition of the Bond had taken place in June 2011 had any further amounts of the Bond for sale. No indication was received that this was the case.
At 08:49, Mr Stevenson contacted a broker at IDB A and asked for the yield spread between the Bond and UKT 1.75% 2017. Mr Stevenson also told the broker at IDB A that he was a buyer of the Bond.
Between 09:00 and 14:30, Mr Stevenson bought a total of £331.1 million of the Bond through the IDB market. This was 2,700% of the Bond’s average daily trading value for the period from June to October 2011 and represented 92% of the purchases of the Bond in the IDB market that day. He purchased the Bond outright and through switch trades as follows:
(a) £57 million by buying the Bond outright at progressively higher prices;
(b) £254.1 million by buying the Bond and selling UKT 5% 2018 at an increasingly more expensive spread (that is, he paid increasingly more to acquire the Bond whilst selling UKT 5% 2018);
(c) £20 million by buying the Bond and selling UKT 1.75% 2017 at an increasingly more expensive spread (that is, he paid increasingly more to acquire the Bond whilst selling UKT 1.75% 2017).
The table at Appendix 1 sets out details of Mr Stevenson’s trading in the Bond on 10 October 2011. Mr Stevenson had never previously traded the Bond in such significant volumes in a single day.
At 09:07, Mr Stevenson received a telephone call from Trader A, who was calling to speak to one of Mr Stevenson’s work colleagues. Mr Stevenson and Trader A initially discussed trading generally and then discussed trading with respect to QE. Towards the end of the conversation, the following exchange took place:
STEVENSON: Well we’re thinking about sort of doing something today but erm…
TRADER A: I think you should just offer way up.
STEVENSON: Well, it depends if you get it way up before, or offer way up.
TRADER A: Yeah [laughs].
STEVENSON: [Laughs]. If you see what I mean.
The Authority has concluded that:
(a) “Doing something today” means making an offer into the QE reverse auction.
(b) “Offer way up” means offering gilts to the BOE at a high price.
(c) “Get it way up before” means moving the price of a gilt higher ahead of the reverse auction.
Buying the Bond outright
Mr Stevenson bought £57 million of the Bond outright in a series of 7 trades between 09:44 and 13:58 as set out in the table at Appendix 1. The increasing price at which Mr Stevenson was prepared to acquire the Bond on an outright basis can be contrasted with the gilt futures price, described by brokers at the time as trading at “a new low”. It can also be contrasted with the increasing yield (and decreasing price) of the Comparator Bonds on 10 October 2011 as set out in Chart 1.
Buying the Bond through switch trades
Trading the Bond vs UKT 5% 2018
Mr Stevenson bought £254.1 million of the Bond on a switch basis against the UKT 5% 2018. These purchases took place in a series of 12 transactions between 10:26 and 14:27 and occurred at a yield spread of between -22 and -28.5 basis points. The most expensive purchase of the Bond by Mr Stevenson at -28.5 basis points at 14:01 and 14:13 represented a richening in the yield spread of the Bond against the UKT 5% 2018 of 10.5 basis points compared to his initial bid of -18 at 09:47.
Through this period of trading, Mr Stevenson consistently improved the level at which he was bidding for the Bond. The effect of this was to increase the price of the Bond relative to that of UKT 5% 2018 (and the other Comparator Bonds).
There was a pattern to Mr Stevenson’s trading whereby he bought and then increased his bid within a short time frame. For example:
(a) At 12:17 he lifted an offer at -25 and he then immediately bid at -25.5.
(b) At 12:20 his bid at -25.5 was hit and he then immediately bid at -26.
(c) Between 12:23 and 12:31 his bid at -26 was hit four times and at 12.42 he bid at -26.5.
(d) At 12:51 he lifted an offer at -26.75 and then immediately bid at -27.
This style of trading, buying at one level and immediately rebidding at a higher level is known amongst traders as “facing”. It is called this because the trading is “in your face” and is unusual in circumstances where the wider market is not moving in the same direction, because it is usually unnecessary to trade very aggressively if buying in a market where prices are generally falling. On 10 October 2011, the wider gilt market was moving in the opposite direction to the Bond and losing value and therefore the aggressive trading undertaken by Mr Stevenson was unlikely to have been necessary.
Mr Stevenson was the only market participant who provided bids in this switch to IDB B on this day and he will have been aware of the absence of other bidders throughout the day through his interaction with the IDBs.
As the day progressed and the level of the Bond’s outperformance increased, it became apparent that some sellers were beginning to enter the market. For example, the broker at IDB B noted to Mr Stevenson that the seller with whom he traded at 12:20 was a different counterparty from the one who sold to him at 12:17. Despite this, Mr Stevenson continued to offer to pay increasingly more for the Bond.
Between 12:23 and 12:31 Mr Stevenson’s bid at -26 was hit by four different counterparties resulting in Mr Stevenson buying £58.4 million of the Bond. Mr Stevenson continued to bid at -26 and then improved his bid at 12:42 to -26.5.
