Epstein Data Wiki

Stewart Oldfield

Properties2
tagsperson, banker, deutsche-bank, compliance, kyc, aml, private-banking, new-york, southern-financial-relationship
aliasesOldfield, Stewart, Stew Oldfield, Stewart Oldfield, CFA, CAIA, Steward Oldfield

48 min read

How he signs himself in these documents”Stewart Oldfield, CFA, CAIA / Director / Deutsche Bank Trust Company Americas / Deutsche Bank Wealth Management / 345 Park Avenue, New York, NY 10154”12
What the bank’s own records call himthe Customer Owner and the Relationship Manager of the “SOUTHERN FINANCIAL RELATIONSHIP” — the relationship under which Jeffrey Epstein, his entities and his trusts banked at Deutsche Bank345
His relationship-manager code82289. A produced Deutsche Bank database extract prints the code beside his name on a Southern Trust Company row; the bank’s automated “RM0082289 Daily Deposit Report” was distributed to the list address built from that code67
When he joined the bank, in his own accountOctober 2014, recruited into the private bank by Charles “Chip” Packard, whom he says he had known for twelve years8
Span of the record read for this page6 October 2014 to 26 July 2019, measured over a floor of 12,205 documents91011
How often he is the senderOn 2,298 of 9,803 documents in the name-form population the sender value carries a form of his name — 23.4% of that population, and 26.5% of the 8,672 documents in it that carry any sender value at all12
What he told a compliance officer who had asked him for an email the AML team wanted, 6 July 2018”As for Chip’s email, it predates me. And the other people on the email are no longer at DB. So I don’t see how I can get hold of the full email. But again, Jeffrey has been through a complete rep risk process here and has been signed off on repeatedly by management as we’ve done any KYC remediation or new account opening.”13
What Epstein was worth to his bookOf $3,576,567 in revenue credited to him for FY18, $703,616 was Southern Financial. By April 2019 the figure credited to his name was zero, against a net outflow the same sheet puts at $85.0m14
What this page does not sayThat he knew of, participated in or facilitated the conduct for which Epstein was charged. Nothing read for this page supports that and this page asserts nothing of the kind

Stewart Oldfield is a banker. From October 2014 until at least July 2019 he was a Director in Deutsche Bank’s wealth management business in New York, and for most of that period he was the officer inside the bank whose name sat in the “Customer Owner” and “Relationship Manager” fields of Jeffrey Epstein’s accounts.

The New York State Department of Financial Services found in July 2020 that “although the Bank properly classified Mr. Epstein as high-risk, the Bank failed to scrutinize the activity in the accounts for the kinds of activity that were obviously implicated by Mr. Epstein’s past”,15 and fined the bank $150,000,000.16 That order names no relationship manager. The release does: across roughly five years of the bank’s own internal correspondence, Oldfield is the person who answered when the bank’s anti-money-laundering and know-your-customer functions asked questions about the relationship, and the documents record what he answered.

This page makes no allegation of wrongdoing against him. He is a living adult, named in the clear on the sheets cited below and named here in full. He is not a survivor. No response by him to any of these documents appears in the release; none is reproduced here, because none was found.

The institutional story (the onboarding, the Approval Email, the 2015 reputational-risk committee, the consent order, the class action and the settlement) is told at Deutsche Bank and is not repeated here. This page is about one officer, what he was asked, and what he wrote back.

Two things to get right before reading further

First, the surname alone is not the population, and the bank’s own bank statements are where most of the hidden record is. The phrase "Stewart Oldfield" returns 9,606 documents and the bare token Oldfield returns 9,815. Both miss a large class: Deutsche Bank’s monthly account statements for Epstein’s entities print a contact line reading “For personal assistance call: Stewart Oldfield”, and on hundreds of those sheets the scanner has mangled the name in both halves at once: Steuart Oldficld, Stekvarl Oldheld, Sicsvart Oldficld, Siouan Oldfield, Steuart Chiliad, Steuart OW ield. The image on those sheets prints his name cleanly. The corruption is a property of the extraction, not of the document, and the damaged strings are finding aids rather than readings.1718

Second, his relationship-manager code reaches more than a thousand documents that carry no form of his name at all. The bank’s automated daily balance report for code 82289 lists the accounts on his book (Epstein’s own account, Southern Financial, Southern Trust, the Butterfly Trust, Gratitude America, Neptune, Hyperion Air, LSJE, Plan D, NES, JEGE, Mort, Prytanee, The Haze Trust, Darren K. Indyke PLLC and others), and the report itself never names the banker. In its prefixed form RM0082289 the code reaches 1,134 documents that no name form and no damaged form of the name reaches; and the same code written three other ways (as the report field “RM CODE 82289”, as the bare number, and as the list address built from it) reaches 495 more that no other arm used on this page reaches at all. Those reports are already described, as reports, at Bank Records of the Epstein Entity Network; what is new here is the sourcing of the code to the man, which is set out below and was not previously established from a document.96

The book of business

A one-page customer-list printout headed “2018” lists 33 client rows and names the same person, Stewart Oldfield, as Customer Owner of every one of them. Epstein’s own row carries Risk “H”, Type “Individual”, Client Since Date 12 September 2013 and Relationship “Southern Financial Relationship”. Twenty further rows on the sheet carry the same relationship number as Epstein’s, and eighteen of those nineteen organisations carry the same “H”. That printout has its own page here: EFTA01433023.3

A separate high-risk KYC list from 2017 shows the shape of the rest of the book. Under Oldfield’s name as Customer Owner it carries, besides the Epstein entities, “Black, Leon D.” under the “ELYSIUM MANAGEMENT RELATIONSHIP” and “Lindemann, Adam M.” under the “LINDEMANN FAMILY RELATIONSHIP”, both rated HIGH.5 Leon Black’s financial dealings with Epstein are documented on his own page and are not repeated here; this sheet records only that the same bank officer carried both relationships.

The bank’s Account Planning Initiative sheets of July 2016 give the coverage picture in the bank’s own template. The Southern Financial LLC sheet names the beneficial owner as Jeffrey Epstein, the risk appetite as “High”, the length of the wealth-management relationship as three years, the last client meeting as 25 May 2016, the main contacts at the client as Jeffrey Epstein, Rich Kahn and Darren Indyke, and the relationship manager as “Stew Oldfield”. Against “Recent changes to DB relationship / contacts?” it reads: “Departure of Paul Morris.” The action plan’s first row is “Conversations with Epstein re ISG capabilities”, responsible “Oldfield”, due 31 December 2016, status “Preliminary”.4 The same July 2016 series carries sheets for Elysium Management and for Third Lake Capital, both also naming Stew Oldfield as relationship manager and both also recording Paul Morris’s departure.1920

His own account of how he got there

On 25 June 2018 Oldfield sent a colleague, Terri Sohrab, a document headed “planning email for Patrick” with the covering line “Please let me know what you think.” It is a career summary, and one of its headings is “MDs who should be supportive of promotion”.8

It gives twenty years in banking, sixteen of them in investment banking and capital markets at JP Morgan (1998–2002) and Credit Suisse (2002–2014), and four in private banking at Deutsche Bank. Of the move it says:

Had 2 competing DB offers in 2014 — Investment Banking (derivatives) and Asset Mgmt (liquid alts). Sought advice from Chip Packard, whom I had known for 12yrs

Chip instead recruited me to join private bank, which was trying to become more investments-focused. Joined October 2014.

