Analytics

FUTURE LABS SUBMISSION

Clinging to Excel as a Security Blanket: Investigating Accountants’ Resistance to Emerging Data Analytics Technology

Pamela J. Schmidt, Ph.D.

Associate Professor School of Business

Washburn University 1700 SW College Ave

Topeka, KS 66621 785-670-2052

Pamela.schmidt@washburn.edu

Kimberly Swanson Church, Ph.D. Assistant Professor

Bloch School of Management University of Missouri - Kansas City

5110 Cherry St Kansas City, MO 64110

816-235-2890 churchk@umkc.edu

Jennifer Riley, Ph.D., CPA (CO)

(Corresponding Author) Associate Professor

Department of Accounting University of Nebraska - Omaha College of Business, MH 228R

6708 Pine Street Omaha, NE 68182

(402) 554-3984 Email: jenriley@unomaha.edu

1

Clinging to Excel as a Security Blanket: Investigating Accountants’ Resistance to Emerging Data Analytics Technology

Summary Introduction

Research Questions: Are accountants heeding the call to adapt to emerging data analytics tools? Does Status Quo Bias Theory explain professional accountants’ resistance to switch to emerging data analytics tools? Motivation: Recently, one of the authors of this proposal hosted a data analytics training session, and communicated this session to the professional accounting community, faculty, and students. The communication described the session as a hands-on learning opportunity featuring an emerging data analytics application, PowerBI. Less than a handful of individuals registered for the session. Undeterred, the author reframed the communication to describe PowerBI as an expanded version of Excel, with much the same functionality to which the professionals were already accustomed. Registration jumped from three to 25 participants. This is anecdotal evidence from an isolated experience; however, it highlights our initial motivation: is the pull of the familiar, the lure of the Excel security blanket, keeping accounting professionals from fully embracing emerging alternative technologies for data analytics? We have all heard the warnings: The accounting profession must embrace and adapt to the new world of big data, data science, and by extension, artificial intelligence and automation, or become the “weavers of the 21st century” (Lukomnik 2018). Making this jump requires a near overhaul of the entire profession, something that some accountants are still fighting, or ignoring. This was evidenced by the recent, visceral uproar to a Wall Street Journal article (Shumsky 2017) calling for the retirement of Excel, and anecdotally, by our first-hand experience. These events motivated us to investigate if the accounting profession is clinging to the familiar, habitual, security blanket of Excel, at the expense of effectiveness, suitability, and even survival. Our initial literature search identified a growing body of research on the importance of data analytics to the profession and the curriculum (e.g., Dzuranin, Jones, and Olvera 2018; Schneider, Dai, Janvrin, Ajayi, and Raschke 2015). Surprisingly, there is a lack of attention directed toward whether and how accounting professionals are actually making this transition. And most importantly, if they are not, why not? There has been a loud, consistent, and often alarming call to action, yet limited empirical evidence exists for the profession heeding this call. In fact, signs from the practitioner literature indicate accounting professionals may be resisting, and as a result, are being left to the dustbin of history. Make no mistake, large public accounting firms and corporations are investing heavily in emerging data analytics technology. The concern lies in whom the firms are hiring to fill the positions created by these investments. Practitioner articles seem to indicate accounting professionals may not be at the top of the list. Rather, the hiring of non-accountant data scientists and technology experts is increasing dramatically in positions once held by an accountant (Tysiac and Drew 2018). Of course, accountants have accounting domain knowledge to contribute, but many fear accountants are moving too slowly with regards to gaining technology skills and adopting innovations. The accounting profession needs “to start making significant inroads” simply to survive and catch up, let alone become leaders of the change (Tysiac and Drew 2018, 6). There has been significant effort and contribution of first-movers, including accounting educators, researchers, and professionals, who have embraced and promoted data analytics (as referenced above and throughout this proposal). However, the practitioner press is replete with pleas for the profession to wake up to “a slow-moving existential crisis” (Lukomnik 2018). Even if the segment of the profession that has leapt forward is highlighted, the perception of accounting is one of a backwards-focused, tunnel-visioned, outdated discipline (Lukomnik 2018). This proposal is aimed at understanding two things: first, identifying the contributors to accountants’ resistance to change and second, identifying

2

characteristics of accountants who resist versus those who adopt emerging technologies. We draw on the integrated Status Quo Bias Theory (SQBT) (Kim and Kankanhalli 2009) from the IT literature to explore individual-level attitudes and behavioral indicators regarding the emerging technological revolution. Method: We will develop a survey instrument from the integrated model of SQBT in Kim and Kankanhalli (2009), and gather survey responses from practicing accountants to test the structural model using Covariance-Based Structured Equation Modelling (CB-SEM) or Partial Least Squares-SEM (PLS-SEM) analysis.

