Earlier this year, the US Department of Justice (DOJ) entered into its 80th, and final, non-prosecution agreement with a Swiss bank as part of its groundbreaking program to combat offshore tax evasion in Switzerland and beyond. The DOJ has collected over $1.36 billion in penalties from these banks, as well as detailed information to trace the assets from the banks’ US-related accounts to other financial institutions worldwide.

On August 29, 2013, the DOJ and the Swiss Federal Department of Finance announced in a joint statement the Program for Non-Prosecution Agreements or Non-Target Letters for Swiss Banks (Swiss Bank Program or Program). The Program provided Swiss banks with a way to resolve potential liabilities in the US for tax-related criminal offenses and be protected from prosecution, in exchange for the disclosure of certain information related to undeclared accounts in which US taxpayers had a direct or an indirect interest and the payment of penalties. (See US Dep’t of Justice, Joint Statement Between the US Dep’t of Justice and the Swiss Fed. Dep’t of Finance (Aug. 29, 2013), available at justice.gov; Justice News, US and Switzerland Issue Joint Statement Regarding Tax Evasion Investigations (Aug. 29, 2013).)

The Program divided the Swiss banks into the following categories:

  • Category 1 Banks. This included banks that the DOJ was already criminally investigating and were ineligible for the Program.
  • Category 2 Banks. This included banks that might have committed tax-related offenses and could request non-prosecution agreements (NPAs).
  • Category 3 Banks. This included banks that did not believe they had committed any tax-related offenses and could apply for non-target letters.
  • Category 4 Banks. This included banks that believed they had met the relevant requirements under the Foreign Account Tax Compliance Act (FATCA) and could apply for non-target letters.

(See US Dep’t of Justice, Program for Non-Prosecution Agreements or Non-Target Letters for Swiss Banks (Swiss Bank Program), at pt. I.A (Aug. 29, 2013), available at justice.gov.)

Against this backdrop, this article explores the details of the Program for Category 2 Banks, including:

  • The disclosure requirements and penalties imposed for a Category 2 Bank to receive an NPA.
  • The issues raised by the Program’s formulaic approach, in light of the Category 2 Banks’ wide range of conduct and vastly different access to account information.
  • The conflicting perspectives of the DOJ and the Swiss Financial Market Supervisory Authority (FINMA), including the tension between the Program’s requirements and Swiss privacy laws.
  • The effectiveness of the Program and whether it achieved its goals.
  • The next steps for the DOJ in its pursuit of offshore tax enforcement, including the potential for similar amnesty programs in other countries.


Any Swiss bank that was not under criminal investigation and had reason to believe it might have committed tax-related or monetary transaction-related offenses under US law was eligible to request an NPA as a Category 2 Bank under the Swiss Bank Program. The deadline for submitting an NPA request was December 31, 2013. (See Swiss Bank Program, at pt. II.) 106 banks submitted requests under Category 2. By the time all of the Category 2 Banks had resolved their situations under the Program on January 27, 2016, 80 banks had executed NPAs with the DOJ. (See Offshore Tax Evasion: The Effort to Collect Unpaid Taxes on Billions in Offshore Accounts: Hearing Before the Permanent Subcomm. on Investigations, S. Comm. on Homeland Sec. and Gov’t Affairs, at 5 (Feb. 26, 2014) (Joint Statement of James M. Cole, Deputy Attorney General, and Kathryn Keneally, Assistant Attorney General, Tax Division); Justice News, Justice Dep’t Announces Final Swiss Bank Program Category 2 Resolution with HSZH Verwaltungs AG (Jan. 27, 2016).)

To receive an NPA, a Category 2 Bank was required to provide detailed information on any US-related accounts and pay penalties, which could be reduced by the value of accounts subject to mitigation under the Program.

Additionally, although the NPAs have four-year terms, the banks must continue cooperating with the DOJ in its ongoing investigations of tax evaders and others involved with the banks’ US-related accounts.