From around 13:45, additional sellers began to enter the market and a number of offers began to be made by other participants. Mr Stevenson’s bids were hit 3 times at -28.5 (at 13:45, 14:01 and 14:13) and his bid at -28 was also hit (at 14:16).
Mr Stevenson’s activity in the IDB market after the BOE reverse auction had commenced at 14:15 is particularly significant and the continued activity in the market informed the DMO mid-price which in turn informed the price of offers made to the BOE. In the 15 minutes between 14:15 and 14:30, Mr Stevenson purchased £56.7 million of the Bond on a switch basis against the UKT 5% 2018.
Trading the Bond vs UKT 1.75% 2017
At 09:00, a broker from IDB A noted in response to a request from Mr Stevenson that there had been a move in the yield spread between the Bond and the UKT 1.75% 2017 (from +10 to +9.5) which reflected a broader movement in the market, that is, a flattening in the yield curve.
As the morning progressed, the Bond outperformed the UKT 1.75% 2017 such that the relationship between the two yields changed significantly. From approximately midday, the yield of the Bond became lower than that of the UKT 1.75% 2017. Thereafter, Mr Stevenson was prepared to buy the Bond against selling the UKT 1.75% 2017 and drop a negative yield rather than pick-up a positive yield, meaning that Mr Stevenson was effectively paying for the trade rather than receiving payment as he would have done earlier in the day.
Between 14:23 and 14:30, Mr Stevenson bought £20 million of the Bond on a switch basis against the UKT 1.75% 2017. These purchases took place in a series of 4 individual purchases and occurred at a yield spread of between -2 and -3 basis points. The average yield spread at which Mr Stevenson bought the Bond through this switch was -2.63 which represented a richening in the yield spread of the Bond against the UKT 1.75% 2017 of 11.63 basis points in comparison to the initial bid of +9 that he placed at 09:01. The most expensive purchase of the Bond by Mr Stevenson at -3 basis points represented a richening in the yield spread between the Bond and the UKT 1.75% 2017 of 12 basis points from the initial bid.
The impact of Mr Stevenson’s trading
The three charts below demonstrate the significant impact Mr Stevenson’s trading had on the Bond’s yield and yield spread relative to other comparable bonds. For example:
(a) Chart 1 shows the Bond outperforming all Comparator Bonds on 10 October 2011. From around 09:20, the Bond’s yield decreases (reflecting a price increase) as the yield of all other Comparator Bonds increase (reflecting a price decrease). The yield spread between the Bond and the Comparator Bonds is at its widest at around 14:30. Following this, Mr Stevenson leaves the IDB market and ceases bidding for or trading the Bond. By approximately 15:30, the Bond’s yield movement had fallen in line with each of the other Comparator Bonds, completely reversing its earlier outperformance.
(b) Chart 2 shows the movement in the yield of the Bond (UKT 8 3 17) and the UKT 1.75% 2017 (UKT 1 75 17) and the movement of the spread between the Bond and UKT 1.75% 2017 (17/17s Yield Spread) on 10 October 2011. The yield spread moves from approximately -9 to +3 between the times Mr Stevenson placed his first bid (09:01) and when he executed his last trade (14:30).
(c) Similarly, Chart 3 shows the movement in the yield of the Bond (UKT 8 3 17) and the UKT 5% 2018 (UKT 5 18) and the movement of the spread between them on 10 October 2011. The yield spread can be seen dropping throughout the day until around 14:30, at which time it begins to increase, until approximately 15:30 when it stabilises.
Offers submitted by Mr Stevenson to the Bank of England
The QE2 reverse auction ran from 14:15 to 14:45 on 10 October 2011. During this period, the GEMMs were able to submit offers to sell gilts through the BOE’s B-Tender system during the auction period. Mr Stevenson was responsible for placing the offers in the Bond on behalf of CSSEL to the BOE.
Mr Stevenson submitted a number of offers through the B-Tender system during this time. Each offer superseded the previous one and only offers in the system as at 14:45 could be accepted by the BOE. As shown in the table below, the quantity of the Bond offered by Mr Stevenson increased from £800 million to £850 million and the price at which Mr Stevenson offered the Bond decreased during the auction.
Bond offers submitted by Mr Stevenson to the BOE on 10 October 2011
The size of Mr Stevenson’s offer of the Bond into the QE2 reverse auction on 10 October 2011 was larger than all of the other offers in the Bond on 10 October 2011 combined and represented two-thirds of the value of the Bond offered to the BOE on 10 October 2011 and the total consideration payable by the BOE to CSSEL (had the offer been accepted by the BOE) would have been 70% of the amount (£1.7 billion) it had allocated to the reverse auction on 10 October 2011.