Practical Implications, Value to Stakeholders and JETA Readers.

• Organizations can use these results to 1) understand their employees’ current attitudes toward emerging data technologies; 2) explore and implement initiatives to increase colleague and organizational support and encourage acceptance; 3) recognize the importance of appropriate communication of emerging technologies to employees; 4) identify appropriate and successful behaviors indicating a willingness and movement towards gaining data analysis skills; and 5) inform ongoing recruitment and hiring initiatives.

• Professional industry groups such as the IMA and AICPA may use the study results to inform and advance continuous education practices. Many professionals depend on these associations for their skills development needs. These groups are critical stakeholders, and can leverage their position to foster greater acceptance of emerging data technologies and an increased willingness to embrace dynamic change as the norm, not the exception.

• Accounting educators can look to these results to develop curriculum that will not only educate students on emerging technologies, but also to encourage greater acceptance and adaptability in students. If students enter the profession with an attitude of technology agility, computer self-efficacy and innovation, they may be better prepared to embrace change as an opportunity rather than a problem to be avoided.

• Researchers in all areas of emerging technologies may be interested in our results. Although this study in the context of data analytics, the resistance of any technology has consequences for its diffusion and use. Whether the emergent technology is blockchain, automation and robotics, artificial intelligence, machine learning, IoT, or something as yet discovered, no technology will be effective without acceptance.

• Finally, all of these stakeholders can use these results to develop and implement interventions for accountants who are resisting the profession’s transformation. Considering the rapid development of technology, it is inevitable that emerging data analytics technology is only the beginning. By understanding the interventions that work and those that do not, organizations can instill an agile mindset throughout their professionals.

Literature Review and Hypotheses

Data Analytics in Accounting There is a growing body of research on the integration of data analytics in the accounting curriculum (e.g., Dzuranin, Jones, and Olvera 2018); the use of data analytics in auditing and business (e.g., Brown- Liburd and Vasarhelyi 2017; Vasarhelyi, Kogan, and Tuttle 2015); and the effects (real and predicted) on the accounting profession for the future (e.g., Tysiac and Drew 2018; Richins, Stapleton, Stratopoulos, and Wong 2017). The overall conclusion of this literature is that the future lies in emerging technologies, not only in data analytics but in all forms of automation, machine learning, IoT, etc. Because data analytics is arguably one of the earliest drivers of this technological revolution in accounting, it also has driven the most research and attention. Large accounting firms and corporations are investing heavily in both technological tools and people with technical knowledge (e.g., Cohn 2017). However, evidence