I was initially teamed up with Paul Morris, who had joined from JPM 2yrs prior and was struggling. KCP was fairly new in 2014 and ISG was not yet formed.

Paul had limited success and was asked to leave in 2016.8

The Department of Financial Services calls a message of 5 May 2013 the Approval Email and gives its sender no name, only the label EXECUTIVE-1.21 Deutsche Bank’s own Divisional Control Officer, setting the history of the relationship out for the head of the business in July 2019, wrote that the reputational risk concerns “were escalated to regional management by Chip Packard, Head WM Americas, including John Caruso, Head AML Compliance Americas” and that “Chip was given the go ahead to proceed “so long as nothing further is identified through KYC and AML client adoption””, the inner quotation marks being the sheet’s own.22 That account, and the order’s findings about the email, are set out at Deutsche Bank. What this page adds is that the banker who held the relationship from 2014 names Chip Packard as the man who recruited him, and that in July 2018 he called the document at issue “Chip’s email”. The passage quoted above is Oldfield’s own account, about his own hiring and about a colleague; it is not a finding by anyone. The consent order does not name the relationship manager who replaced the first one, and this page does not identify anyone with that unnamed person.

Under “Client highlights” the same document says of the Epstein relationship:

Southern Financial — one of the most complicated client situations I’ve seen. They have been extremely vocal about our lack of trading capabilities since I joined. Withdrew large portion of assets during 2016 (not DOJ related). Client was offboarded by Global markets at the end of 2016 due to lack of profitability. Further internal issues nearly caused us to offboard the client completely. Client was quite close to Paul and viewed his departure as a negative. I’ve managed to salvage and massively improve this relationship in the last two years. To start, we won $50mm of deposits into DBNY in 2017. We’ve also turned around the GM/trading issues via KCP. Client has been re-onboarded and ISDA reestablished, and is now the largest trading counterparty of the KCP capital markets group. Current balances are ~$230mm across brokerage and deposits.8

The document’s own financial progression reads: 2014 revenues (Paul Morris) $0.9mm; 2015 (combined with Paul) $1.3mm; 2016 (with Paul until May) $1.6mm; 2017 (Stew alone) $1.7mm with a $2.5mm exit run rate; 2018 $4mm annualised. Net new assets: 2016 negative, 2017 $112mm, 2018 $211mm.8 The “DOJ crisis” and “DOJ news” the document blames for 2016’s negative net new assets are not explained anywhere on its face, and the same document separately says the Epstein withdrawal that year was “not DOJ related”; this page does not supply what the document leaves unstated.

June 2016: “Are we doing business with Epstein?”

On Friday 10 June 2016 at 3:22 PM, Andrew King, a vice president in Deutsche Bank’s Investment Solutions Group, wrote to Zbynek “Ziggy” Kozelsky, a director and branch administrative manager in Key Client Partners, copying Oldfield:

I’ve attached a list of preferred equities that was asked for by Richard Kahn; the CFO for Southern Financial, which is an investment entity for the personal wealth of Jeffrey Epstein (one of Stew Oldfield’s clients).

The list isn’t a recommendation. Instead, it is a list of commonly traded preferred equities, which Richard and Jeffrey can select from to build their own basket.23

Nothing came back over the weekend. On Monday 13 June at 5:00 PM King chased: “I thought I’d send a quick note to see if you were able to review the list. Stew had a conversation with the client today, and he’d like us to send something as soon as possible.”23

On Tuesday 14 June, at a stamp the produced header prints as 2:43:44 PM, Oldfield answered both men: “We can’t take three days to send an email with basic information to one of our most sophisticated clients.”23 A separate message, forwarding the same chain to Todd Stevens, carries an RFC date header of Tue, 14 Jun 2016 14:44:16 -0400. Its entire body is: “Fyi, this process is completely broken.”24

Kozelsky replied at 2:50:11 PM: “You are absolutely right. And I have not even looked at it yet. Will send to MRG this PM.”25

At 4:56 PM Kozelsky wrote again, to King, copying Oldfield:

Guys,

Are we doing business with Epstein? This is a KCP client, why is ISG involved with the client?

Thanks26

At 5:06:14 PM Oldfield answered:

He is a private banking client … And ISG and KCP report to the same person. Who is now copied on this email because this is getting silly.26

The thread ran on into the evening: Todd Stevens at 5:31 PM, then Zia Memon, a managing director and head of the wealth management business supervisory office, at 5:56 PM setting out who had to approve a marketing piece and how long it should take; Oldfield forwarded Memon’s message to Terri Sohrab at 5:59:18 PM, and she replied at 7:07:03 PM: “Go Zia!”2728

Nothing in this thread is about Epstein’s history. Kozelsky’s question is stamped 4:56 PM and Oldfield’s answer 5:06:14 PM, and what is answered is that Epstein is a private banking client and that two internal groups report to the same person. No document read for this page records the question being taken any further, by anyone.

What he told the bank’s compliance functions

December 2016: the Haze Trust alert

On 20 December 2016 the bank’s anti-money-laundering compliance function pressed, for the third time, for the origin of funds behind an alert on The Haze Trust. Vahe Stepanian wrote to the distribution at 10:09 AM: “We don’t know the answer. Are we supposed to reach out to client about this?” Two minutes later Oldfield replied:

Client is a billionaire. It’s not like this is a specific $6mm of his. Most of his wealth comes from investments and advisory fees.29

Compliance answered both of them at 10:27:41 AM, still without an origin: “Please find out where these funds originate. We need it for audit purposes.”29 That exchange, and the thread it sits in, has its own page at EFTA01369019; The Haze Trust is documented separately.

May–June 2018: the Panama negative media

On 24 May 2018 Vijay Sawant, a KYC Case Representative working from Deutsche Bank’s CIB Centre in Mumbai, opened the 2018 periodic review of Prytanee, LLC, “above high risked account due for remediation on revised date — 6/30/2018”, and set out the related parties he had identified: Prytanee, LLC, owned half by Southern Trust Company, Inc. and half by The Pierre Trust; Etienne Binant as managing member; Jeffrey E. Epstein as sole shareholder and a director of Southern Trust, with Darren K. Indyke and Richard Kahn as the other directors; and Caroline Sophie Camille Lang as grantor, trustee and ultimate beneficial owner of The Pierre Trust. He asked whether the list was missing anyone or whether there were material changes, and noted that he could not find Prytanee, LLC in the US Virgin Islands registry at all: “The search attempts reflected ‘No Records Found’ for Prytanee, LLC.”30

On 29 May 2018 at 16:58:26 −0400 Oldfield answered in full: “No material changes. The last Virgin Island cert I have is attached. Please let me know what else we need”.30

Sawant came back on 1 June 2018:

The BIS searches have returned with a few alerts for whom I’ve emailed AML team for clearances. Will attach to case when received.

