Invest1 distinct publisher3 min readUpdated
Frankfurt prosecutors accuse four former staff of costing Hesse more than 20 million euros. The bank says it is not a party to the case.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
Frankfurt's General Prosecutor's Office has charged four former Commerzbank employees over Cum-Ex dividend-stripping trades executed in 2008, alleging more than 20 million euros in tax losses to the state of Hesse [1][2]. The relevant fact for operators is not the sum but the reach: positions taken in 2008, four years before the German legislature closed the loophole in 2012, are still generating individual criminal exposure [2][6][17].
The mechanics are well documented. Shares were bought and sold rapidly around a dividend date so that more than one party appeared to own the same stock, and each then claimed a refund on withholding tax that had been paid only once [5]. The activity clustered between roughly 2006 and 2011, with Germany absorbing the largest share of the damage and comparable schemes surfacing in Denmark, Belgium and other European markets [6][14].
The defendant list is the part worth studying if you run a cross-border team. Prosecutors name two British nationals aged 66 and 59, a 61-year-old German and a 60-year-old American [3]. Three were based in London and one in Frankfurt, meaning three quarters of the charged group sat outside German territory [4][18]. Prosecutors allege they worked across borders to exploit what were then gaps in German tax law, and according to the Cryptobriefing account of the filing, the precise role of each defendant will be tested at trial [15].
Commerzbank itself is not in the dock. The bank has said it is not a party to these proceedings and that the charges are directed at the four individuals rather than the institution [9], and it has maintained that it cooperates fully with authorities [8]. That is a narrower comfort than it sounds. German authorities searched the bank's offices in 2017 and 2019 as part of broader Cum-Ex investigations [7], and the pattern since has been to pursue people.
Bonn district court delivered Germany's first Cum-Ex criminal conviction in 2020 [10]. Hanno Berger, the tax lawyer described in the same report as a central architect of the structures, was convicted and sentenced in 2022 after years spent avoiding prosecution from Switzerland [11]. Recent convictions have produced prison sentences rather than fines [13]. Set against estimates that put the cumulative cost to European treasuries above 55 billion euros, contested as those figures are, the 20 million euros at issue here is roughly four hundredths of one percent of the total [12][16].
Two things to watch. First, whether the case reaches trial and how a German court apportions individual responsibility among four people spread across two financial centres and three nationalities [15][3][4]; the London-based defendants may not move at the same pace as the Frankfurt one [4]. Second, whether further charges emerge from the same 2006 to 2011 window, which is where the volume was [6]. For compliance functions, the practical read is that an institution's cooperation posture does not settle a former employee's personal position on trades this old [9][8].
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
The Frankfurt General Prosecutor's Office has charged four former Commerzbank AG employees for their alleged roles in orchestrating Cum-Ex trades.
The charges accuse the group of causing more than 20 million euros in tax losses to the state of Hesse through trades executed in 2008.
The defendants include two British nationals, ages 66 and 59, a 61-year-old German national, and a 60-year-old American.
Three of the defendants were based in London, with one operating out of Frankfurt.
In Cum-Ex, traders rapidly bought and sold shares in the narrow window around a dividend payout, creating the appearance that multiple parties owned the same shares simultaneously, with each party claiming a refund on dividend withholding tax that was only paid once.
The scheme peaked roughly between 2006 and 2011, before German legislators closed the loophole in 2012.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single-outlet account of untested allegations
Every factual element traces to one article from one publisher. The core charging facts are attributed to prosecutors but no indictment, case number, prosecutor statement or court filing is cited, defendants are unnamed, and the only corroborating voice is Commerzbank's own statement. Historical anchors such as the 2020 Bonn conviction and Berger's 2022 sentencing are checkable but likewise unsourced here, and the aggregate damage figure is explicitly hedged.
No adoption signal in scope
This is a criminal-charging story. The supplied source records no release, deployment, benchmark, pricing or usage disclosure, and no adoption observations could be drawn without inventing facts, so adoption is not measurable here.
Mildly inflated framing around a modest, unproven case
The article's own hedges pull it close to alignment: it says the case sum is a fraction of the total, that precise figures are contested, and that roles will be tested at trial. The inflation comes from scale juxtaposition, presenting a 20 million euro unproven case against a 55 billion euro unattributed aggregate and 'Europe's largest tax fraud scandal' framing, plus recounting 2017 and 2019 raids alongside a case the bank says it is not party to.
Self-interested statements and one outlet outside its core beat
Two incentive layers are visible in the material. Commerzbank's cooperation-and-transparency posture and its 'not a party' line are its own characterisations of matters affecting its reputation and liability, and prosecutors' allegations are advanced by a charging authority before any trial. Separately, a crypto-focused publisher covering a German banking prosecution carries a scandal-scale framing incentive, evident in the unattributed aggregate figure.
Low to moderate: one publisher, allegations unproven
The narrative is internally consistent and appropriately hedged, and the historical Cum-Ex framing is plausible, which supports moderate confidence in the broad picture. But a single publisher, unnamed defendants, no primary court record, no adoption dimension and an explicitly contested headline number cap confidence well below the midpoint.
Distinct publishers with included, body-backed reporting in this cluster.
cryptobriefing.com
1 article · August 20, 2026