3

seems to indicate some of this investment is going towards redefining accounting-related job roles and filling these technology-related jobs with non-accounting professionals. Some have even suggested this as a best practice: “It’s not just about bringing in the brightest CPAs and CFAs anymore. Fill your team with data scientists and broaden the notion of what a modern financial team truly is” (Silverman 2019, np). This is very concerning for the profession; if we cease to be seen as leaders and owners of the business language/data arena, one in which accountants are the go-to experts for financial information and metrics of firm value, we may cease to exist as a profession. While tangible assets are becoming less and less valuable in the global economy, the worth of the intangible “asset” of information and the ability to turn information into value has grown (Lukomnik 2018). The path into this new reality is through data, analytics, technology, and agility. The vice chair of the AICPA, Bill Reeb, noted accountants are “going to go kicking and screaming” into the future (Tysiac and Drew, 4), but shouldn’t we be running head first to the front of the pack instead? Why are accountants resisting? Why are they kicking and screaming? Emerging Technologies, Resistance, and Status Quo Bias Reinking, Arnold, and Sutton (2015) examine the antecedents and consequences of early adoption of emerging IT to explore the apparent failure of technology investments to lead to increased productivity (Brynjolfsson and Hitt 1998). They posit the duration of the protracted productivity paradox described in the Model of Technology Diffusion (MTD) (Atkeson and Kehoe 2007) will be shortened by technological advancement. Two critical factors in the MTD are the level of built-up knowledge and the continued investment in old technology. Intuitively, it makes sense the more substantial the level of built-up knowledge, the less likely someone is to replace it with new technology. More surprising is the continued investment in the old technology: employees persist in learning and upgrading their skills and companies continue to dedicate resources (Atkeson and Kehoe 2007; Reinking et al. 2015). Reinking et al. (2015) used the resource-based view of the firm to hypothesize the time lag for diffusion will be shortened by today’s extreme pace of technological advancement. Results of their archival study are mixed, and the authors conclude this “will be disappointing to many in that, 40 years into the third technological revolution, productivity gains by the early adopters of emerging technologies are not discernable” (69). We propose another critical component influencing the slow diffusion of new technology may lie in the behavioral aspects of technology acceptance and resistance. Specifically, we propose a study to investigate whether employees’ individual-level biases toward maintenance of the status quo contribute to the resistance of IT implementation. We draw our proposal from the substantial body of research on technology acceptance and resistance in the information systems literature. This literature includes a number of theoretical perspectives to explain IT resistance such as the widely studied Technology Acceptance Model (TAM) (Davis 1989), the Theory of Planned Behavior (TPB) (Ajzen 1991), and the Equity-Implementation Model (EIM) (Joshi 1991). In 2009, Kim and Kankanhalli proposed an integrated status quo bias theory (SQBT) as an alternative approach to understanding the phenomena of IT user resistance. The integrated theory adds concepts from TPB and EIM to the status quo bias perspective. The integrated SQBT aims to explain an individual’s preference for maintaining their current state through cost-benefit analysis, perceived value, colleague opinion, and self-efficacy and organizational support for change. Hypotheses Rational decision-making principles indicate perceived value of any choice is a function of the net benefits of that choice (Kahneman and Tversky 1979). In regards to SQBT, this requires consideration of the perceived switching benefits (the perceived utility a user would enjoy in switching from the status quo to the new IS” (Kim and Kankanhalli 2009, 573)) and perceived switching costs (“the perceived

4

disutility a user would incur in switching to the new IS” (Kim and Kankanhalli 2009, 572)). SQBT proposes the aggregate of transition, uncertainty, and sunk costs are directly associated with an individual’s resistance to switch. In addition to the direct effect of switching costs on user resistance, these costs influence user resistance indirectly through their influence on perceived value. Switching benefits influence perceived value through improvements in job performance such as efficiency, effectiveness, and quality. We suggest practicing accountants will assess the value of emerging data analytics technology through this rational, cost-benefit approach, and propose the following hypotheses to test these propositions:

H1: The perceived value of the new data analytics tool is negatively associated with an individual’s resistance to change from Excel to a new data analytics tool.

H2: Switching costs are positively associated with an individual’s resistance to change from Excel to a new data analytics tool.

H3: Switching costs are negatively associated with an individual’s perceived value of a new data analytics tool.

H4: Switching benefits are positively associated with an individual’s perceived value of a new data analytics tool.

SQBT conceptualizes the perceived behavioral controls of TPB as self-efficacy and organizational support for change. Individuals with lower levels of self-efficacy are more likely to feel threatened by change and fearful of their own ability to master the new situation (Bandura 1995). This results in the expectation that self-efficacy will have an inverse relationship with resistance to change from Excel to a new data analytics tool. At the same time, self-efficacy may also influence an individual’s assessment of switching costs through their perceptions of uncertainty (Bandura 1995; Compeau et al. 1999):

H5: Self-efficacy for a change from Excel to a new data analytics tool is negatively associated with user resistance to change.

H6: Self-efficacy for a change from Excel to a new data analytics tool is negatively associated with perceived switching costs of the change.