During Google Panama searches, I encountered attached negative media on RP’s Etienne Binant & Jeffrey E. Epstein. Please note there were multiple articles on Mr Epstein all relating to the same case addressed in the attachment.

I look forward to your feedback, thanks.2

Oldfield’s reply is stamped 9:37:59 AM the same day:

Thanks. We’ve taken Epstein through high risk review on a regular basis, so there shouldn’t be new news there. Will review the other one Monday when I’m back unless we have a simpler way to clear it. Thanks.2

Neither timestamp on that sheet carries a time zone, so this page states both as printed and computes no interval between them.

July 2018: “As for Chip’s email, it predates me”

The Prytanee case came back rejected. On 4 July 2018 Sawant sent Oldfield a list of rejection points on KYC Case#01946825, among them: “For Point 6 – Could you provide the email that the AML team is seeking? I only have the attached email (2nd pdf attachment) from the repositories.” and “For Points 8 & 9 – Kindly provide SOW info for Jeffrey & Caroline.”13

Oldfield answered on Friday 6 July 2018 at 12:14 AM:

Let’s set up a time to discuss when Brad is back next week. Some of these requests seem unreasonable, and AML should already have a lot of this info, particularly anything relating to Jeffrey Epstein personally (since we have so many accounts for him).

As for Chip’s email, it predates me. And the other people on the email are no longer at DB. So I don’t see how I can get hold of the full email. But again, Jeffrey has been through a complete rep risk process here and has been signed off on repeatedly by management as we’ve done any KYC remediation or new account opening.

Anyway, let’s regroup next week so we can go back to AML with a comprehensive set of responses.13

Set beside the regulator’s findings, and against what the bank itself did a year later, that is the most consequential paragraph Oldfield wrote in the release.

The Department of Financial Services found that the initial onboarding of the first account “was not reviewed by the Bank’s regional reputational risk committee but was instead approved in what appears to have been an off-hand conversation reflected only in the Approval Email”, and that “[t]hat Approval Email was then relied upon, substantially without additional scrutiny, to open numerous other Epstein-related accounts.”15 Of the email itself the order records: “The Bank has represented to the Department that it has no other record of this communication between EXECUTIVE-1 and the other officers, and the ARRC did not meet in connection with the initial onboarding of Mr. Epstein.”21

A year after Oldfield’s answer, an email of 2013 was attached to the bank’s own internal account of the onboarding. On the evening of 8 July 2019, two days after Epstein’s arrest, Kimberly Hart, Managing Director and Divisional Control Officer for Wealth Management Americas, set the history of the relationship out for Patrick Campion, Managing Director and Head of Wealth Management Americas. Her message is stamped 6:28 PM on the sheet. It recites the 2008 conviction and the onboarding, names Chip Packard and John Caruso, and ends that paragraph: “Email from 2013 is attached.” It continues: “A number of KYC updates have been processed for Mr. Epstein since 2013 and all alerts have been cleared – often relying on the original clearance by Chip Packard and John Caruso. Sample of an RDC clearance from 2016 is attached.” Campion’s reply, stamped 7:26:25 PM in the produced header, is one line: “What was the catalyst for our decision last year to exit?”22

That the document Oldfield could not get hold of in July 2018 is the document Hart attached in July 2019 is an inference by this page. Neither sheet says it, and the two are set side by side here as an argument, not as a finding. What supports it: the clearance wording Hart recites, quoted above, is the operative clause of the Approval Email itself, which the consent order reproduces at ¶22 as the closing words of the sentence in which EXECUTIVE-1 cleared the relationship, differing only in that the order reads “adoptions” where Hart reads “adoption”; both documents concern a pre-2014 email in which Chip Packard figures; and Oldfield’s own word for the document the AML team was seeking is “Chip’s email”.21 What is not settled: the order has the Approval Email sent by an unnamed EXECUTIVE-1 to an unnamed RELATIONSHIP MANAGER-1, while Hart has Packard given the go-ahead, and this page does not identify Packard with either label. No document read for this page explains why an email of 2013 could not be got hold of in July 2018 and was attached a year later, and this page does not explain it either. The order of the July 2019 thread is worth stating because it is easy to get backwards: Hart wrote first, at 6:28 PM, and Campion’s question came after, at 7:26:25 PM.

What the documents do not show is part of this too. No document read for this page records anyone above Oldfield querying that answer, and none records the full email reaching the 2018 KYC case. The page does not assert that neither happened; it records that the searches set out below did not reach any sheet in which it did.

The other reviews

Oldfield’s answers to compliance about individual payments are documented on the pages of the people and entities they concerned, and are not restated here. Asked in May 2018 why an entity account was being used to make one-off transfers, he told compliance “Amar Siad is a professional photographer. The other two are students,” and that Epstein and his finance staff “have the flexibility to use any account they like that is funded”; see Daniel Siad. Asked in November 2018 about the beneficiaries of Epstein’s insurance trust, he wrote “There are later amendments that remove the original bens and name Celina”; see Eva Andersson-Dubin. The Zorro Management rush KYC of December 2016 names him as relationship manager; see Jared Kushner. The 2018 periodic review of the “Southern Financial Relationship” records that “Jeffrey Epstein reached out to Stew Oldfield on 3/7/17 to have a new acct opened under this relationship”; see Enhanced Education. A 2018 KYC print’s approval block carries his name among four; see Bill Clinton.

On 14 September 2018 Daphne Cales, a director and branch manager, wrote to a colleague about a rejected Gratitude America know-your-customer case. Three grounds are recorded on the sheet. The first is a ten-million-dollar gift to the charity from BV70 LLC, recorded in a 2015 return attached to the case file, which on the ground as stated made the giver itself a party the bank had to examine in full. Cales asks how the client can be approached on that ground, and names Oldfield as the banker to go to if that is needed.31 The link between BV70 LLC and Gratitude America is documented at Deutsche Bank and at Leon Black.