Events following submission of the offer
At 14:56, the BOE announced the “Asset Purchase Facility gilt purchase operation results” which included the following statement:
“The Bank has decided to reject all offers against UKT_8.75_250817 following significant changes in its yield in the run up to the auction."
The Bank therefore decided to reject all offers in the Bond – the only time that the BOE has ever taken such a step.
Shortly afterwards, the BOE contacted CSSEL. In a telephone conversation between senior representatives of the BOE and CSSEL, the BOE explained that it believed that the Bond had been traded in such a way so as to increase its price in order to sell it to the BOE at a distorted level, and that someone would contact the relevant trading desk at CSSEL to further investigate.
The BOE was concerned that the Bond was being offered to it at an inflated level, such that there would have been an additional, unacceptable cost to the taxpayer in buying at that level. The BOE therefore rejected all offers in the Bond.
The market’s view of the trading
Several gilt market participants commented on the Bond’s significant outperformance on 10 October 2011.
By 09:39 (38 minutes after Mr Stevenson began trading in the Bond), a market participant had telephoned the BOE regarding the Bond’s outperformance. Several other market participants telephoned the BOE throughout the day, suggesting that the Bond had been “squeezed”, “rammed”, and that someone “was messing around with” it. One market participant said he could see no reason for the significant trading in the Bond on 10 October 2011 other than deliberately pushing the price higher in order to sell to the BOE later in the day.
Some market participants declined to trade the Bond with clients on 10 October 2011, with one indicating that he would be prepared to do so only at the previous trading day’s closing price in view of the unusual price movements. Other market participants commented on the scale of the outperformance.
Mr Stevenson’s explanation for the trading
Mr Stevenson has stated that he bought the Bond on 10 October 2011 because he believed it was cheap and not with the definite intention of offering to sell it to the BOE later that day. He stated that he traded the Bond throughout the day on 10 October 2011 in an open and transparent manner. He said that he continued to purchase the Bond until it reached what he believed was its fair value. He said that once the Bond reached what he believed was its fair value, he was entitled to offer it into the reverse auction.
Mr Stevenson said that one reason for offering the Bond into the QE reverse auction on 10 October 2011 was in order to rebalance the trading book. He said that the Bond was offered because it had reached what he believed was its fair value, but that he could have offered a number of other gilts instead.
The Authority’s findings about Mr Stevenson’s trading in the Bond
The Authority has reached the view that Mr Stevenson made the decision to purchase the Bond with a view to offering it to the BOE in the auction. He was aware that his trading in the Bond on 10 October 2011 would not only increase CSSEL’s holding in the Bond but also increase its price. This was done in order to increase the return for CSSEL if its bid in the auction was accepted by the BOE.
The Authority has concluded that Mr Stevenson had formed an intention to offer the Bond into the next round of QE before 10 October 2011. He had an opportunity to buy more of the Bond before 10 October 2011 (both before and after QE2 was announced on 6 October 2011) but instead chose to buy an extremely large quantity on 10 October 2011 (92% of the value traded in the Bond through the IDBs and 2,700% of the Bond’s 5 month daily average volume, the only time he has traded the Bond in this volume).
Mr Stevenson’s trading in the Bond on 10 October 2011 had a significant effect on the price of the Bond, something Mr Stevenson would have been aware of at the time yet he continued to buy the Bond aggressively throughout the day. The comments made during the day by other market participants indicate how remarkable the movements in the Bond’s price and yield were compared to the Comparator Bonds during the day. The Authority has concluded that Mr Stevenson’s trading was designed to move the price of the Bond on 10 October 2011 and that he was not simply trying to acquire more of a bond that he perceived as being cheap.
Having considered these factors, the Authority has concluded that Mr Stevenson’s conduct is at level 5 in terms of seriousness and therefore, the percentage of relevant income for step 2 is 40 %. This amounts to £946,800. As this amount is larger than £100,000, the step 2 figure is £946,800.
The Authority therefore imposes a total financial penalty of £662,700 on Mark Stevenson for market abuse.
The Authority considers that Mr Stevenson deliberately acted to increase the price of the Bond on 10 October 2011 and considers that this amounted to deliberate market abuse. His behaviour is particularly egregious as it took place on the first day of QE2 and effectively sought to deprive the economy from QE’s full effect in order to maximise the potential profit from selling the Bond to the BOE in the QE2 reverse auction. The Authority considers that, as a result of this behaviour, Mr Stevenson lacks fitness and propriety in terms of his integrity.
The Authority therefore makes a prohibition order pursuant to section 56 of the Act prohibiting Mr Stevenson from performing any function in relation to any regulated activity carried on by any authorised or exempt person or exempt professional firm. This order takes effect from 20 March 2014.
That's ok - we are confident Mr. Stevenson will live comfortably for the rest of his life away from trading, with the millions of bonuses he has collected during his tenure trading bonds in a fair and honest manner.
As for everyone else, now that you know how POMO manipulation is done, keep doing it.
- advertisements -