Kim and Kankanhalli (2009, 573) define organizational support for change as “the perceived facilitation provided by the organization to make users’ adaptation to new IS-related change easier.” Organizations can create a culture of acceptance and positivity around the new technology. This will directly reduce their resistance to change. Additionally, the organization can provide resources such as training and dedicated time to employees in order to assist them in adopting the new technology. Users may then perceive reduced transition costs and indirectly reduce resistance to change:

H7: Organizational support for a change from Excel to a new data analytics tool is negatively associated with user resistance to change.

H8: Organizational support for a change from Excel to a new data analytics tool is negatively associated with perceived switching costs of the change.

SQBT conceptualizes subjective norms as colleague opinion. Individuals seek social approval and fear negative reactions from their colleagues, resulting in the tendency to conform (Ajzen 2002). Accordingly, we expect the acceptance of or resistance to a change from Excel will be directly affected by colleagues’ opinion regarding the new data analytics tool. If colleagues present a positive opinion toward a change from Excel to a new tool, it can reduce the uncertainty of the change, thereby reducing switching costs. Similarly, colleagues’ promotion of the advantages of a new technology can manifest in an increased assessment of the switching benefits:

H9: Colleague opinion is negatively associated with an individual’s resistance to change from Excel to a new data analytics tool.

5

H10: Colleague opinion of a change from Excel to a new data analytics tool is negatively associated with perceived switching costs of the change.

H11: Colleague opinion of a change from Excel to a new data analytics tool is positively associated with perceived switching benefits of the change.

We present our hypothesized model of SQBT in Figure 1.

Insert Figure 1

Method We propose the use of Kim and Kankanhalli’s (2009) instrument to test our hypotheses. The instrument includes several scales adopted from prior research and others developed by those authors:

• Three questions for perceived value, adapted from Sirdeshmukh et al. (2002) • Four questions for switching benefits, developed by Kim and Kankanhalli (2009) from similar

constructs of Moore and Benbasat (1991) • Four questions for switching costs, adapted from Jones et al. (2000) • Three questions for colleague opinion, developed by Kim and Kankanhalli (2009) from

research by Venkatesh and Davis (2000) • Three questions for self-efficacy, adapted from Taylor and Todd (1995) • Three questions for organization support, developed by Kim and Kankanhalli (2009) from

research by Thompson et al. (1991) • Four questions on user resistance, developed by Kim and Kankanhalli (2009) from the

framework of resistance behaviors created by Bovey and Hede (2001) The instrument concludes with questions to gather data for standard demographic and control variables. The target sample is practicing accountants and finance professionals in all industries and business sizes. After data is collected, survey validity is established with Confirmatory Factor Analysis, before testing the model using CB-SEM or PLS-SEM. CB-SEM tests whether survey data supports the operationalization of theory (Geffen, Straub, and Boudreau 2000), but requires a relatively large sample size (Hampton 2015). If this is not achieved, PLS-SEM can be effectively utilized to test a model with smaller sample and/or non-normal data (Hair, Ringle, and Sarstedt 2011). Prior research serves to guide tests of proposed hypotheses. Additional analyses of control variables allows for exploration of individual demographic characteristics and/or firm-related characteristics common to those who display a tendency to resist versus those who embrace emerging data analytics technology. Control variables include the demographic data such as gender, age, and position, plus additional data regarding years and depth of Excel use and method of learning Excel. Basic information about firms is also included to better understand the context of the workplace that might be influential. This information will spur additional research that paints a better picture of the individuals suited and evolving into the new accounting hybrid professional, one that combines accounting domain knowledge, value-adding decision ability, and technological agility and skill. Future Research A number of additional avenues exist for further research in this area. For issues related to those already in the profession, we suggest studies on interventions to address resistance behavior. How can organizations reduce the switching costs incurred by professionals? This necessitates not only attention toward optimizing identifiable resources such as time and training, but also toward intangible aspects such as accountants’ tendencies to cling to “what we know” and to protect our “turf”. To what extent can simple changes in communication language and the framing of an emerging technology affect acceptance? Do cognitive biases and sunk costs dictate professionals’ rational cost-benefit analyses?