11 June 2018: the ATM card

At 11:32 AM on Monday 11 June 2018, Richard Kahn of HBRK Associates wrote to Oldfield under the subject “JEE ATM card we use to withdraw cash”:

would it work in paris to take out euros from an ATM machine? please advise thanks32

At 12:09:11 −0400 Oldfield forwarded it to Bradley Gillin and Rodney Macelus, copying Daphne Cales, over two words: “Should work, right?”33

At 12:10:37 −0400 Cales answered: “Yes, it would work but you need to add a note on the system with the travel dates and locations.”34 At 12:13 PM Gillin instructed a colleague to “put up a ticket to let back office know the client will be in Paris”.35

At 12:15:24 −0400 Oldfield answered Kahn, copying Gillin and Macelus:

Yes, it will work. Let us know what dates he will be there and we will make sure we notify the network that transactions in France are to be expected (so they don’t get blocked by some fraud prevention software)32

At a stamp the produced header prints as 12:22:32 PM, Cales wrote once more, to Gillin: “We need the dates.”35

What this exchange establishes is what it says. A client’s chief financial officer asked whether Epstein’s cash card would work abroad; the relationship manager and the branch cleared it and arranged that the bank’s own fraud controls would not stop the withdrawals. The Department of Financial Services’ separate finding, that the bank did little or nothing “to inquire how Mr. Epstein was using, on average, more than $200,000 per year in cash”, concerns branch counter withdrawals by Epstein’s attorney and is set out at Deutsche Bank.15 No document read for this page connects the two, and this page does not connect them. No sheet read here records any question being asked, by anyone in the thread, about what the cash was for.

2019: the accounts come off the book

By April 2019 the relationship was being wound down inside the bank. On Monday 15 April 2019 at 11:59:19 −0400 Ana Hunter, in Deutsche Bank Wealth Management’s Jacksonville office, wrote to Oldfield and a colleague: “Hi — the below client ended up on the CIB offboard list. Client will be offboarded today unless you confirm we need to retain. Please confirm: SOUTHERN FINANCIAL, LLC”.36 Eighty-four seconds later, at 12:00:43 −0400, Oldfield replied: “I think we closed out the last CIB trade they had, in which case we can off board. Liam - please confirm.”1 At 12:16 PM Liam O’Sullivan confirmed: “There are no open trades.”37

On 26 March 2019 at 14:31:10 −0400 Oldfield had written to Kahn: “Rich, I’ve been asked for another update on this process. Please let me know what you expect balances to be by end of the month and how much longer until you can close the remaining accounts. Thanks for your help.”38

On 24 May 2019 Robert Frohlich of Deutsche Bank AG’s New York branch sent Oldfield a table of his own financials. Its rows are headed “Steward Oldfield”, the bank’s own spelling on this sheet, and read: revenues, FY18 actuals $3,576,567, of which Southern Financial $703,616; April 2019 year to date $1,600,929, of which Southern Financial 0; FY19 target $4,026,207. A separate “Southern Financial House Acc.” row carries $486,673 for FY18 and $363,457 for April 2019 year to date. Net new assets, in $m: Oldfield 73.2 for FY18, of which Southern Financial (49.2); April 2019 year to date 360.7, of which Southern Financial 0.0, with the House account at (85.0). The covering message asks “whether the 2019 YTD Southern Financial revenues of $363k currently mapped to a House account should be mapped back to your name alongside the $-85m NNA outflow (no further flows and revenues expected).”14

Both movements belong in the same sentence. Oldfield’s total revenue line is rising, $3.58m for FY18 against a $4.03m target for FY19, while the Epstein component of it falls from $703,616 to zero and the assets leave at $85.0m. The document is a question about where to book a departing relationship, and it is dated forty-three days before Epstein’s arrest.

Epstein was arrested on 6 July 2019; the arrest and its sourcing are set out at Deutsche Bank. At 7:33 the following Monday morning, Kimberly Hart sent Oldfield and three colleagues the instruction headed “URGENT!!! Need to close accounts ASAP - please prioritize today”; Oldfield passed it on at 8:34 with the words “FYI. Thanks.”, and forwarded the answer on at 8:55. That sequence, and the closures that followed, are set out at Deutsche Bank.

The last dated documents in which Oldfield writes are of 25 and 26 July 2019, to Patrick Campion, Head of Wealth Management Americas, under the subject line of a Wall Street Journal article. On 25 July at 7:23 PM: “Paul Barrett was the head of GIO at JPM who handled trading for Jeffrey through his (Paul’s) RIA. Would need to look back to see when he started that, but I think it was probably 2017.” On 26 July at 11:39:45 −0400, under the classification “Strictly Confidential”: “There’s an article today about Epstein’s charity receiving IPO allocations from Morgan Stanley as recently as 2017. We understood him to have been in the syndicate at MS and at least one other bank while Paul was running some of his trading.”11 Those are Oldfield’s statements to his own management about third parties, nineteen and twenty days after the arrest. This page does not adopt them, and no document read for it independently establishes either.

How the population was measured

Every figure in this section was run. None was estimated. All counts are documents, not pages, taken on the authoritative pages_fts route over the full-text index of the release; page counts differ and are not used here.

The name forms, with every zero. "Stewart Oldfield" 9,606 · Oldfield 9,815 · "Stew Oldfield" 156 · "Oldfield Stewart" 194 · "S. Oldfield" 4 · "Stewart J. Oldfield" 0 · "Stewart Oldfields" 0 · Oldfields 0 · "Stuart Oldfield" 2 · "Steward Oldfield" 26 · "stewed oldfield" 3 · "stenett oldfield" 1 · "stew oldielf" 1 · "Oldfield/db" 200 · "S_Oldfield" 4 · "Stew Oldfield's" 64 · "Relationship Manager: Stew Oldfield" 7 · "Account Manager: stew oldfield" 2. The surname is a superset of the full-name phrase: no document carrying "Stewart Oldfield" falls outside Oldfield, and 209 documents carry the surname without the phrase.

The middle-initial and legal forms return nothing. "Stewart J. Oldfield" returns 0 and "S. Oldfield" returns 4, all four already inside the surname population. That is a negative worth stating, because on another subject measured this month the middle-initial form and the bare name proved to be disjoint in both directions, each holding hundreds the other missed. Here it does not arise.

The enumeration. The index was enumerated with Damerau–Levenshtein distance ≤ 2 against the target oldfield (8 characters) over a length window of 6 to 11 characters (target −2 to +3), using an fts5vocab row table built in the connection’s temp schema on a read-only handle. It returned 133 distinct index terms besides oldfield itself. The whole-index form of this enumeration did not complete: a scan over all index terms was run twice under explicit timeouts, at 600 s and at 2,300 s, and finished neither time; the second left a zero-byte output file, which on disk is indistinguishable from a null and is reported here as an unfinished scan rather than as a result. What was run instead were prefix ranges, exhaustive within each range: the whole of o (278,458 terms scanned in 158 s), the whole of old, and 55 two-character heads: 35 of the form Xl, one for every digit and every letter but o; four lX ranges (lc, ld, lf, li); and sixteen oX ranges. The head bound, stated so it can be checked: a term with two errors in its first two characters, outside that head set, is unreachable by this enumeration.