The Program required Category 2 Banks to provide, among other things:

  • Information on the structuring of cross-border business for US-related accounts, including the names and positions of the bank employees involved.
  • The total number of US-related accounts and the total maximum aggregate value of those accounts in the following categories: zzUS-related accounts that existed on August 1, 2008;
  • US-related accounts that were opened between August 1, 2008 and February 28, 2009; and
  • US-related accounts that were opened after February 28, 2009.
  • For each US-related account:zzthe maximum account value;
  • the name of any relationship manager, client advisor, or other individual or entity functioning in a similar capacity associated with the account; and
  • all information about incoming and outgoing transfers.

(See Swiss Bank Program, at pt. II.D.1-2.)

The three time periods were intended to correlate to public awareness of the DOJ’s investigations into offshore tax evasion in Switzerland, in particular the DOJ’s investigation of UBS and its subsequent entry into a deferred prosecution agreement with UBS (see Justice News, UBS Enters into Deferred Prosecution Agreement (Feb. 18, 2009)).

Gathering information about incoming and outgoing transfers was a time-consuming and expensive process for the banks. The Program required the banks to retain a qualified attorney or accountant as an independent examiner to verify each bank’s due diligence standards and findings (see Swiss Bank Program, at pt. II.D.3). The production and certification of the transfer information was intended to allow the DOJ to trace the assets of the so-called “leavers,” who closed their accounts and transferred their assets to other banks, each of which then became a potential target of additional DOJ investigations. Under Swiss law, the banks were obligated to notify the identified banks that information about the transfers into and out of accounts at those banks would be disclosed to the DOJ (see below Swiss Privacy and Bank Secrecy Laws).


A bank’s agreement to pay certain penalties was a key component of an NPA request. The Category 2 Banks had to pay:

  • 20% of the maximum aggregate value of all US-related accounts in existence on August 1, 2008.
  • 30% of the maximum aggregate value of all US-related accounts opened between August 1, 2008 and February 28, 2009.
  • 50% of the maximum aggregate value of all US-related accounts opened after February 28, 2009.

(See Swiss Bank Program, at pt. II.H.)

For each of these three groups of accounts, the maximum aggregate dollar value was calculated when the value of all of those accounts (for example, all accounts opened after February 28, 2009) was at its highest point, typically using a single end-of-month date (see US Dep’t of Justice, The Tax Division’s Further Comments About the Program for Non-Prosecution Agreements or Non-Target Letters for Swiss Banks (June 5, 2014), available at justice.gov). This resulted in some extremely high penalties (see Box, Penalty Statistics).


Certain types of accounts could be mitigated and therefore excluded from the penalty calculation. However, exactly how banks should prove mitigation for these accounts was unclear when the Program was announced (see below Flaws in the Mitigation Concept). The DOJ issued guidance in a March 2015 letter to the Program’s participants, after the deadline passed for banks to submit mitigation proof. It instructed that an account’s status could be confirmed to credit the bank with penalty mitigation if the bank provided the DOJ with either:

  • Client information with which the DOJ could confirm the relevant taxpayer’s position.
  • An unredacted document showing the verified facts about the client’s account.

A mitigation committee, comprised of a group of senior Tax Division attorneys, oversaw the Program and made the ultimate decision on the amount of each bank’s penalty.

The following types of accounts were eligible to be mitigated:

  • Accounts that were not undeclared. For these accounts, the banks could provide a copy of either:
    • a Report of Foreign Bank and Financial Accounts (FBAR); or
    • an amended or original tax return filed in any year prior to August 29, 2013.
  • Accounts that the banks timely declared. For these accounts, the banks could provide a copy of a Form 1099 that was filed with the Internal Revenue Service (IRS) prior to August 29, 2013.
  • Accounts for which the bank encouraged the client to enter into the US Offshore Voluntary Disclosure Program (OVDP). For these accounts, the banks could provide unredacted documents showing that the account holder participated in the OVDP and some type of evidentiary support that the bank encouraged the account holder to enter the OVDP, for example:
    • an affidavit from the account holder or the responsible bank employee;
    • a pre-Program bank policy instructing bank employees to encourage clients to enter the OVDP; or
    • a letter sent after the Program began encouraging clients to enter the OVDP.

