The turn of the 21st century has been laden with high-profile corporate scandals, prompting widespread concern about the standards of conduct followed by big business. Intrigued by the complexity of managing corporate behavior in a global context, three Harvard Business School professors decided to investigate corporate conduct standards around the world. But what started as a seemingly straightforward search soon grew into a major undertaking that required a dedicated server to process multilingual data from thousands of employees at four large global firms.
The results of their research are both heartening and disturbing.
“We were looking for 'Generally Accepted Conduct Principles.'"
—Rohit Deshpandé
The good news is that the researchers found broad consensus among the employees of the four multinationals who were asked for their opinions about appropriate corporate behavior.
"Along the lines of Generally Accepted Accounting Principles, we were looking for 'Generally Accepted Conduct Principles,' " says Rohit Deshpandé, who conducted the research along with HBS colleagues Lynn S. Paine and Joshua D. Margolis. "And we actually have them. We found them."
The bad news is that the researchers discovered a wide divergence in opinions between how respondents thought a company should behave and how their company actually behaved. Furthermore, the lower a respondent's position in the company, the greater the gap between "should" and "actual." "People at the top of the hierarchy generally have a more positive view of an organization than the people in the middle and on the frontlines," Paine says. "The top of the organization doesn't really know what the frontlines are seeing."
The professors are now working on an article that explores various uses of the corporate assessment tool that they developed for their research.
They recently sat down with HBS Working Knowledge to discuss the origins of the project, and where they plan to take it from here.
Three Phases
The effort began in 2004, evolving from the School's required first-year MBA course Leadership and Corporate Accountability, which focuses on the economic, legal, and ethical responsibilities of business leaders. As they studied various business cases, the students were looking for objective criteria for making ethical decisions. "They were asking, 'What standards are we supposed to be adhering to?' " Paine recalls, adding that she heard the same question from executives who, at the time, were facing pressures from many external groups to follow various codes of corporate conduct.
In launching the Global Business Standards project, the professors aimed to provide an ethical reference point for both future and current business leaders. "We wanted a way to help companies answer these questions," Margolis says. "How do we know if we're doing well on ethics? And how do we know how we're doing relative to others?"
Phase 1 of the project looked for key similarities and differences in conduct guidelines and regulatory requirements from around the world: the Caux Round Table Principles, the Global Reporting Initiative, the UN Global Compact, the OECD Guidelines for Multinational Enterprises, the Interfaith Center on Corporate Responsibility principles, the Sarbanes-Oxley Act, various US Securities and Exchange Commission regulations, and several stock exchange listing rules, as well as the company codes of 14 top global firms.
“People at the top of the hierarchy generally have a more positive view of an organization than the people in the middle and on the front lines."
—Lynn Paine
The team noticed key differences according to who authored each code: Those written by business practitioners generally emphasized the company's financial health and employees' responsibilities to the company, while codes written by multi-sector groups (such as the UN Global Compact) emphasized companies' responsibilities to employees and the general public. But the team found eight recurring principles among all the codes, covering specific mandates such as obey the law, protect the environment, avoid cooking the books, keep promises, respect and protect human rights, and refrain from bribery. The research team created a codex of widely endorsed standards based on these similarities.
Phase 2 of the project focused on developing a survey instrument derived from the codex and centered on two crucial issues: one, the extent to which business practitioners agreed with the common precepts of the world's leading conduct codes for business; and two, the extent to which their companies actually followed these precepts—that is, the "shoulds" and the "actuals."
"We started discussing doing something empirical to test this notion of an ethical gap," Deshpandé says.
The team developed 62 questions based on the principles in their newly created codex, and then pilot-tested their survey with volunteers among executives enrolled in the Advanced Management Program at HBS. The team then refined the survey and since 2006 has administered it in 11 cohorts of AMP participants to nearly 900 executives from more than 600 companies. (The results of the AMP surveys have not yet been published since this phase of the project is ongoing.)
Phase 3 of the project aimed to address the same crucial questions as phase 2, but across a stratified sample of employees at several multinational companies. The researchers wanted the impressions of frontline workers as well as corporate executives. The goal was to develop a benchmark of real-world business conduct that practitioners could use to evaluate and track their own company's ethical performance.
And this is when the project got really complicated. For starters, there was the issue of buy-in. The researchers discovered that some companies balked at the idea of employees potentially airing shortcomings in company behavior, even if the information were to remain anonymous. In other cases, top executives agreed to participate initially, but then hesitated when lawyers voiced reservations or HR managers raised concerns about "survey fatigue" among employees. "Some of the others wanted to be able to modify the questions to suit their particular needs, but that didn't meet our research objectives," Deshpandé recalls.
“We just hung on for dear life and moved as quickly as we could."
—Joshua Margolis
Executives who did choose to participate saw the survey as a unique opportunity to gain insight into perceptions of the company's conduct among a large cross section of employees around the world, and to benchmark themselves against other leading companies. When the global financial crisis hit, the project team decided to close the survey and analyze the data gathered up to that point. In the end, phase 3 included only four companies—but all four were major multinationals, headquartered in the United States, Europe, and Japan.
In the course of research, the team inadvertently discovered some cultural truths.
"I experienced firsthand what I've learned in my research on cross-cultural management, which is that in the United States you can get a quick yes in agreeing to participate, but when you get into actual implementation, the yes may turn into a no," Paine says. "In Japan, it's likely that they'll take a long time getting to yes. But after they commit to do it, they are really engaged. It was an interesting cultural difference."
Because all four companies involved in phase 3 were multinationals, the surveys had to be translated from English into nine additional languages. "That was actually quite a task because a lot of these ethical issues are fairly nuanced, and it took a while to make sure that we had not only the correct syntactical interpretation but also the correct idiomatic interpretation," Deshpandé says.
There was also the matter of trying to complete the surveys in a short amount of time, to respect the participants' schedules and to avoid the possibility of more companies pulling out of the project. "We just hung on for dear life and moved as quickly as we could after they said yes," Margolis says.
Collecting and processing data from a 5,000-person cross section of invited respondents in each company in a matter of weeks was a bandwidth-heavy proposition. (In the end, the researchers received 7,600 responses, but they had to be ready to process up to 20,000.) "We realized we did not have the computing wherewithal to handle it because our system would crash if employees all over the world were all responding at once," Deshpandé says, explaining that the School ended up investing in a dedicated server for the project. "We were not aware of anyone else at the School engaged in a project of this magnitude—both in the sheer number of responses and in the cultural complexity."
Future Phases
After publishing the final results of the multinational survey, the team plans to explore more focused research on companies headquartered in emerging markets such as Brazil, India, China, and Russia. In the meantime, the Global Business Standards project already has garnered a great deal of interest among multinational firms, including, Margolis observes, those that hesitated to participate in the initial survey.
In focusing on the perceived differences between what a company should do and what it actually does do, this research has yielded a valuable insight: firms should adopt a performance perspective on conduct as opposed to a compliance mentality, Margolis says. "It's not just about thou shalt not, it's about taking a path of continual improvement."