Every candidate was run twice, as a bare token and as the phrase "stewart <form>", and the two disagree in both directions. 0ldfield (a zero for the letter o) is the largest single damaged form: 657 documents as a token, 647 as the phrase, 280 outside the base population. Other substantial forms: oldficld 127 (101 outside), oldfiel 210 (78 outside), oldlield 36 (29 outside), oldfieid 34 (22 outside), oldfleld 38 (13 outside). The bare-token route reached 114 documents the phrase route could not; every one of those 114 was read in context and assigned individually. 83 are him, almost all of them bank statements whose contact line the scanner mangled at both ends. 31 are not, and they are named rather than discarded: oilfield, outfield and battlefield in news copy and in a soccer rulebook; Orefield, PA in an address list; a fund-performance table naming Geoff Oldfield; a reference to the Edinburgh Handedness Inventory, “Oldfield, 1971”; and David Oldfield, a vice president, named once. Those named homographs, and Oldfield Pipeline, were subtracted from the surname population.

Two forms are outside edit distance altogether and were found only by reading. The scanner renders his surname as the English word Chiliad on ten sheets of the statement family, and splits it as OW ield on fifteen more. Neither is generable by any substitution table, and Chiliad is beyond Damerau–Levenshtein 2 from oldfield.

The spider. Reading the hits produced identifiers the name cannot reach, and each was run:

armdocumentsoutside every name form and the enumeration
RM0082289 (his relationship-manager code, prefixed form)1,1891,134
the same code written three other ways: "RM CODE 82289" 1,345, the bare "82289" 1,669, "[email protected]" 289union 1,671495 further, beyond every other arm in this table, one homograph subtracted
his direct telephone line, in the two forms the release prints it30032 reachable by no other arm
the mobile line in his signature block1081
"Stewart Oldfield CFA CAIA"1,8850
"stewart.oldfield"9,6060 — the index splits on the dot, so the address is not a separate reach
"Steuart Chiliad" · "OW ield" · "Oktficld" · "Stewart Old Field"10 · 15 · 4 · 39 between them
the personal address he mailed himself from in October 201410 — the extraction damages it and no clean form is in the index
"Geoff Oldfield" · "David Oldfield" · "Oldfield Pipeline" · "Oldfield Annand"1 · 1 · 12 · 0subtracted as homographs

A code is a name form, and it has to be enumerated like one. The four written forms of code 82289 are four different populations, and the prefixed one is not the largest: RM0082289 returns 1,189 documents, the report field "RM CODE 82289" returns 1,345, the bare "82289" returns 1,669, and the list address returns 289; their union is 1,691. Of that union, 496 documents lie outside every other arm used on this page, and every one of the 496 was classified from its own text rather than counted. One is a homograph and is subtracted: an unrelated message whose mail-client metadata happens to carry the same five digits as a conversation identifier. The other 495 are all the same automated series, balance and deposit reports keyed to the code. 460 of them carry the report heading “LARGE, ZERO AND NEGATIVE BALANCES” somewhere in the document and 113 carry “Daily Deposit Report”, of which 106 carry both, these being bundled multi-report extracts; so those two groups are not exclusive and do not sum to the total. The remaining 28 carry a form of one of those headings that the scanner has broken, and none of the 495 is left unclassified. The narrower forms miss them for a reason worth recording: on the documents the report-field query cannot reach, the extraction has detached RM from CODE 82289 or dropped a column between the two, so a phrase query fails on sheets whose face carries the field intact.

Neither telephone value is reproduced on this page. Both print in the clear on sheets this page cites and links, so a reader who wants to re-run either arm has the sheet. The direct line is the more useful of the two: it is the string that reaches 32 documents no name form, no damaged form and no code reaches. The mobile line reaches one document nothing else reaches, which makes it decoration by the desk’s own test.

The code was sourced to the man at the image. A produced Deutsche Bank database extract prints, on a single row beside SOUTHERN TRUST COMPANY, INC., the fields DDA 82289 Stewart Oldfield … Stewart Oldfield; a second extract in the same series carries the same code against his name on Southern Trust Company rows of 2017; and an internal message of 29 March 2018 answers Oldfield’s request to add a colleague to the distribution with “Brad Gillin is already member of list [email protected]”. Read at 400 dpi.6397

The floor. Name forms with homographs subtracted 9,803, plus 728 from the enumeration, plus 1,134 from the relationship-manager code in its prefixed form, plus 495 from the other three written forms of the same code with one homograph subtracted, plus 36 from the direct and mobile lines, plus 9 from the space-split and far-OCR forms, giving 12,205 documents. This is a floor and not a population. It does not reach: a sheet on which the producing party barred his name; a sheet whose scan damaged the name beyond the window and head bound above; a reference to him by first name alone inside a thread that never spells the surname; or any document reachable only through an identifier nobody has yet read off a sheet. The commission that produced this page carried 9,606; the measured floor is 27.1% larger, and the whole of that difference came from reading documents rather than from guessing at spellings.

The act rate. Measured exhaustively over the 9,803-document name-form population rather than sampled: 2,298 documents carry a form of his name in the sender field: 23.4% of the population, and 26.5% of the 8,672 documents in it that carry any sender value at all. On the full 12,205 floor the same measure is 2,314, or 19.0%, because the relationship-manager reports are automated and carry no personal sender: not one of the 495 documents added by the code’s other written forms carries a sender value in his name, and only 29 of them carry any sender value at all. He appears in a recipient value on 1,945 of the 9,803, or 19.8%. State the denominator with this number: the commission carried 26.0% from a 400-document sample, which matches the with-a-sender denominator and not the population one.

Dates. The classifier’s date field is wrong on at least two documents that matter, and on both the sheet’s own header was read at the image at 400 dpi to settle it. It dates one Oldfield message to 2009; the sheet’s own RFC header reads Tue, 26 Mar 2019 14:31:10 -0400. It dates another to 2011; the sheet prints Sent: Friday, July 06, 2018 12:14 AM, and the extraction’s 2011t for 2018 is why. Corrected, the earliest document in the floor in which Oldfield writes is of 6 October 2014 and the latest is of 26 July 2019. The distribution of the 2,278 datable sender documents runs 2014: 15 · 2015: 60 · 2016: 326 · 2017: 627 · 2018: 888 · 2019: 360. Those six buckets are taken from the classifier’s date field and sum to 2,276, two short of the total beside them, and the two missing documents are the two the field mis-dates, the ones named in this paragraph, which it files under 2009 and 2011 and whose sheets read March 2019 and July 2018. Read by the sheets rather than by the field they belong in 2019 and 2018, which is where the difference between 2,276 and 2,278 goes; the buckets are printed as the field records them so that both numbers and the item between them are on the page.