To receive credit, the DOJ had to be able to independently confirm the existence and accuracy of the relevant documents with the IRS.

Alternatively, the banks could provide the DOJ with the name and taxpayer identification number of the relevant US clients, which the DOJ would then use to confirm each taxpayer’s status.

Finally, if the bank could not provide the above information, but produced other relevant evidence, the DOJ team investigating the bank could recommend a penalty reduction to the mitigation committee. (See Letter from US Dep’t of Justice, Tax Division, to Program Participants (Mar. 6, 2015) (on file with the DOJ).)

After the mitigation committee reviewed the mitigation evidence, the DOJ provided the banks with the committee’s determination of the applicable penalties. The DOJ did not disclose details on which accounts received mitigation credit or how the committee calculated the final penalty amount. The banks typically received about three days to either accept the DOJ’s penalty calculation and enter into an NPA, or withdraw from the Program.


An analysis of the banks that received an NPA shows that there is no correlation between participation in the Program as a Category 2 Bank and the conduct of any particular bank. Moreover, although the Category 2 Banks exhibited a wide range of conduct, the Program’s penalty system was predetermined and objective (see above Penalties), leading to disproportionate effects and unjust results.


Within Category 2, there were many banks that had almost no bad conduct and others that had extraordinarily problematic conduct. The variation was huge. Further, an examination of the Statements of Facts (SOFs) of some Category 1 Banks and some Category 2 Banks reveals similar conduct. Many of the Category 1 Banks that settled with the DOJ seem to have had bad luck in being selected for an investigation when compared with the Category 2 Banks whose SOFs exhibited bad behavior.

Most of the Category 2 SOFs contain relatively benign facts outlining an entity that would have little chance of being prosecuted. Many of the Category 2 Banks, for example:

  • Did not believe their actions were wrong given that the qualified intermediary agreements they signed with the IRS clearly indicated that they could keep undeclared money as long as they reported any US investments.
  • Reviewed the UBS SOF and deferred prosecution agreement and concluded that their own behavior was different because, for example:
    • they did not have US-specific banking teams;
    • they did not intentionally target US clients;
    • their relationship managers did not travel to the US; and
    • their relationship managers did not assist clients with secretly moving assets.
  • Misinterpreted the DOJ’s position and took in certain US clients.
  • Started exiting US clients when they realized there was a problem.

By contrast, some Category 2 Banks actively engaged in behavior to help clients avoid detection by US authorities. These banks, for example:

  • Processed withdrawals and then maintained the funds in a safe deposit box at the bank.
  • Maintained segregated accounts in the names of insurance companies for which the bank was aware that the policyholders or premium payers were US persons.
  • Allowed accounts to be closed with cash withdrawals.
  • Converted client assets to precious metals, such as gold or silver.


The Program’s formulaic approach to penalties did not reflect the severity of a bank’s conduct. For example, an account where a dual-national account holder lied to the bank and did not disclose her US citizenship would be treated the same as an account where the bank assisted a US client in setting up a foreign entity as the account holder to conceal US ownership.

Additionally, the procedures for mitigating penalties by providing documentation that the clients were US tax compliant were much more difficult for some banks than others. Category 2 Banks that were disadvantaged under the Program’s penalty structure included:

  • Banks that exited their US clients early. If the clients and their assets were no longer with a bank, it was much harder to obtain documentation of tax compliance. These former clients had no incentive to assist the bank, and convincing them to disclose their undeclared and untaxed assets to the IRS proved difficult. Several banks paid penalties where more than half of the penalty amount was attributable to a single account that had been closed prior to the Program and the client could not be reached. By contrast, banks that were “bad” and continued to service US clients as late as 2013 had far more leverage to convince these clients to self-disclose or provide documentation that the banks needed for penalty mitigation.
  • Larger banks with many smaller US clients. There is a perception that most banks in Switzerland service only a small number of ultra-high net worth clients. However, some banks, including larger universal banks with private banking businesses in Switzerland, have many relatively low-value, US-related accounts. It would have been very challenging for these banks to negotiate arrangements with thousands of clients to provide documentation of US tax compliance. By contrast, banks that had a small number of high-value accounts were able to collect the necessary documentation and mitigate their penalty more easily.
  • Banks that were unwilling or unable to pay for information from clients. A bank that decided it would not invest in penalty mitigation had a far higher penalty when compared to a bank that made significant payments to current or former clients to provide documentation showing US tax compliance. Indeed, several banks paid more to obtain this information than they did in penalties, in an effort to avoid the negative public perception of a high penalty. By contrast, the ownership structure of certain publicly owned banks prevented them from making payments to current or former clients, and other banks had fewer concerns about the publicity surrounding the penalty amount and instead wanted an NPA executed as quickly as possible.

