Shoko Chukin Bank's improper crisis-response loans. Staff discarded genuine trial balances and created altered versions themselves

When amounts on a paper trial balance were physically cut and pasted, changing only part of the document could break its arithmetic. The sum of individual lines might no longer equal total sales, or the current balance might no longer reconcile with the previous balance and the month’s debits and credits.

Yet the report says many supervisors checked only whether the figure needed for the crisis-response eligibility test matched the corresponding figure on the trial balance. They did not recalculate whether the document as a whole was internally consistent.

As a result, even trial balances whose internal arithmetic had been broken by cut-and-paste alteration could pass the loan approval process without the manipulation being detected.

The report also identified a different approach: instead of modifying paper, employees created whole trial balances themselves in applications such as Excel.

At the Kagoshima Branch, the Committee found that the existence of misconduct had become common knowledge among sales employees. Under pressure to meet allocated lending amounts, employees discussed at drinking gatherings and smoking areas how supporting evidence could be altered so that a case would satisfy the crisis-response criteria, and then put those methods into practice.

Employees could also see colleagues using scissors and glue on paper trial balances at their desks or creating documents in Excel on the same floor.

The Committee’s interviews found a culture in which employees who refused to manipulate documents and therefore could not meet their allocations were described in terms equivalent to “honesty doesn’t pay.”

Some employees did refuse, saying they disliked such conduct. The Committee concluded that widespread misconduct fostered a sense that “everyone is doing it,” weakening norms and helping further misconduct spread.

In examining the Kagoshima Branch, the Committee asked whether section managers, deputy branch managers and branch managers had failed to notice the misconduct, tacitly tolerated it, or actively directed it.

Employees who had engaged in misconduct told the Committee that supervisors must have noticed. Some said they were not explicitly told to alter documents, but were told to “do something” when a case did not meet the requirements.

One employee recalled telling a section manager that, including the time needed to fabricate the trial balance, the approval request would be submitted in another three hours—and receiving no response.

Most managers interviewed denied knowing of or participating in the conduct. The Committee said objective evidence did not allow it to establish tacit approval or instructions by particular managers. Considering the surrounding circumstances and testimony, however, it assessed that many section managers and deputy managers were likely to have noticed misconduct and tolerated it.

This episode did not itself involve a crisis-response loan. The Committee included it as another example of conduct that an employee attributed to the same pressure to meet lending allocations.

After a planned deal fell through near month-end, a sales employee persuaded another customer to use an overdraft facility. The explanation had not been fully understood, however, and on the repayment date at the start of the next month the account lacked enough money to cover the interest.

The employee panicked, paid the missing interest personally, and made the deposit while impersonating the customer.

A later customer complaint exposed the episode, and the employee and managers in the reporting line received an oral warning. The employee said that even after this incident, earlier manipulation of crisis-response loan materials continued.

Two separate interview accounts illustrate how different ideas of “normal” could develop from branch to branch.

An employee who transferred from Kagoshima said it was surprising to find that nobody at the new branch was altering documents. There, the section manager accompanied employees on customer visits and coached them on sales conversations. The employee regretted not having worked in that environment first and said that, while in Kagoshima, it had been difficult to know what was normal.

At Ikebukuro, another employee had the reverse reaction after encountering fabricated trial balances soon after moving to Tokyo.

That employee said the initial reaction was surprise that “this is what they do in Tokyo,” but, being new to the area and unsure of how things worked, came to accept it as simply the way things were done.

The two accounts are from different employees and branches, but together they show how local workplace practice could reset an employee’s sense of what was ordinary.

At the Ikebukuro Branch, employees did more than alter figures in Excel files received from customers. They reused one customer’s trial-balance data to create a document in another customer’s name, or built a new Excel trial balance from scratch to resemble a paper document received from a customer.

The report also describes steps designed to make discovery less likely. Employees kept multiple Excel formats and varied them by customer. Some took a fabricated or altered trial balance to a customer, gave little explanation, and obtained the company’s seal so the document could appear to have been customer-created.

Employees also discarded genuine paper trial balances received from customers so that the authentic documents would not remain alongside the fabricated or altered versions.

