Companies are starting to care more about corporate social responsibility (CSR). Among the largest 250 companies in the world, 92% produced a CSR report in 2015, informing shareholders and the public about the firm’s activities. That’s up from 64% having such a report in 2005. Today, Fortune Global 500 firms spend around $20 billion a year on CSR activities.
Companies are also finding that CSR efforts — such as sustainability initiatives, corporate foundations, employee volunteer programs, and donations to charity — can be important tools for attracting and motivating employees. Research has shown that various forms of prosocial incentives (workers get rewarded not with money, but with the firm engaging in some act to benefit society) indeed increase productivity in simple and complex tasks, increase retention, and even lower employees’ wage demands.
It’s no surprise then that more firms might be investing in CSR. But our research shows that firms should not pursue CSR simply for benefits like greater productivity. We found that if employees think their company is using CSR initiatives instrumentally — trying to engage in prosocial activities only to benefit from it — then they’ll react negatively and put in less effort. In other words, while these initiatives will benefit society, they will backfire for companies if people think they’re being used for the wrong reasons.
To investigate this, we designed an experiment with an Italian firm. We hired around 3,000 workers on Amazon Mechanical Turk (M-Turk) to create taglines for products that the firm could use on their English website. Workers were asked to come up with three slogans as the baseline. We randomized participants into four main treatments that had differing incentives. We offered incentives to get participants to generate an additional three slogans: We told half the main sample they would get private, monetary bonuses; we told the other half that the firm would donate to charity for the additional work.
But while we told half of both groups that the company would give the incentive whether or not they did the extra work, we told the other half the incentive would be given only if they came up with the extra slogans. This condition was supposed to make them see that the incentive was profit-maximizing (instrumental) for the firm — because the size of the incentive was half of the baseline pay, that meant it would be cheaper for the firm to incentivize them to contribute three more slogans than to hire a new worker. Finally, we also had a control group of workers who received only the baseline payment without any other incentives.
Additionally, we conducted a vignette study in which we presented a different pool of respondents from M-Turk with one of two scenarios. In both scenarios, a firm committed to annually donating part of its profit to a renowned charity. However, one scenario said that the CEO only decided to commit after market research confirmed that the donation would be profitable for the firm. The other scenario said the CEO decided to make the donation without doing any market research, just because it was the right thing to do. We then asked respondents how attractive it would be to work for the firm and how they viewed the firm in general.
Our experiment revealed four main findings. First, monetary incentives and charitable incentives worked differently. While making monetary incentives conditional on performance increased effort (that is, people were more likely to submit three additional slogans when the bonus was tied to outcome than when it was just given), making charitable incentives conditional backfired. Only 49% of workers submitted additional slogans when the donation to a charity was tied to their extra work, compared with 54% of workers who did extra work when the donation was made independent of their effort. In other words, when the company only donated if it gained from employees’ extra work, employees were less motivated to do that work.
Second, when we surveyed these workers about their opinions of the Italian firm, we found that those who were offered conditional incentives (both monetary and charitable) rated the firms as less socially responsible than those who were offered non-conditional incentives.
Third, we saw that the effect of a conditional charitable donation was particularly strong for workers who seemed less interested in charity in the first place. Our survey of these workers asked about whether they regularly donate or volunteer to charity. We found that workers who said they seldom donate or volunteer were even less likely to submit extra slogans to support charity when that support was conditional on their work. This makes sense if you think about it: They were not as motivated to work harder to support a social cause, and they still got a negative impression of the firm’s motives. We saw that 54% of those workers created more than three slogans when they were told a donation would be automatic, but only 43% did so when the donation actually depended on their effort.
Fourth, prosocial incentives in general backfired compared with the control group: We found that more workers (61%) were doing three more slogans when they were simply asked to do it than when they were offered a charity incentive. The proportion of workers who did extra work dropped to about 52% with charitable incentives. So even the workers who saw that the company would donate to charity irrespective of whether they sent in more slogans were less likely to do the extra work than workers who were asked for more slogans with no additional incentive. We believe this is because the simple act of announcing the firm’s charitable donation could be perceived as instrumental or strategic to increase workers’ effort.
The results from our vignette study also suggest that employees care about the intentions behind companies’ CSR activities, not just the positive impact those activities have on society. We found that when CSR was made strategically — that is, conditional on a positive outcome from the market research study — respondents rated the firm as less attractive, viewed the firm as less socially responsible, thought the donation was less generous, thought the donation would be less effective in motivating workers, and said they would be less likely to accept a lower wage to work at the company.
Taken together, these results suggest that using CSR instrumentally to increase profits might destroy the very benefits it hopes to achieve. But it’s important to note that our setting had three important limitations. First, our workers were recruited on Amazon Mechanical Turk and are a selection of (gig) workers. We couldn’t investigate the effect of intentions behind CSR on a firm’s actual employees. Second, charitable giving is only one form of CSR, and making charitable incentives performance-based is only one type of instrumental CSR. Workers could react differently to other types of activities. Third, in our experiment workers were randomly assigned to the treatments, which means they were not allowed to sort into specific incentive schemes. One might expect that workers who care more about charitable initiatives are more likely to choose companies that invest highly in CSR. This sorting is likely to mitigate the negative effect of prosocial incentives for such companies. However, companies that are not renowned for their CSR activities may have a pool of less-prosocial workers, which could accentuate the negative effect found in our study. Hopefully these limitations will motivate follow-up studies.
Overall, our study indicates that firms probably should not use CSR initiatives as merely another tool to benefit their HR strategy. Workers care about the genuine use of prosocial incentives and may react negatively to firms using social initiatives to increase productivity or profits. Just as we care about other people’s intentions as well as their actual behavior, workers care about why firms offer CSR incentives, not just how much good they do. It is a tricky balance to get right. But our research suggests that prosocial incentives whose main goal is to increase the firm’s profit can backfire. Managers need to genuinely care about the social impact of their CSR actions — only then will workers reward it with higher productivity.
Stephan Meier is a professor of business at Columbia Business School. He investigates the impact of psychology and economics on human decision-making and its implications for public policy and firms’ strategy.
Lea Cassar is an assistant professor in Behavioral Managerial Economics at the University of Cologne. Her research studies how to design non-monetary incentives to foster workers’ motivation.