The costs of liability of newness for teams: Somewhere between 33% and death

To what extent do we understand the liability of newness of teams? Arthur Stinchcombe wrote a highly influential chapter on organizational ecology in which he created the term ‘the liability of newness’. For Stinchcombe this was in reference to the disproportionate amount of work young firms have to do to overcome the costs of acquiring knowledge, trust and social ties (internally and externally). The concept of liability of newness has been adopted principally within entrepreneurship and startup research, with a focus on startup costs and obstacles related to legitimacy and innovation.

The liability of newness of a team is something that has been given less attention. Part of that may be explained by the fact that a ‘new’ team is a slightly vague concept. Secondly, there is an implicit handling of ‘newness’ within the extensive literature on small group and team formation (forming and storming…). So, is there something more specific we can identify about the liability of newness that affects teams? Below I talk through two examples where the liability of newness is at play. Both involve pairs: one is a comparatively trivial kind of pair (tennis doubles) and the other anything but trivial (aviation).

The liability of newness costs you 33% performance (...on the tennis court)

I play tennis for my local club and the focal point of our competitive year is the Summer Doubles Leagues. In this competition each tennis club fields a team of 6 people (there are separate men’s and women’s leagues) forming 3 doubles pairs. We are not talking Wimbledon here; these are leagues for enthusiastic amateurs who are interested as much in the social side of organised sport as they are in the competition. In these matches, three ‘sets’ are played regardless of the outcome of each set; so, the best result a doubles pair can achieve is 3-0. A few years ago I set about building our results into a database. The results of this showed an interesting characteristic to our performance: people who played together as a pair only once almost never won a match. In fact, doubles pairs that only played together once rarely even managed to win a single set. I analysed 8 teams across 6 seasons, 1,152 tennis matches in total. I found that a team that has played together at least once before on average wins 1 more set per match than a pair that is playing together for the first time. In our league matches where you play 3 sets per tie, that is a difference of 33% in performance and could be the difference between winning 2-1 and losing 1-2. Further still, when I ran my team’s results adjusted for better team selection (i.e. avoiding ‘new’ pairings where it would have been possible), if we had simply ensured that we fielded established pairings instead of new pairings we would have finished at least one place higher in the league every season and in some cases as many as three places higher.

The liability of newness could cost you your life (in a plane)

Although the tennis example may have resonance, it is a relatively small sample and there are potentially interacting factors, so it may be considered indicative at best. But, the liability of newness of teams has been identified in a far more serious and consequential environment - commercial aviation. Dismukes et al. (2017) reflect on the accidents they investigated for the National Transport Safety Board in the US. In a series of 37 accidents analysed over a 12 year period between 1978 and 1990, 73% of the accidents involved pilots who were flying together for the first time that day.

If 73% of car accidents involved a certain model of tyres, those tyres would be banned pending an investigation. The question is, how do you take a team beyond the newness state? My tennis example showed that the threshold from newness to non-newness was just one match, after one match the performance increases on average by 33%. Maybe pilots should have to do some simulation time specifically with other pilots before they fly together. I think there is something valuable in just giving this dimension attention, too. For example, if you read accident reports on air crashes there is always a deliberate passage that summarises the pilots’ respective experience (the Captain had logged over 15,000 hours with the airline and acquired 3,075 hours in command of the Airbus a330; the First Officer had logged 11,140 hours with the company and 1,502 with the aircraft). These statements always seem to be passively suggesting that there is no case to answer in questioning the skills or experience of the pilots. But, it would make less comfortable reading if you saw 'the pilots had logged zero hours flying together'. Do pilots, and indeed tennis doubles pairs, even consider what they are doing together to be teamwork? Either way, my conclusion is that the liability of newness of teams is worth giving some attention and zeroing-in (pun intended) on the vulnerability to error and diminished performance of teams who 'haven't flown together' before.

References

Dismukes, R. K., Berman, B. A., & Loukopoulos, L. (2017). The limits of expertise: Rethinking pilot error and the causes of airline accidents. Ashgate: Hampshire, UK.

National Transportation Safety Board [NTSB] (1994). A Review of lightcrew-involved Major Accidents of US Air Carriers, 1978 through 1990 (Report No. PB94-917001, NTSB/SS-94/01).

Stinchcombe, Arthur L., 1965. Social structure and organizations. In: March, J.G. (Ed.), Handbook of Organizations. Rand McNally: Chicago, IL, pp. 142 - 193.