Restructuring case2020

You are a management accountant at a firm that is undertaking a restructuring of their sales department. As part of this process, you will have to estimate the restructuring costs for the company. You are preparing for a meeting with the division manager to work through the plan to identify the employees who will be terminated. Your manager is asking for your advice, given your background in data analytics, on how to be use information in web browser data to pick between employees that have had equivalent performance evaluations. The manager has heard a lot about how big data, especially data collected from employees, is being used in the new field of Human Resource (HR) analytics. Your manager has also asked if you think it would be valuable to screen email data of the employees being considered for termination.

Broadly speaking this case is about the ethical use of big data and how accounting judgments may interact with other business decisions. Estimates of the expense of terminating employees as part of a disposal or restructure include a large amount of judgement when managers provide choices between voluntary and involuntary termination benefits as well as termination benefits that require continued service to a future service date. The CFO is hoping that your team will be able to work together towards minimizing the costs of restructuring in the current period by deferring some of the costs to the future. The CFO believes that this can be achieved by offering a mixture of short-term and long-term termination benefit packages to employees, and wants you to consider offering lower short-term termination benefits to individuals expected to leave voluntarily. The big data “revolution” has provided tools and insight into how to collect data from individuals, especially web based data, often without individuals being aware of how much data is being collected. This raises ethical questions relating to the use of individual data in sensitive contexts such as offering different retrenchment packages. You will have to tackle this ethical question as you prepare for the meeting and cover your beliefs on the ethics of the use of web-browser data, as well as whether email data should also be used to identify the employees targeted for termination.

You will have the opportunity to discuss the rationale for and the practicalities of your conclusions in your response and in class. Your conclusions are to be backed up with a discussion of a hypothetical analysis of individual web-browser history data. Note that what some find as invasive or over-monitoring, others may think is perfectly fair. These are evolving ethical issues that need to be discussed.

Background

Accounting for restructuring

Accounting for restructuring is covered in ASC ¶420 Exit or Disposal Cost Obligations. The transactions that can be recognized as part of the restructuring costs are discussed in ASC ¶420-10-15, §3, as follows:

“The guidance in the Exit or Disposal Cost Obligations Topic applies to the following transactions and activities:

a. Termination benefits provided to current employees that are involuntarily terminated under the terms of a benefit arrangement that, in substance, is not an ongoing benefit arrangement or an individual deferred compensation contract (referred to as one-time employee termination benefits)

b. Costs to terminate a contract that is not a capital lease (see paragraphs 420-10-25-11 through 25-13 for further description of contract termination costs and paragraph 840-30-40-1 for terminations of a capital lease)

c. Costs to consolidate facilities or relocate employees

d. Costs associated with a disposal activity covered by Subtopic 205-20

e. Costs associated with an exit activity, including exit activities associated with an entity newly acquired in a business combination or an acquisition by a not-for-profit entity.”

As our focus is on employee termination, we are most interested in the accounting for termination benefits, which the FASB defines as “Benefits provided by an employer to employees in connection with their termination of employment. They may be either special termination benefits offered only for a short period of time or contractual benefits required by the terms of a plan only if a specified event, such as a plant closing, occurs” (ASC ¶712-10-20). Often these are cash payments that occur at the end of the employee’s employment period with the firm and are often labelled severance packages (or redundancy payouts or lay-off payoffs).

Accounting for termination benefits

The guidance in the ASC suggests that the recognition of a liability as part of the restructuring charge is not at the time of the commitment by management to undertake the restructuring (see ASC ¶420-10-25-2). In the case of accounting for termination benefits, the liability is triggered at the time of the communication date, defined as meeting the criteria in ASC ¶420-10-25-4:

“An arrangement for one-time employee termination benefits exists at the date the plan of termination meets all of the following criteria and has been communicated to employees (referred to as the communication date):

a. Management, having the authority to approve the action, commits to a plan of termination.

b. The plan identifies the number of employees to be terminated, their job classifications or functions and their locations, and the expected completion date.

c. The plan establishes the terms of the benefit arrangement, including the benefits that employees will receive upon termination (including but not limited to cash payments), in sufficient detail to enable employees to determine the type and amount of benefits they will receive if they are involuntarily terminated.

d. Actions required to complete the plan indicate that it is unlikely that significant changes to the plan will be made or that the plan will be withdrawn.”