The Excel files were stored in shared folders. When another sales employee asked for a format that could be reused for a different customer, colleagues could identify the folder containing an earlier fabricated trial balance; the recipient then reused the format and entered figures considered suitable for the new case.

An Ikebukuro sales employee identified as E described to the Committee how trial-balance fabrication and alteration began for him.

E recalled being asked to “make one” while helping with a senior colleague’s loan approval rather than his own customer. Later, E said, he told his section manager that one of his own cases did not meet the crisis-response criteria and received a response pointing toward creating the necessary trial balance.

E told the Committee that the section manager effectively said there was no choice but to make one, and E took that as a sign that this was simply how the work was done.

Other employees in the same section also said the manager had explained that modifying a trial balance could make a case fit the criteria and that other sections or branches did the same. The report uses these statements, among other evidence, to analyze how fabrication and alteration spread within the section.

At another branch, multiple employees built new Excel trial balances resembling customer documents because changing only one part of a paper document could create arithmetic inconsistencies and make manipulation easier to detect.

The Excel files contained formulas.

If an employee changed sales to a figure that met the eligibility requirement, related items such as gross profit, operating profit and net income changed automatically as well, avoiding the vertical and horizontal mismatches created by simple cut-and-paste alteration.

Files created by one employee were shared with others and used in other improper cases. Genuine customer trial balances were shredded, which the report says made the fabricated documents harder to detect.

In an interview, one employee described creating different customer-specific formats as “art” and also described what had been done as a “perfect crime.”

Employee L at a branch identified as Y told the Committee that a conversation during a gathering of employees from the same intake became the trigger for misconduct. The conversation took place in the lobby of the training facility; it was not part of the formal training program.

When L complained that recent cases did not meet the crisis-response criteria, a peer said there was a way to make them fit and described alteration methods used by senior employees. Cut-and-paste techniques became part of the conversation.

After the gathering, L actually swapped the date portions of trial balances from two different years, making a business whose performance was improving appear to be deteriorating.

L said the method had not occurred to him before that conversation. In the later interview, he also described a sense that everyone did this sort of thing and that any punishment for being caught would be minor.

Through its employee questionnaire, the Third Party Committee learned that suspected fabrication and alteration of trial balances had surfaced at the Ikebukuro Branch in late 2014. Headquarters investigated 110 loan accounts at the time and ultimately treated both the alteration issue and crisis-response eligibility as presenting no problem, handling the matter as a violation of internal rules.

The Committee therefore re-examined those 110 accounts as part of its review of how the earlier case had been handled.

The Committee found misconduct in 106 of the 110 accounts and said that doubts about misconduct could not be dispelled for another three.

When Shoko Chukin subsequently re-evaluated crisis-response eligibility for those 106 accounts plus the three with unresolved doubts, a total of 109, it determined that 52 did not satisfy the requirements. The report treats headquarters’ earlier handling as a substantial investigation topic in its own right.

In January 2015, the Bank’s compliance office prepared a document setting out how investigators should question the employees who had created the trial balances. The Third Party Committee calls it the “leading-question paper.”

The document did not contain questions alone. It set out possible answer patterns and follow-up questions to use depending on the response. Members of the special audit team were asked to conduct the interviews in line with it.

In a passage urging accurate recollection, the paper even included a parenthetical message telling the interviewee, in substance, that the questions were being asked to keep the person from making a mistake and “to save you.”

A compliance-office manager involved in the document said it was intended to clarify the issues, not to force answers. The Committee disagreed with treating it as mere issue organization or anticipated Q&A, concluding that it was clearly aimed at obtaining answers that fit the scenario. After those answers were obtained, the compliance office concluded that alteration of trial balances had not been established for the 110 accounts.

After headquarters developed a method for treating all 110 accounts as satisfying the crisis-response criteria, the organizational finance division instructed branch managers to replace material in the loan files.

The original eligibility check sheets and supporting trial balances were removed from the loan documents and placed in a separate file for submission to headquarters. New check sheets were then prepared using the newly selected eligibility grounds.

The new sheets were given the dates of the original loan approvals and stamped by the original employees and approvers; some replacement evidence attached to them had not even existed when the loans were originally approved.

The Third Party Committee concluded that this work created the appearance that the 110 loan applications had been submitted from the outset on the basis of the replacement eligibility grounds.