Another issue with the Program’s penalty mitigation framework was that the DOJ did not provide any guidance on how mitigation would be applied until after the deadline for all required data to be submitted had passed. For example, the DOJ required that it be able to independently confirm the veracity of the mitigation data the banks provided, which was not part of the Program’s initial requirements (see Letter from US Dep’t of Justice, Tax Division, to Program Participants (Mar. 6, 2015) (on file with the DOJ)). Therefore, it was extremely challenging for the banks’ attorneys to plan a proper mitigation strategy when the banks were collecting information. This was a significant issue that created tension between the banks and their advisors. The banks expected unequivocal answers, but the Program’s lack of clarity and the DOJ’s changing rules made certainty or even predictability nearly impossible.


A country-specific program with country-specific terminology like the Swiss Bank Program is a unique concept. This type of program has never been implemented before and it is unclear whether it will be again (see below Potential Bank Programs in Other Countries). When it was first announced, the Program raised confusion among banks and their attorneys because, on the one hand, the Program contained elements that gave due consideration to Swiss law, but on the other hand, the DOJ administered it. This led to many banks taking different approaches.

Ultimately, this was a US program under US law, with Swiss elements. Once the initial negotiations were completed, the DOJ administered the Program with limited influence from the Swiss authorities. However, Swiss data protection and bank secrecy laws played a significant role in the Program and caused tension between the Swiss banks and the DOJ over what information could be disclosed, particularly relating to employees, non-clients, and other third parties.


Because there was not a preset manner in which the Program would be administered and no history to draw on, much was left to the interpretation of those involved. Although it was a voluntary program, the DOJ’s perspective was that any bank entering the Program:

  • Would not receive better or worse treatment than had it not entered the Program.
  • Was immediately subject to a comprehensive review of all issues.
  • Should be viewed with skepticism on all issues as if the bank had either chosen to self-disclose or had been caught by an investigation committee.

The DOJ’s approach was unsurprising given that the Category 2 Banks admitted to being involved in a possible crime. However, the banks and their attorneys expected to receive some amount of credit for their willingness to participate in the Program and believed it would be less onerous than going through a full self-disclosure procedure. This perspective conflicted with the DOJ’s position.

Another issue arose because the DOJ underestimated the number of banks that would enter the Program, which resulted in a staffing problem. To effectively run the Program, the DOJ Criminal Tax Division required assistance from the DOJ Civil Tax Division. As a result, the Program was administered in a way that deviated slightly from the standard approach to handling criminal tax matters.


FINMA, which regulates all Swiss banks, took the position that each bank had to decide for itself whether to enter the Program. FINMA then monitored the banks’ participation in the Program to ensure that the banks acted reasonably. If a bank chose to withdraw from the Program, it needed to provide FINMA with its rationale. Occasionally, when a bank settled, FINMA inquired about the settlement and required the bank to complete certain questionnaires. However, FINMA’s ability to influence the process and help facilitate resolutions between the banks and the DOJ was relatively limited.


The political context in which the Program arose was an important factor in its development and administration. In the two or three years before the Program’s announcement, some of the Swiss banks already under investigation had disclosed substantial amounts of information on US clients to the US government. While the number of negatively affected individuals represented a small percentage of the total number of individuals whose information was disclosed, the privacy concerns became a significant political issue.