Clocks. Where a sheet carries an RFC Date: header with an offset, that offset is quoted and used. Where a sheet carries only a produced Sent: field with no zone, this page prints the value and computes no interval across it. On the June 2016 thread the two can be checked against each other: a message the produced header stamps 6/14/2016 2:43:44 PM is quoted in a sibling copy as Sent: Tuesday, June 14, 2016 2:44 PM, and a second message from the same sender that afternoon carries the RFC header Tue, 14 Jun 2016 14:44:16 -0400. The produced field and the quoted header agree to the minute on the same message, so on this family the produced field is Eastern and not coordinated universal time. That is a finding about this family of sheets and is not carried to any other.

What the record read for this page does not establish

  • Anything about what Oldfield knew of Epstein’s conduct. No document read here records him being told anything about it, or asking.
  • What was said on the telephone. More than once the written record stops at an arrangement to speak. In December 2016 Richard Kahn told compliance he had “provided this information” to Oldfield and proposed a call; Oldfield answered that he would “be driving at 11” and could call “around noon”; the produced thread ends there; see Bank Records of the Epstein Entity Network. In July 2018 he proposed to “set up a time to discuss when Brad is back next week”. No document read for this page records what was said at any of those calls.
  • Whether the email the AML team was seeking in 2018 is the Approval Email, whether it was ever produced to that KYC case, and why an email of 2013 could not be got hold of then and was attached a year later. Searches of the full-text index for "Chip's email" (43 documents), "complete rep risk process" (39) and "signed off on repeatedly by management" (38) return copies of Oldfield’s own message and nothing that answers it. Those three searches were written down before they were run and are reported here whatever they returned; they are not built from words already on this page. They do not settle the question, and this page does not assert that the email was never produced. It records only that no sheet read here shows it reaching the KYC case, and that no sheet read here records anyone above Oldfield querying his answer.
  • Any statement by him about any of this. No public statement by Stewart Oldfield about the Epstein relationship was located, and this page does not assert that none exists.
  • Whether he is the relationship manager the consent order leaves unnamed. The order describes a second relationship manager in place by April 2016 who had the know-your-customer file and knew of the earlier escalation but was not told of the 2015 committee’s conditions, and does not name him. This page does not name him either.

Notes

Footnotes

  1. Stewart Oldfield to Ana Hunter and Liam Osullivan, copied to Monica Gilkinson and Armen Brash, “Re: CIB Offboard”, RFC date header Mon, 15 Apr 2019 12:00:43 -0400, quoting Hunter’s message of “Apr 15, 2019, at 11:59 AM”. Read at the image at 400 dpi. His signature block on this sheet prints his name, charterholder designations, both Deutsche Bank entities and the 345 Park Avenue address in the clear, together with a direct telephone line and a mobile line. Telephone-class values are not reproduced on this page. https://epstein-data.com/EFTA01426551 pp. 1-2. ↩ ↩2

  2. Stewart Oldfield to Vijay-A Sawant, copied to Bradley Gillin, Hildegarde Berrios and Daphne Cales, “Re: 2018 Periodic Review of acct Prytanee, LLC … under SOUTHERN FINANCIAL RELATIONSHIP”, produced header stamped 6/1/2018 9:37:59 AM with no time zone, quoting Sawant’s message of “Jun 1, 2018, at 9:25 AM”, which also carries no zone. Read at the image at 400 dpi. Sawant’s own signature block gives “KYC Case Representative / Deutsche CIB Centre Private Limited - Deutsche Bank Group / Wealth Management”, Mumbai. https://epstein-data.com/EFTA01369499 p. 1. ↩ ↩2 ↩3

  3. Deutsche Bank customer-list printout headed “2018”, one page of data, naming Stewart Oldfield as Customer Owner of all 33 rows. Already documented in full at EFTA01433023 and not independently re-sourced here. https://epstein-data.com/EFTA01433023 p. 1. ↩ ↩2

  4. Deutsche Bank, “Account Planning Initiative July 2016 (v1.0)”, account review sheet for Southern Financial LLC, marked “Illustrative Draft”. Read at the image at 400 dpi. The sheet is a produced rendering of a two-column form flattened into a single stream of text, so this page reports only pairings the flattening does not put in doubt — the field labels quoted and the values immediately following them. https://epstein-data.com/EFTA01422032 p. 1. ↩ ↩2

  5. Deutsche Bank internal message quoting a high-risk KYC schedule, listing under Customer Owner “Stewart Oldfield” the rows “Black, Leon D.” (ELYSIUM MANAGEMENT RELATIONSHIP, client since 4/17/2014), “Lindemann, Adam M.” (LINDEMANN FAMILY RELATIONSHIP, client since 2/1/2010), “Southern Trust Company, Inc.”, “Southern Financial LLC”, “Butterfly Trust” and “The 2007 Jeffrey E. Epstein Insurance Trust #3”, each rated HIGH. Read in the extraction only; nothing above rests on the position of a value on the sheet. https://epstein-data.com/EFTA01356260 p. 1. ↩ ↩2

  6. Deutsche Bank database extract produced in the DB-SDNY series; a single row beside “SOUTHERN TRUST COMPANY, INC.” prints the fields “DDA 82289 Stewart Oldfield” and repeats “Stewart Oldfield”. Read at the image at 400 dpi. No account-number-class value is reproduced on this page. https://epstein-data.com/EFTA01419809 p. 138. ↩ ↩2 ↩3

  7. Rishab Sarawagi to Stewart Oldfield, copied to Bradley Gillin, “RE: RM0082289 Daily Deposit Report 20180329 [I]”, RFC date header Thu, 29 Mar 2018 15:10:33 -0400, answering Oldfield’s “Please add Brad Gillin to this list” with “Brad Gillin is already member of list [email protected]”. Read in the extraction only. https://epstein-data.com/EFTA01420506 p. 1. ↩ ↩2

  8. Terri Sohrab and Stewart Oldfield, “planning email for Patrick”, 25 June 2018; Oldfield’s message of 5:17 PM enclosing the career summary, Sohrab’s reply of 18:12:45 −0400. Seven sheets. Read at the image at 400 dpi (pp. 1, 2, 3, 4); the heading “MDs who should be supportive of promotion” quoted above prints on p. 4. The summary is Oldfield’s own composition and is quoted as such. The image prints ”~$230mm”; the extraction renders the tilde as an em dash. https://epstein-data.com/EFTA01421399 pp. 1-7. The same document is cited at Deutsche Bank. ↩ ↩2 ↩3 ↩4 ↩5

  9. Populations for this page were measured on the authoritative pages_fts full-text route over the released corpus, counting distinct document identifiers rather than pages. The arms, their counts and the bounds of the floor are set out in full under “How the population was measured” above. Floor: 12,205 documents. ↩ ↩2