In the case that operations are ceased immediately, the restructuring expense and cash payments will coincide, and there is little need for estimation or judgement. There are more complex cases that do require judgement, for example:

In the case where the company requires operations to continue for some period of time in the future, they may announce that the operations will be ceased on a future date, and employees still with the company on that date will receive a termination benefit. Employees who choose to leave the company in the interim period will not receive the termination benefit. This event triggers the estimation of the fair value of the liability.
In addition, a firm may only require 50% of the workforce between the time of the announcement and the ceasing of operations. In this case, the firm may offer voluntary termination benefits to employees who are will to leave at the time of the announcement. This triggers a conditional liability as well as fair value estimates in some cases.

In the cases that require judgment, often the estimate involves considering how many individuals will be subject to the various termination benefits being offered by the firm. The accounting treatment is provided in the following two examples from ASC ¶420-10-55:

“Example 2: One-Time Employee Termination Benefits—Stay Bonus-Future Service Required

55-4 This Example assumes that an entity has a one-time benefit arrangement established by a plan of termination that meets the criteria in paragraph 420-10-25-4 and has been communicated to employees.

55-5 An entity plans to shut down a manufacturing facility in 16 months and, at that time, terminate all of the remaining employees at the facility. To induce employees to stay until the facility is shut down, the entity establishes a one-time stay bonus arrangement. Each employee that stays and renders service for the full 16-month period will receive as a termination benefit a cash payment of $10,000, which will be paid 6 months after the termination date. An employee that leaves voluntarily before the facility is shut down will not be entitled to receive any portion of the termination benefit. In accordance with paragraph 420-10-25-9, a liability for the termination benefits would be measured initially at the communication date and, in accordance with paragraph 420-10-30-6, based on the fair value of the liability as of the termination date and recognized ratably over the future service period. The fair value of the liability as of the termination date would be adjusted cumulatively for changes resulting from revisions to estimated cash flows over the future service period, measured using the credit-adjusted risk-free rate that was used to measure the liability initially (as illustrated in this Example).

55-6 The fair value of the liability as of the termination date is $962,240, estimated at the communication date using an expected present value technique. The expected cash flows of $1 million (to be paid 6 months after the termination date), which consider the likelihood that some employees will leave voluntarily before the facility is shut down, are discounted for 6 months at the credit-adjusted risk-free rate of 8 percent. In this case, a risk premium is not considered in the present value measurement. Because the amounts of the cash flows will be fixed and certain as of the termination date, marketplace participants would not demand a risk premium.

55-7 Therefore, a liability of $60,140 would be recognized in each month during the future service period (16 months).

55-8 After eight months, more employees than originally estimated leave voluntarily. The entity adjusts the fair value of the liability as of the termination date to $769,792 to reflect the revised expected cash flows of $800,000 (to be paid 6 months after the termination date), discounted for 6 months at the credit-adjusted risk-free rate that was used to measure the liability initially (8 percent). Based on that revised estimate, a liability (expense) of $48,112 would have been recognized in each month during the future service period. Thus, the liability recognized to date of $481,120 ($60,140 × 8) would be reduced to $384,896 ($48,112 × 8) to reflect the cumulative effect of that change (of $96,224). A liability of $48,112 would be recognized in each month during the remaining future service period (8 months). Accretion expense would be recognized after the termination date in accordance with the guidance beginning in paragraph 420-10-35-1 and in paragraph 420-10-45-5.

Example 3: One-Time Employee Termination Benefits—Voluntary and Involuntary Benefits Offered

55-9 This Example assumes that an entity has a one-time benefit arrangement established by a plan of termination that meets the criteria of paragraph 420-10-25-4 and has been communicated to employees.

