Trust me: how cultural behaviors impact supply chain efficiency

New research quantifies the role of trust in supply chain transactions between Chinese and American partners

October 8, 2014

Assistant Professor Karen Zheng

Assistant Professor Karen Zheng

Trust, it turns out, has a measurable impact on bottom lines.

New research by MIT Sloan assistant professor Karen Zheng and her colleagues examine the impact of trust between firms in increasingly global supply chains. Zheng, a member of the school’s operations management group, has devised a way to quantify the effect of trust in supply chain transactions between Chinese and American associates. She and her research partners concluded that Americans are perceived to be more trustworthy, and that the Chinese phenomenon of guanxi, or personal ties, can increase that trust.

“If you’re building a new business relationship, it pays to spend time building trust with Chinese partners,” Zheng says.

The research is chronicled in a new article—Trust, Trustworthiness, and Information Sharing in Supply Chains Bridging China and the United States—co-authored with Özalp Özer at The University of Texas at Dallas and Yufei Ren at Union College. The article appears in the October 2014 issue of Management Science.

Let’s say you plan to sell skateboards in the U.S. that are produced in China. You don’t want to run out before Christmas, so you puff up your demand forecast. Factory managers now have to determine how many skateboard decks to build. If your forecast is too ambitious, they’re on the hook for the extra boards collecting dust in the warehouse.

“The parties need to share information, but incentive conflicts deter people from always sharing,” Zheng says. She brings a fresh approach to this problem, which has long plagued supply chains.

“This is a somewhat classic model that has been analyzed by many researchers,” says MIT Sloan operations management professor Retsef Levi. But until now, Levi says, “researchers were oblivious to the behavioral and cultural factors that actually play a major role. If you ignore them, you may get the wrong answer.” That’s a mistake few companies can afford. Merchandise exchanged between the U.S. and China now accounts for three percent of global trade volume, and China has become the world’s largest exporter and trader.

The research builds on earlier work Zheng conducted for her PhD at Stanford University, where she interned at HP Labs. She noticed that HP used simple price agreements with non-binding forecasts instead of contracts loaded with financial motivations. “Retailers tell them demand, and HP decides how many to build. Theoretically, that should never work,” Zheng says. In reality, she says, “It worked sometimes, and not other times.” She decided to figure out why.

In her first study, Zheng recruited 32 Stanford undergrads for a two-hour simulation. In an experiment echoing MIT Sloan’s Beer Game supply chain exercise, each student was cast as either a retailer submitting demand forecast for a product, or a supplier who had to interpret that forecast and decide how many units to manufacture. The pretend product was never named; Zheng and her fellow researchers did not want subjects to introduce a personal perception about the cost or demand of, say, smartphones or jumbo jets. The researchers set demand volatility and cost of production both high and low and recorded the participants’ decisions. The students were paid up to $75 depending on the outcome of their faux business dealings. Zheng found that production cost made the most impact on how much partners trusted each other.

Using that result as a baseline, Zheng investigated how different cultural orientations impacted trust. She ran the study again, this time with 186 students at the University of Texas at Dallas and Tsinghua University in China. In one-time transactions, Zheng says, Chinese subjects playing the retailer role inflated their forecasts to be twice as high as those of their stateside counterparts; when playing the manufacturer role, they adjusted production to be more than 50 percent less than what their American counterparts delivered. These lower levels of spontaneous trust and trustworthy conduct resulted in a 10 percent loss in supply chain efficiency and profit. However, when Chinese playing the manufacturer role interacted with Americans playing the retailer role, the Chinese trusted the forecast more and produced more.

“Our data show that that’s because the Chinese perceive the Americans to be more trustworthy in general,” Zheng says. In repeated interactions, guanxi kicked in—and stayed. Chinese trust levels increased to match those of Americans, and persisted at a high level even if their counterparts provided inaccurate or inflated data.

“In each round, the participants could infer with some probability if their partner had exploited them,” says Zheng, who grew up in China. “Even if the Chinese students observed their partners manipulating them, they were not willing to decrease their trust immediately. They still maintained a cooperative relationship in the long run. That’s what was most surprising to me.”

So how can you ensure your skateboard startup is a success, from that first forecast onwards? Capitalize on the positive image that Chinese have of Western companies.

“Companies should go beyond the minimum standards to ensure good working conditions for the workers producing their products,” Zheng says. “Think of ways to help address poverty and inequality in Chinese society, and show support in natural disasters. These kinds of activities can help build and maintain a positive image.”

Then, build guanxi. “Appoint an executive to engage in constant and frequent interactions with the Chinese partners,” Zheng advises. “Many American companies practice rotation of management positions, which hurts the build-up of the interpersonal relationships necessary for guanxi.”

Recently, Zheng finished collecting data for a third version of the experiment. This time the subjects are MIT Executive MBA students. “I’m excited to find out if the results are the same when we use a population with more business experience,” she says.