The image was created with the help of Microsoft Designer

This piece has originally appeared on Medium.

I like to argue that providing workers with immunity for failed innovation projects is a much better way to promote corporate innovation than by “celebrating failures.” One way to do this would be to modify corporate employment policies. For example, companies can place employees involved in strategic innovation projects on fixed-term employment contracts instead of employment at will.

This point of view is based on research by Viral V. Acharya and his colleagues. They showed (here and here) that laws that limit firms’ ability to discharge employees promote corporate innovation. The authors call this phenomenon an “insurance effect”: feeling increased protection from the negative consequences of failure, employees are more committed to engaging in risky innovative projects.

The dose makes the poison

Not everyone agrees with Acharya and his co-authors. For example, a 2019 study tracked employment protection across 20 OECD countries from 1950–2003. The study found a 5% decrease in the number of patents and their quality submitted by firms after a major increase in employment protection (relative to a set of firms operating in the same industry at the same time but located in countries without changes in employment protection). The study concluded that labor protection impedes, rather than promotes, innovation.

This conclusion is supported by findings showing that strong labor protection effectuated by trade unions negatively affects innovation within firms. One mechanism causing this negative effect might be reduced investment in innovation at the firm and industry levels.

I think the best way to reconcile all the results described above is to suggest that the relationship between labor laws and innovation is not linear but follows an inverted-U pattern instead. When the level of workers’ protection is too small, increasing it through adopting stricter labor laws provides job insurance against failures and therefore spurs innovation. However, providing too much protection — e.g., forced by the unions — stifles innovation via multiple mechanisms, including reduced investment in innovation, restricted labor mobility, etc.

Is religious diversity good or bad for your team’s performance?

It appears that an inverted-U dependency applies to other fields. For example, a 2016 article describes the effects of team religious diversity — the degree to which team members differ in their religious beliefs — on the performance of 66 health teams in three large hospitals in Dubai, UAE, a predominantly Muslim country.

Using the proportion of Muslims vs. non-Muslims within a team as a measure of religious diversity, the authors found an inverted U-shaped relation between diversity and performance: moderate diversity was associated with higher performance, while homogeneous and highly heterogeneous teams underperformed moderately diverse ones (see the picture below, reproduced from the article).

Interestingly, the positive effect of moderate diversity on performance was stronger for teams charged with more complex tasks (e.g., surgical vs. clinical teams), which reminds us of the fact that the positive effect of social policies on innovation is especially strong in knowledge-intensive industries, such as technology and finance.

When more is too much

Another article analyzed the role of domain experts (people whose primary professional experience is within a specific industry) on the performance of corporate boards. The authors wanted to know how the proportion of domain experts affected the board’s performance.

They looked at financial data for 1,300 community banks and found that when banks faced increased levels of uncertainty, the higher proportion of domain experts on the board resulted in a higher likelihood of banks’ failure. The major problem with having too many domain experts was “cognitive entrenchment,” the inability of expert-dominated boards to effectively respond to new information or unfamiliar situations.

Sounds like “too much” knowledge and expertise harms performance. Weird, isn’t it?

In fact, we shouldn’t be surprised: our own experience tells us that many things in life are good “in moderation.” For example, at small doses, the rattlesnake venom can be used to treat arthritis and cancer, but it becomes deadly at larger doses. Similarly, we all long for attention from our loved ones, yet we begin protecting our space when this attention gets overwhelming.

So, in innovation, business — and life in general — dose is everything. You may remember this wisdom tonight when adding salt to a dish you cook for dinner.