Social Interaction at Work (Digest Summary)

Using a sample of individuals in Norway, the authors show that the stock market trading decisions of employees appear to be affected by the advice and actions of their coworkers. In one test, they show that the correlation between the trading activity of new employees and that of longer-serving employees rises sharply from zero as the new employees adapt to their new working environment. The authors also note that investment decisions tend to be concentrated in the industry where the individual is employed and that the stocks purchased tend to underperform. In short, individuals should be cautious of heeding advice from their coworkers.

The study is a useful reminder to investment practitioners of how certain parts of the equity market (e.g., themes, sectors, stocks) can be subject to crowding as investors converge on the latest popular idea. Such convergence can lead to volatility when the idea does not perform as expected and investors try to sell simultaneously. The authors also highlight how topical “glamour” growth stocks can become overpriced and subsequently underperform. Of course, the overall market impact will depend to a large degree on the relative size of retail investment versus institutional investment.

From a broader corporate governance perspective, the authors unintentionally illustrate the importance of providing new employees with guidance on the cultural expectations of an organization. Otherwise, they will tend to adopt the practices of their peers. Ideally, the cultural expectations and practices of current employees are somewhat aligned.

The authors combine stock trading data from the Oslo Stock Exchange with detailed sociodemographic data from the government statistical agency to provide rich insight into the trading behavior of Norwegians. The data cover the period 1994–2005.

In the merged dataset, the authors match the details of any share trades by an individual with such employment identification as plant identification number, location, and industry, as well as such personal information as income, wealth, age, gender, and education. Immediate relatives and more distant relatives (e.g., cousins, nieces, and nephews) are identified.

A series of regressions are then performed. The initial regression tests for correlations between the stock purchases of an individual and those of his or her coworkers in a given month and includes a series of sociodemographic control variables, as well as dummy variables, to capture trading by an individual’s family and other individuals living in the same area.

Subsequent regressions refine these ideas further. To confirm that the results are not due solely to unobservable similarities between coworkers, the authors test the correlation of new employees’ trading behavior with that of their coworkers for a period of 12 months prior to and after the new employees are hired. The authors find that correlations with future coworkers are essentially zero prior to being hired and rise sharply as the individuals become immersed in the culture of their new employer. Correlations with ex-coworkers correspondingly fall sharply as their influence becomes less relevant.

Stock ownership levels vary across countries as a function of many variables, including investing and social culture, the legal framework, tax treatment, previous investment experiences, and the balance between institutional and retail investment. It would be interesting to determine whether similar findings hold across other markets. From a research perspective, however, it is rare to have access to such a rich repository of data; thus, analysis across other regions might be difficult.

A number of other slow-moving forces may also affect these findings over time. As the population ages, investor behavior may change. Older employees may be less attracted to equity investments after the global financial crisis. The evolution of social media has been significant over the past decade: Individuals now have access to broader, more knowledgeable investment resources, perhaps leading to less reliance on their coworkers for ideas. Finally, investment product offerings have matured over time with the rise of such products as exchange-traded funds. All of these factors may lead to a reduction in relevance of what is essentially “water cooler” discussions—that is, casual conversations among coworkers.