As a result, when the Swiss Federal Department of Finance allowed banks to participate in the Program, the Swiss regulators became extremely concerned that the Program would also become a political issue in Switzerland if data on a significant number of individuals was to be disclosed. Therefore, the Swiss regulators stressed that a bank’s disclosure of any employee or third-party data without explicit or silent consent would be a violation of the bank’s Article 271 waiver to participate in the Program (see Schweizerischees Strafgesetzbuch [StGB] [Criminal Code] Dec. 21, 1937, SR 311.0, art. 271 (Switz.); Swiss Model Order of July 3, 2013 ¶ 1.4, available at www.news.admin.ch).

This created a novel and complex situation. In its joint statement with the DOJ announcing the Program, the Swiss Federal Department of Finance represented that Swiss law would “permit effective participation by the Swiss Banks on the terms set out in the Program” (US Dep’t of Justice, Joint Statement Between the US Dep’t of Justice and the Swiss Fed. Dep’t of Finance, at ¶ 2). However, under the Swiss Federal Data Protection Act, third parties and employees could block the disclosure of their data to US authorities by applying for provisional relief from competent Swiss courts to prevent the participating banks from providing part of the information the Program required (see Systematische Sammlung des Bundesrechts [SR] [Classified Compilation of Swiss Internal Laws] June 19, 1992, SR 235.1, art. 6 (Switz.)). The DOJ probably did not appreciate that this component of Swiss law would remain applicable, and there seemed to be different understandings of the concept of “effective participation” by the US and Swiss sides. The Swiss data protection laws, as opposed to, and in combination with, its bank secrecy laws, led to some of the ensuing confusion.

Civil lawsuits in Switzerland brought to block the disclosure of certain employee or third-party data are still pending, despite the banks’ execution of NPAs. If a bank fails to continue litigating those cases against the objecting employees or third parties, the DOJ could view the bank as having violated the NPA’s terms, thereby voiding it.

Finally, it is worth noting that the preservation of banking secrecy has been a challenge since the negotiation and announcement of the Program. The Swiss government made the first concession when it accepted that the Program would enable participating banks to cooperate with US authorities in the preparation of treaty requests, which, albeit in compliance with Swiss law, could eventually result in the disclosure of information protected by Swiss bank secrecy laws against the wishes of the concerned clients. The Swiss government also accepted that banks could rely on waivers from their clients to disclose account information to the DOJ. However, some clients were reluctant to waive banking secrecy, even though they had provided proof of tax compliance. The DOJ did not react favorably to the production of redacted documents, and this complicated the penalty mitigation process.


It is too early to determine whether the Swiss Bank Program achieved its objectives. As detailed above, the Program did not necessarily provide a better resolution for the banks than if they had disclosed and cooperated with the DOJ outside of the Program. Further, while the DOJ collected over $1.36 billion, in many instances, the penalties imposed did not correlate in proportion to each bank’s conduct, whether good or bad.

Banks that received an NPA can, in some ways, move forward with a clean slate. However, many banks chose to withdraw from the Program and some unresolved issues remain.


Entering into an NPA, while important for the US side, also provided value to the Swiss banks. For the next 10 to 15 years, anytime a potential acquisition or merger with another bank arises, a potential acquirer will question if the potential target participated in the Program and received an NPA. The failure to have an NPA will make it much more difficult to sell or merge a Swiss bank. Moreover, it might be harder for a bank to deal with custodians in the future if it did not participate in the Program.


As mentioned above, 106 banks had elected to participate in the Program, but only 80 remained in the end. About 25% of the banks withdrew. While each bank made its own decision and there is no one reason for withdrawal, there were certain common trends, including:

  • The lack of time to make a decision. The banks received only four months to decide whether to participate in the Program. No bank, other than the largest banks, had ever conducted the type of investigation that was necessary to fulfill the Program’s requirements. Many of the banks had a limited ability to generate and gather the required data. There were also many fundamental questions unanswered about the level of review the Program required. Additionally, one month before the Program’s deadline for entry, the then-head of FINMA wrote an article in the preeminent Swiss newspaper, stating that, when in doubt, a bank should go into the Program under Category 2 (see Patrick Raaflaub, Entscheidende Phase für Schweizer Banken, Neue Zürcher Zeitung (Nov. 29, 2013), available at nzz.ch). As a result, many banks joined the Program because they did not have enough information to determine if there was sufficient criminal conduct to justify participating in the Program and did not want to risk losing the opportunity.
  • The DOJ’s changing requirements and the perception of a bait-and-switch. Most banks thought the Program would be a straightforward and objective arrangement, where if a bank took certain actions, it would receive an NPA and have a clear-cut resolution. However, once the Program started, the DOJ did not adhere to its initial framework and imposed additional requirements. This resulted in some banks choosing to exit the Program after perceiving a bait-and-switch. Banks also withdrew because they concluded that the risk of an indictment or other punishment was disproportionate to the difficulties they encountered being part of the Program and meeting the DOJ’s new conditions.
  • The DOJ’s inflexibility on penalty calculations. The DOJ was unwilling to consider any ability-to-pay arguments. Banks with weak financial status that wanted to remain in the Program might have been unable to absorb the penalties. In this case, a bank would have no incentive to stay in the Program and would likely trade the risk of going bankrupt against that of potential prosecution by the DOJ.


Because not all Swiss banks entered the Program, and many withdrew from it, there are still banks that the DOJ could investigate in the future. Therefore, the Program did not fully achieve its goal to move the US tax evasion hunt forward and out of Switzerland to other locations.

Additionally, many Swiss banks are worried that the DOJ will investigate certain employees and third parties with which the banks conduct business. This kind of paralysis makes it extremely difficult for a bank to continue operating its business.


Offshore tax enforcement remains a top priority for the DOJ. In remarks made to the Federal Bar Association Tax Law Conference, Acting Assistant Attorney General Caroline D. Ciraolo confirmed that “investigations of both individuals and entities are well beyond Switzerland at this point, and no jurisdiction is off limits” (Justice News, Acting Assistant Attorney General Caroline D. Ciraolo Delivers Remarks at the Fed. Bar Ass’n Tax Law Conf. (Mar. 4, 2016)).

In a recent interview, Ciraolo also stated that the DOJ is following the assets that flowed from Switzerland to the following jurisdictions, among others:

  • The British Virgin Islands.
  • The Cayman Islands.
  • The Channel Islands.
  • Hong Kong.
  • Israel.
  • Lichtenstein.
  • Luxembourg.
  • Panama.
  • Singapore.

Ciraolo noted that other countries have inquired about whether the DOJ would implement more programs similar to the Swiss Bank Program, however, she did not confirm that it would. Instead, she cautioned individuals and entities in other jurisdictions that assisted clients with evading US taxes to disclose immediately and not wait to see if a potential country-specific program provides a better resolution. (See Jeremy H. Temkin, DOJ Tax Division Today: Interview with Acting Assistant Attorney General, N.Y.L.J. (Mar. 23, 2016).)

Days after Ciraolo’s interview, the DOJ announced that two Cayman Island financial institutions pleaded guilty to conspiring with US taxpayer clients to hide more than $130 million in offshore accounts from the IRS and evade taxes on the income earned in those accounts. The DOJ announced that these were the first convictions of non-Swiss banks for tax evasion conspiracy. (See Justice News, Two Cayman Island Fin. Insts.

Plead Guilty in Manhattan Fed. Court to Conspiring to Hide More Than $130 Million in Cayman Bank Accounts (Mar. 9, 2016).)

Local laws might preclude banks in other countries from voluntarily disclosing to the DOJ any conduct that occurred before the banks entered into FATCA, which requires banks to report to the IRS information about accounts held by US taxpayers or foreign entities in which US taxpayers hold a substantial ownership interest. Banks that comply with FATCA should not encounter any issues related to current US clients. However, to address pre-FATCA conduct related to US clients that did not disclose their accounts to the IRS, the US government will likely have to engage with other governments to allow their banks to disclose this information.

* This article was first published in Practical Law by Thomson Reuters