55-10 An entity initiates changes to streamline operations in a particular location and determines that, as a result, it no longer needs 100 of the employees that currently work in that location. The plan of termination provides for both voluntary and involuntary termination benefits (in the form of cash payments). Specifically, the entity offers each employee (up to 100 employees) that voluntarily terminates within 30 days a voluntary termination benefit of $10,000 to be paid at the separation date. Each employee that is involuntarily terminated thereafter (to reach the target of 100) will receive an involuntary termination benefit of $6,000 to be paid at the termination date. The entity expects all 100 employees to leave (voluntarily or involuntarily) within the minimum retention period. In accordance with paragraphs 420-10-25-6 through 25-8, a liability for the involuntary termination benefit (of $6,000 per employee) would be recognized at the communication date and, in accordance with paragraphs 420-10-30-4 through 30-6, measured at its fair value. In this case, because of the short discount period, $600,000 may not be materially different from the fair value of the liability at the communication date. As noted in paragraph 420-10-25-10, a liability for the incremental voluntary termination benefit (of $4,000 per employee) would be recognized in accordance with paragraph 712-10-25-1 through 25-3 (that is, when employees accept the offer).”

Accounting and corporate actions: The accounting for restructuring costs could induce managerial overestimation of the costs relating to the restructure. For example, Al Dunlap of Sunbeam was criticized for using the flexibility in accounting judgment to overestimate the restructuring costs significantly and reverse the liability in future years (an extreme case of the overestimation in ASC ¶420-10-55-8 above). In cases where the goal is to shift the expense to the future, however, it is not that easy, as many underestimates could result in SEC comment letters or restatements. With limited flexibility in accounting measurement, managers may alter corporate actions to achieve a desired earnings outcome. Often this behavior is labeled “real activities” management. In terms of a restructuring, management may consider selectively targeting individuals for redundancy.

Business Analytics

“Big data”

Attention to the uses of “big data” for business applications has increased dramatically over the past years. Big data is often defined in terms of the four V’s: Volume: the amount of data that can now be processed is in terabytes, petabytes, zettabytes versus megabytes and gigabytes in the recent past. Variety: new forms of data are now being processed typically referred to as unstructured (text, voice, video) versus the reliance on a structured form like a relational database. Velocity: the speed at which data is being created due to streaming sources versus a more static data environment in the recent past. Veracity: data is often more untrusted and unclean versus the trusted and cleansed datasets created in the past. The IT research firm Gartner Inc., describes big data as “information assets that demand cost-effective, innovative forms of information processing that enable enhanced decision making and process automation.”

The most well-known business applications of big data are often centered on customers (such as the customer suggestion engines used by Netflix Google and Amazon, and the data-driven targeted advertising by Facebook and Target). But big data can be used in many other environments. For example, Deloitte reports that 79% of respondents to a survey in the US stated that predictive models for HR are important or very important, with similar responses from executives in other countries (Deloitte 2016).

Predictive analytics for HR purposes

People analytics: is the analysis of individual-level HR and other business data to predict uncertain behavior, such as the probability of an individual employee leaving the firm. In particular, Hewlett Packard and Best Buy have used people analytics to minimize employee turnover. There are fewer reports of companies using analytics to identify who to fire. It is no secret, however, that managers will often take a subjective approach to determining who to fire. Eric Siegel, author of Predictive Analytics states that “In general, the point of making data-driven decisions is to move away from the gut and more toward empirically validated decisions” (Siegal 2016). The potential for big data in people analytics is therefore to remove subjectivity in the evaluation of employees. Without getting too complex, something as simple as a browser history search may reveal that an individual has been searching for a new job, a signal that they are considering leaving. More subtly, increased use of social media websites and personal searches may reveal lower engagement in the company and be predictive of future turnover. A summary score, that firms have used includes a “flight-risk” score, which could be represented in a normalized score between 1 and 0, with an increased flight risk score, say 0.87, suggesting an employee is more likely to leave the firm than an employee with a flight-risk score of 0.3.

