In an experiment, a team of researchers found that referred workers who shared a high number of friends and connections on Facebook with the person who referred them put more effort into the job than those who had been hired anonymously via the open market.

And the more social connections shared between the two parties, the greater amount of effort provided by the referred worker, the experiment revealed.

The findings were detailed in a new paper, Hiring an employee’s friends is good for business: Overcoming moral hazard with social networks, published in the journal Labour Economics. The paper is co-authored by Professor Amrita Dhillon, from King’s College London; Professor Ronald Peeters, from the University of Otago; Oliver Bartrum, from King’s College London; and Ayse Muge Yuksel, from Maastricht University.

In their experiment, the researchers used a group of more than 130 undergraduate students from the School of Business and Economics at Maastricht University, with the students divided into the roles of employer, referrer, and worker. Prior to the experiment, all participants allowed the researchers to retrieve information on their Facebook friends and connections.

Across a two-stage process, those in the referrer and worker groups were made aware of any direct friendships and then mutual friendships that existed between them, but identities were kept secret.

The researchers found that people who were hired by someone with whom they shared a number of Facebook friends and connections put in an average 7.48 units of effort, compared with those who were hired anonymously, who put in an average of 5.65 units of effort.

This increased effort could be more pronounced in a real world setting, the researchers said, because the identities of the referrers and workers might not be anonymous.

The researchers noted: “We demonstrated that limited information on social proximity extracted from Facebook for a large number of participants is sufficient to increase the level of effort provided by referred workers.

“Social preferences between workers in the field who know one another’s identity, and who are likely to be more socially proximate anyway, will be greater compared to those among the individuals in our participant pool. This points to a greater reduction of moral hazard [risk] in the field than our quantitative results suggest.”

You can read the paper in full here.