Mange vil nok umiddelbart mene, at sygeplejersker i højere grad udviser omsorg for andre end ejendomsmæglere. De er trods alt ”velfærdsstatens frontsoldater”.

Men helt så enkelt er det ikke, viser Jacobsen m.fl. (2022), som bruger en twist af diktatorspillet til at vise deres pointe. I diktatorspillet får én person tildelt en portion penge, og skal efterfølgende vælge, hvor stor en andel af pengene, personen vil beholde selv, og hvor stor en del af pengene, der skal gå til andre (i dette tilfælde er ”andre” Amnesty International).

Jacobsen m.fl. (2022) finder i det klassiske diktatorspil – som mange nok ville forvente – at sygeplejersker rent faktisk ER mindre selviske end ejendomsmæglere, når de skal tage stilling til, hvor mange af pengene Amnesty International skal have.

In Part 1 of our experiment, every subject was asked to share 100 Norwegian kroner (NOK), or about 16.5 USD, between themselves and Amnesty International. Our results show that brokers contribute, on average, a substantial 61 percent of their endowment; nurses‘ contributions were significantly larger, however, reaching an average of 75 percent. Thus, in this particular context, nurses were indeed more generous than brokers.

Men her stopper Jacobsen m.fl. (2022) ikke. De udvider spillet ved at give deltagerne mulighed for at betale 10 kr. for at undgå at skulle tage stilling til, om de vil give Amnesty International pengene. Og på dette punkt er sygeplejerskerne faktisk mere selviske, for de er mere tilbøjelige til at betale for at undgå at skulle tage stilling.

In the second part of our experiment, therefore, subjects were given the choice between two options, A and B. Option A implied repeating the game from Part 1 with certainty. Option B implied a 50 percent probability of repeating the game from Part 1, and a 50 percent probability of receiving 90 NOK with no opportunity to donate to Amnesty International. Thus, option B implied a 50 percent probability of costly exit from the dictator game situation. Nurses were substantially more likely to opt for the exit alternative than brokers: 36 percent of brokers and 51 percent of nurses chose B.

Aha. Så selvom sygeplejersker gav mere under part 1, så er de også mere tilbøjelige til at vælge at undgå at skulle tage stilling til at give penge til Amnesty International. Og effekten er betydelig. I gennemsnit gav de 75% af pengene væk. Hvis de ”slipper ud af spillet” og undlader at skulle tage stilling, koster det dem kun 10%. En forskel på hele 65%-point.

Men. Det bliver endnu mere komplekst:

Interestingly, however, those subjects who opted for exit but drew a repetition of the dictator game contributed substantial amounts: 31 percent for brokers and 42 percent for nurses. Even after accounting for exit behavior, nurses were more generous than brokers (in Part 2 and in total).

Som forfatterne påpeger, er resultaterne svære at forklare. Sygeplejerskerne er mere tilbøjelige til at forsøge at undgå at blive sat i en situation, hvor de skal tage stilling til, om de vil give penge til Amnesty International. Men hvis de bliver sat i situationen, donerer sygeplejerskerne mere. I hvert fald i gennemsnit.

As we discuss in more detail below, this pattern of behavior is hard to reconcile with most standard models of social preferences. One possible explanation, however, could be that nurses experience a stronger sense of duty to conform with moral ideals, and that this sense of duty applies more strongly to sharing decisions than to exit decisions.

Man kan spekulere over, om det måske netop er fordi de ved, at de vil give mere, hvis de bliver sat i situationen, at de forsøger at undgå den…

Jacobsen m.fl. (2022) er udgivet her (en gratis arbejdspapir-version kan hentes her).