Application to the case

You have been briefed by the CFO that the CEO wants to hit earnings estimates, and the impact of the restructure on earnings is to be as minimal as possible. The fiscal year end is December 31^st 2020, and the division will cease operations on August, 31^st 2020. Subject to the following complications:

Under no circumstances is the accounting to violate GAAP.
The restructure of the division requires a 75% cut in total workforce, with the 25% being relocated to other areas of the company.
The 25% of employees to be relocated have been identified and will be relocated within a month.
Of the 75% to be cut, 25% are to be terminated immediately or by the fiscal year end and the remaining 50% by August 31^st 2020.
Monthly employment costs are $6,000 per employee. The firm will offer a two-month severance payment ($12,000) to employees that leave voluntarily within a month and a one-month severance payment ($6,000) for those that leave involuntarily on August 31^st 2020.
Those that leave after the voluntary window and before August 31^st, 2020 will not receive any severance payments.

Your role is to suggest a plan of action that can lead to the lowest restructuring cost estimate for the quarter ending March 31^st 2020 and scenario analyses for the total restructuring cost for the fiscal year ending December 31^st 2020. You have received the scenario estimates from the Accounting Department. Some of the scenarios involve selecting individuals based on the internally generated flight risk score calculated by the HR and IT departments. The use of this flight score is the contentious part of the decision, and you will need to consider various legal and ethical issues.

Legal and ethical issues: The firm’s legal counsel has noted that there are no contractual obligations that are in place, and suggest that employees can be terminated with or without cause as long as the selection of individuals does not violate antidiscrimination law. Federal Employment Law prohibits discrimination on the basis of race, sex, color, national origin, or religion (Title VII of the Civil Rights Act of 1964), on the basis of age for employees over the age of 40 (Age Discrimination in Employment Act), those with disabilities (Americans with Disabilities Act), those who are pregnant (Pregnancy Discrimination Act of 1978), those that fail polygraph tests (Employee Polygraph Protection Act), genetic information (Genetic Information Nondiscrimination Act) among other provisions and specific State employment laws.

The general counsel does not believe that the use of data collected from individuals, such as browser history and email, violates employee privacy as the company policy which has been communicated to all employees is that all electronic communication and retrieval is subject to internal audit and monitoring. You have also heard the HR manager mention that they are concerned that many individuals are unaware that their email and web browser history might be monitored. They raise the concern that it may be unethical to fire individuals for poor performance, or even workplace reduction, based on web browser monitoring.

Browser history: The IT department has been able to generate reports on individuals’ browser history using searches to identify certain internet domains. They have identified job search websites, competitors’ websites, and social media websites, including Facebook and twitter, personal searches including political websites, news websites, and medical related searches. They can provide output on the total time spent on various websites, both from any company issued desktop computers used by employees and any personal wireless devices that connected through the company wireless, both of which require a login. They will keep this data in a secure location until it is requested by the group you are working with requests it. They are also holding back on analyzing email content or any other further monitoring until it is requested.

Accounting Analytics

Black box measures: The difficulty with looking at “black-box” measures, such as the internally generated flight-risk score is that we do not know how whether the score is purely objective or has inadvertently added an implicit bias. Implicit bias is defined on Dictionary.com as “bias that results from the tendency to process information based on unconscious associations and feelings, even when these are contrary to one’s conscious or declared beliefs.” Implicit bias is widely understood to be a cause of unintended discrimination that leads to racial, ethnic, socioeconomic, and other inequalities, and is a significant issue when making business decisions, including those related to hiring and retention.

Two-Sample Tests: The way we can test for any bias in the measure is to use a statistical test that compares two sample means. In this case, we will use these statistical tests to investigate whether the flight risk score has an unintended bias. Specifically, we will consider unintended gender bias by examining whether those identified as high flight risk are more likely to be male or female.

Data and Resources

The following data and resources are available in the case supplement:

There are three datasets related to this case:
1. The flight risk scores and surveillance of web searches data is recorded in “employeedata.xlsx”
2. The “Resutructuring_estimates_and_impact.xlsx” has the accounting department estimates
3. The “name_gender.csv” file has the likelihood of a first name being male or female.

References

Deloitte. 2016. Global Human Capital Trends 2016: The new organization: Different by design: Deloitte University Press.

Siegal, E. 2016. Predictive Analytics. Hoboken, NJ: John Wiley & Sons.

Acknowledgements: This case written by Asher Curtis in Winter 2017. Revised Fall 2020.