  10. Stewart Oldfield to Stewart Oldfield, no subject, RFC date header Mon, 06 Oct 2014 17:10:10 -0400. Read in the extraction only. This is the earliest dated document in the floor in which he writes. The non-bank sender address on it is damaged in the extraction and is not reproduced here. https://epstein-data.com/EFTA01426370 p. 1. ↩

  11. Stewart Oldfield to Patrick Campion, “RE: WSJ: Jeffrey Epstein Burrowed Into the Lives of the Rich and Made a Fortune [I] [SC]”, RFC date header Fri, 26 Jul 2019 11:39:45 -0400, classification “Strictly Confidential”, quoting his own message of Thursday 25 July 2019 7:23 PM. Read at the image at 400 dpi. https://epstein-data.com/EFTA01410768 p. 1. ↩ ↩2

  12. Measured exhaustively over the 9,803-document name-form population by matching a regular expression for the damaged and clean surname forms against the sender value recorded for each document, not by sampling. 2,298 of 9,803 (23.4%); 2,298 of the 8,672 documents in that population carrying any sender value (26.5%); 2,314 of the full 12,205 floor (19.0%). Recipient: 1,945 of 9,803 (19.8%). ↩

  13. Stewart Oldfield to Vijay-A Sawant, copied to Bradley Gillin, “RE: Assistance Required in solving rejection points on KYC Case…”, printed on the sheet as “Sent: Friday, July 06, 2018 12:14 AM”, quoting Sawant’s message of Wednesday 4 July 2018 9:15 AM on KYC Case#01946825 relating to Prytanee, LLC. Read at the image at 400 dpi. The extraction of this sheet is heavily damaged and was not used for the quotation: it renders the date “July 06. 2011t”, “people” as “Foyle”, “remediation” as “remcdiation”, “Anyway” as “Amway” and “solving” as “soh ing”. The corpus date field for this document reads 2011 in consequence; the sheet reads 2018. https://epstein-data.com/EFTA01388741 p. 1. ↩ ↩2 ↩3

  14. Robert Frohlich, Deutsche Bank AG Filiale New York, Wealth Management, to Stewart Oldfield, copied to Leslie Roth and Donna Kerrigan, “Southern financial [C]”, produced header stamped 5/24/2019 2:52:03 PM, classification “Confidential”, with a two-block table headed “FY18 Actuals”, “Apr19 YTD” and “FY19 Target”. Read at the image at 400 dpi. The table’s row label prints “Steward Oldfield” — the misspelling is the sheet’s own. https://epstein-data.com/EFTA01372308 p. 1. ↩ ↩2

  15. Consent Order under New York Banking Law §§ 39 and 44, In the Matter of Deutsche Bank AG, Deutsche Bank AG New York Branch, and Deutsche Bank Trust Company of the Americas, New York State Department of Financial Services, July 6, 2020, ¶¶56 and 58, p. 19. Read at the image at 400 dpi. On the sheet the words “in cash” at the end of ¶56 are set in italic; they are quoted here in roman. https://epstein-data.com/EFTA00151495 p. 19. The same order is cited throughout Deutsche Bank. ↩ ↩2 ↩3

  16. Consent Order, ¶114, under the headings “SETTLEMENT PROVISIONS” and “Monetary Penalty”: Deutsche Bank “shall pay a penalty to the Department, pursuant to New York Banking Law §§ 39 and 44, in the amount of one hundred fifty million U.S. dollars ($150,000,000.00)”. Read at the image at 400 dpi. The penalty is on a different sheet of the order from the findings quoted at 15. https://epstein-data.com/EFTA00151495 p. 33. ↩

  17. Deutsche Bank Trust Company Americas monthly statement for JEGE, LLC, 1 January 2019 to 31 January 2019, carrying the contact line “For personal assistance call: Stewart Oldfield”. Read at the image at 400 dpi. No account-number-class value is reproduced on this page. This sheet is cited only for the printed form of his name and for the class of document; no transaction on it is relied on anywhere above. https://epstein-data.com/EFTA01253913 p. 1. ↩

  18. The damaged forms named in the text were produced by the fts5vocab enumeration and by reading, and each was run as a query. They are recorded here as finding aids, not as readings: on the statement sheet cited at 17 the image at 400 dpi prints “Stewart Oldfield” cleanly while the extraction of the same sheet renders “Steuart Oldficld”. Read at the image at 400 dpi. ↩

  19. The same July 2016 series, account review sheet for Elysium Management, naming “Stew Oldfield” as relationship manager and recording “Departure of Paul Morris” against recent changes to the DB relationship. Read in the extraction only. https://epstein-data.com/EFTA01422028 p. 1. ↩

  20. The same July 2016 series, account review sheet for Third Lake Capital, naming “Stew Oldfield” as relationship manager and recording “Paul Morris departure”. Read in the extraction only. https://epstein-data.com/EFTA01422036 p. 1. ↩

  21. Consent Order, ¶22, p. 7. Read at the image at 400 dpi. https://epstein-data.com/EFTA00151495 p. 7. ↩ ↩2 ↩3

  22. Patrick Campion and Kimberly Hart, “Epstein” and “RE: Epstein”, July 8, 2019; Campion’s message stamped 7/8/2019 7:26:25 PM in the produced header and Hart’s quoted as “Sent: Monday, July 08, 2019 6:28 PM”, neither field carrying a time zone. Read at the image at 400 dpi. Hart’s signature block gives “Managing Director | Divisional Control Officer / Wealth Management Americas / Deutsche Bank Trust Company Americas”; Campion’s gives “Managing Director | Head of Wealth Management Americas”. The dash in Hart’s sentence is an en dash on the sheet and is quoted as one. https://epstein-data.com/EFTA01356821 p. 1. The same sheet is cited at Deutsche Bank. ↩ ↩2

  23. Stewart Oldfield to Andrew King and Zbynek Kozelsky, copied to Joseph Dursi, “RE: Approval Needed - List of Preferreds”, produced header stamped 6/14/2016 2:43:44 PM with no time zone, quoting Andrew King’s messages of Monday 13 June 2016 5:00 PM and Friday 10 June 2016 3:22 PM. Read at the image at 400 dpi. The extraction renders “Joseph Dursi” as “Joseph Oursi”; the image prints Dursi. https://epstein-data.com/EFTA01373637 p. 1. ↩ ↩2 ↩3

  24. Stewart Oldfield to Todd Stevens, “FW: Approval Needed - List of Preferreds”, RFC date header Tue, 14 Jun 2016 14:44:16 -0400, body in full: “Fyi, this process is completely broken.” Read at the image at 400 dpi. https://epstein-data.com/EFTA01417315 pp. 1-2. ↩

  25. Zbynek Kozelsky to Stewart Oldfield, copied to Joseph Dursi and Andrew King, “RE: Approval Needed - List of Preferreds”, produced header stamped 6/14/2016 2:50:11 PM, quoting Oldfield’s message as “Sent: Tuesday, June 14, 2016 2:44 PM”. Read in the extraction only; this footnote is cited for the quoted-header time, which is the cross-check described under “Clocks” above, and not for any statement about where a value sits on the sheet. https://epstein-data.com/EFTA01374216 p. 1. ↩