6

Future research should also address issues related to those preparing for the 21st century accounting and technology hybrid profession. For example, how can accounting educators develop students’ acceptance of rapid technological change? Studies of innovative projects, learning methods, and degree structures, may help educators teach not only a tool’s nuts and bolts, but also the behavioral influences that may foster resistance to emerging technology. The hybrid professional suggested above can spur additional avenues of research to understand how they developed the characteristics that encourage their innovative spirit, how these characteristics can be best utilized, and how they can be fostered in those who show a tendency to resist. From a broader view, if this hybrid professional is to rise from the ashes of the 20th century accountant, how can the reputation of the profession in the popular press be restored and revamped so that the hybrid becomes the sought-after team member once again? All of these avenues investigate biases from the perspective of the currently active accounting professional. Taking a broader view, accounting firms and the profession as a whole also play a role. Accounting firms have developed structures, often hierarchical and bureaucratic, that may reinforce the benefits of clinging to a status quo. Accountants are under tremendous time pressure, which discourage and even prohibits development of new skills and expertise. The need to “get the job done” strengthens professionals’ tendencies to continue using the tools they are comfortable with rather than undertaking challenging new options. The profession as a whole also supports behavioral biases of the status quo that impede adoption of new technology (Hood 2018). Real-time reporting, continuous auditing, and similar on-demand services driven by consumers rather than accountants and auditors challenges the profession’s ingrained revenue model. Many firms continue to base fees on billing hours by person by level, which incentivizes time spent over value-added output (Baker 2008). Institutional Theory offers an intriguing framework for analyzing the behavioral biases from a firm and industry level. Institutional Theory describes the development and embedding of processes and procedures into an organization or industry, such that certain behaviors and expectations become rigid rules necessary for legitimacy (DiMaggio and Powell 1983). This theory can be used to understand the current rigidity of the accounting profession towards its status quo and how technologies become institutionalized in the industry. It can then be used to explore how new technologies, models, and behaviors can replace these structures to create a more efficient and effective value-adding industry (Agyekum and Singh 2018).

Deliverables We plan to administer the survey through the authors’ participation in several professional accounting meetings; we plan to achieve a sample of at least 175 responses. Timetable is as follows: Incorporation of revision suggestions from the Future Labs panel by 5/31, data collection through July 31, statistical analysis through September 30, working paper completion through November 30, journal submission shortly thereafter.

7

References Agyekum, A.A.B., and R.P. Singh. 2018. How technology is changing accounting processes: Institutional Theory and Legitimacy Theory perspective. Journal of Accounting and Finance, 18 (7): 11-23. Ajzen, I. 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes 50: 179-211. Ajzen, I. 2002. Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior. Journal of Applied Social Psychology 32 (1): 1-20. Atkeson, A., and P. Kehoe. 2007. Modeling the transition to a new economy: Lessons from two technological revolutions. The American Economic Review 97 (1): 64–88. Baker, R.J. 2008. The firm of the future. Journal of Accountancy, 206 (5): 68-73. Bandura, A. 1995. Exercise of personal and collective efficacy in changing societies, in Self Efficacy in Changing Societies, A. Bandura (ed.), New York: Cambridge University Press: 1 -45. Bovey, W.H., and A. Hede. 2001. Resistance to organizational change: The role of defense mechanisms. Journal of Managerial Psychology 16 (7): 534-548. Brown-Liburd, H., and M.A. Vasarhely. 2015. Big data and audit evidence. Journal of Emerging Technologies in Accounting 12: 1-16. Brynjolfsson, E., and L. Hitt. 1998. Beyond the productivity paradox. Communications of the ACM 41 (8): 49–55. Cohn, M. 2017. Audit technology evolving quickly at Big Four. Accounting Today (Nov 14). Available at: https://www.accountingtoday.com/news/audit-technology-evolving-quickly-at-big-four-firms. Compeau, D., C.A Higgins, and S. Huff. 1999. Social cognitive theory and individual reactions to computing technology: A longitudinal study. MIS Quarterly 23 (2): 145-158. Davis, F. D. 1989. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly 13 (3): 319-340. DiMaggio, P., and W.W. Powell. 1983. The iron cage revisited: Collective rationality and institutional isomorphism in organizational fields. American Sociological Review, 48 (2): 147-160. Dzuranin, A., J.R. Jones, and R.M. Olvera. 2018. Infusing data analytics into the accounting curriculum: A framework and insights from faculty. Journal of Accounting Education 43: 24-39. Geffen, D., D. Straub, and M. Boudreau. 2000. Structural equation modeling and regression: Guidelines for research and practice. Communications of the AIS, 7 (7): 1–78. Hair, J. F., C.M.Ringle, and M. Sarstedt. 2011. PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice, 19 (2): 139-151. Hampton, C. 2015. Estimating and reporting structural equation models with behavioral accounting data. Behavioral Research in Accounting 27 (2): 1-34. Hood, D. 2018. The profession’s biggest challenges. Accounting Today (Oct 01). Available at: https://www.accountingtoday.com/news/the-accounting-professions-biggest-challenges Joshi, K. 1991. A model of users’ perspective on change: The case of information systems technology implementation. MIS Quarterly (June): 229-242. Kahneman, D., and A. Tversky. 1979. Prospect theory: An analysis of decision under risk. Econometrica 47 (2): 263-292. Kim, H-W., and A. Kankanhalli. 2009. Investigating user resistance to information systems implementation: A status quo bias perspective. MIS Quarterly 33 (3): 567-582. Lukomnik, J. 2018. Will accountants become the weavers of the 21st century? AccountingToday (Nov 19). Available at:

https://www.accountingtoday.com/opinion/will-accountants-become-the-weavers-of-the-21st- century

Reinking, J., V. Arnold, and S. Sutton, S. 2015. Antecedents and consequences of early adoption of emergent technologies: The IT revolution. Journal of Emerging Technologies in Accounting 12: 51-70.

8

Richins, G., A. Stapleton, T. Stratopoulos, and C. Wong. 2017. Big data analytics: Opportunity or threat for the accounting profession? Journal of Information Systems 31 (3): 63-79. Samuelson, W. and R. Zeckhauser. 1988. Status quo bias in decision making. Journal of Risk and Uncertainty 1: 7-59. Schneider, G., J. Dai, D.J. Janvrin, K. Ajayi, and R. Raschke. 2015. Infer, predict and assure: Accounting opportunities in data analytics. Accounting Horizons 29 (3): 719-742. Shumsky, T. 2017a. Stop Using Excel, Finance chiefs tell staffs. The Wall Street Journal, November 27. Shumsky, T. 2017b. Finance pros say you’ll have to pry Excel out of their cold, dead hands. The Wall Street Journal, November 30. Silverman, G. 2019. How the CFO can build a data-driven company. AccountingToday (Mar 18).

Available: https://www.accountingtoday.com/opinion/how-the-cfo-can-build-a-data-driven- company?utm_campaign=daily- mar%2019%202019&utm_medium=email&utm_source=newsletter&eid=2e33055c6575d3d020 af2e376ee35177.

Sirdeshmukh, D., J. Singh, and B. Sabol. 2002. Consumer trust, value and loyalty in relational exchange. Journal of Marketing 66 (1): 15-37.

Taylor, S., and P.A. Todd. 1995. Understanding information technology use: A test of competing models. Information Systems Research 6 (2): 144-176.

Thompson, R.L., C.A. Higgins, and J.M. Howell. 1991. Personal computing: Toward a conceptual model of utilization. MIS Quarterly 15 (1): 124-143.

Tysiac, K., and J. Drew. 2018. Accounting firms: The next generation. Journal of Accountancy 225 (June), (6): 3-9.

Vasarhelyi, M.A., A. Kogan, and B. Tuttle. 2015. Big data in accounting: An overview. Accounting Horizons 29 (2): 381-296.

Venkatesh, V., and F. Davis. 2000. A theoretical extension of technology acceptance model: Four longitudinal field studies. Management Science 46 (2): 186-204.

9

FIGURE 1: Hypothesized Model

H1 H4

H3

H9

Switching Benefits

Switching Costs

Perceived Value

Resistance to Change

H2

Self-efficacy

Org. Support

Colleague Opinion

H11 H10

H6 H5

H8 H7