  26. Stewart Oldfield to Zbynek Kozelsky and Andrew King, copied to Todd Stevens, “RE: Approval Needed - List of Preferreds”, produced header stamped 6/14/2016 5:06:14 PM with no time zone, quoting Kozelsky’s message of Tuesday 14 June 2016 4:56 PM. Read at the image at 400 dpi. The image prints “ISG” in both Oldfield’s sentence and Kozelsky’s question; the extraction of Kozelsky’s line renders it “156”, and the damaged form is recorded here because it is the only string that will find that line in the index again. Oldfield’s sentence prints with a spaced ellipsis, as quoted. https://epstein-data.com/EFTA01369942 p. 1. ↩ ↩2

  27. Stewart Oldfield to Terri Sohrab, “FW: Approval Needed - List of Preferreds”, produced header stamped 6/14/2016 5:59:18 PM, forwarding Zia Memon’s message of Tuesday 14 June 2016 5:56 PM and quoting a Todd Stevens header of 5:31 PM. Read in the extraction only. https://epstein-data.com/EFTA01373748 p. 1. ↩

  28. Terri Sohrab to Stewart Oldfield, “Re: Approval Needed - List of Preferreds [I]”, produced header stamped 6/14/2016 7:07:03 PM, body in full: “Go Zia!” Read in the extraction only. https://epstein-data.com/EFTA01375554 p. 1. ↩

  29. Danya Friedman on behalf of AML Compliance Inquiries to Stewart Oldfield, Vahe Stepanian, Zbynek Kozelsky, Cynthia Rodriguez and PWM AML Compliance, copied to Joshua Shoshan, Donald Summer and Cherie Quigley, “RE: URGENT - THIRD REQUEST!!!!! RE: Inquiry Regarding THE HAZE TRUST…”, produced header stamped 12/20/2016 10:27:41 AM, quoting Oldfield’s message of Tuesday 20 December 2016 10:11 AM and Vahe Stepanian’s of 10:09 AM. Read at the image at 400 dpi. The same sheet is cited at EFTA01369019. No account-number-class value is reproduced on this page. https://epstein-data.com/EFTA01369017 p. 1. ↩ ↩2

  30. Stewart Oldfield to Vijay-A Sawant, Bradley Gillin and Hildegarde Berrios, copied to Daphne Cales, “RE: 2018 Periodic Review of acct Prytanee, LLC (GCIS#550868) under SOUTHERN FINANCIAL RELATIONSHIP”, RFC date header Tue, 29 May 2018 16:58:26 -0400, quoting Sawant’s message of Thursday 24 May 2018 6:28 AM in full. Read at the image at 400 dpi (pp. 1, 2). Oldfield’s sentence ends without a full stop, as quoted. https://epstein-data.com/EFTA01413443 pp. 1-3. ↩ ↩2

  31. Bradley Gillin to Richard Iarossi, “FW: KYC …”, produced header stamped 9/14/2018 2:40:54 PM, forwarding a message Daphne Cales had sent Derek St-Hilaire, copied to Vinita Advani, Vijay-A Sawant, Bradley Gillin and Stewart Oldfield, Friday 14 September 2018 2:33 PM, setting out the case reference, the relationship, the customer and three grounds of rejection. Read at the image at 400 dpi. https://epstein-data.com/EFTA01360004 p. 1. ↩

  32. Stewart Oldfield to Richard Kahn, copied to Bradley Gillin and Rodney Macelus, “RE: JEE ATM card we use to withdraw cash”, RFC date header Mon, 11 Jun 2018 12:15:24 -0400, quoting Kahn’s message of Monday 11 June 2018 11:32 AM in full. Read at the image at 400 dpi. https://epstein-data.com/EFTA01430415 p. 1. ↩ ↩2

  33. Stewart Oldfield to Bradley Gillin and Rodney Macelus, copied to Daphne Cales, “FW: JEE ATM card we use to withdraw cash”, RFC date header Mon, 11 Jun 2018 12:09:11 -0400, body in full: “Should work, right?” Read at the image at 400 dpi. https://epstein-data.com/EFTA01352573 p. 1. ↩

  34. Daphne Cales to Stewart Oldfield, Bradley Gillin and Rodney Macelus, “RE: JEE ATM card we use to withdraw cash [I]”, RFC date header Mon, 11 Jun 2018 12:10:37 -0400. Read in the extraction only. Cales’s signature block gives “Director | Banking Services & Branch Manager / Deutsche Bank Trust Company Americas / WM Americas / 345 Park Avenue, 10154-0004 New York, NY, USA”. https://epstein-data.com/EFTA01423411 pp. 1-2. ↩

  35. Daphne Cales to Bradley Gillin, copied to Stewart Oldfield and Michele-M Quattrocchi, “RE: JEE ATM card we use to withdraw cash [C]”, 11 June 2018, quoting Gillin’s message of Monday 11 June 2018 12:13 PM and Oldfield’s of 12:09 PM. Read at the image at 400 dpi. Gillin’s salutation and his instruction are separate lines on the sheet and are not run together here; Cales’s reply is likewise quoted without its salutation. No account-number-class value is reproduced on this page. https://epstein-data.com/EFTA01371013 p. 1. ↩ ↩2

  36. Ana Hunter to Stewart Oldfield and Monica Gilkinson, copied to Armen Brash, “CIB Offboard”, RFC date header Mon, 15 Apr 2019 11:59:19 -0400. Read in the extraction only. https://epstein-data.com/EFTA01432353 pp. 1-2. ↩

  37. Liam Osullivan to Stewart Oldfield and Ana Hunter, copied to Monica Gilkinson and Armen Brash, “RE: CIB Offboard [C]”, produced header stamped 4/15/2019 12:16:00 PM, body: “There are no open trades. Thanks”. Read in the extraction only. https://epstein-data.com/EFTA01379375 p. 1. ↩

  38. Stewart Oldfield to Richard Kahn, “RE: Bank Balances”, RFC date header Tue, 26 Mar 2019 14:31:10 -0400, quoting Kahn’s message of Friday 8 March 2019 2:51 PM and Oldfield’s of 8 March 2019 2:46 PM. Read at the image at 400 dpi. The corpus date field for this document reads 2009; the RFC header printed at the head of the sheet reads 2019. https://epstein-data.com/EFTA01426549 p. 1. ↩

  39. A second Deutsche Bank database extract in the same series, carrying code 82289 against Stewart Oldfield’s name on Southern Trust Company, Inc. rows dated 2017. Read in the extraction only. https://epstein-data.com/EFTA01428201 p. 3. ↩

Backlinks