The use of behavioral science interventions, and particularly social norms, in tax compliance is a growing industry for scholars and practitioners alike in recent years. However, the causal mechanism of these interventions is unknown, where effects could be explained by a pro-social desire to support one's community, conditional cooperation, desire to conform, or fear of reprisals. We conduct a field experiment in local government taxation in the United Kingdom which tests the effectiveness of a social (descriptive) norm against a control condition and against messages that highlight the enforcement process. The social norm outperforms enforcement salience, suggesting that this explanation, although more powerful than the control, does not fully explain compliance effects. This study further provides evidence that social norm type interventions can be effective at the subnational level, a context where previous work has shown they may produce null effects.

The use of behavioral science interventions, and particularly social norms, in tax compliance is a growing industry for scholars and practitioners alike in recent years. However, the causal mechanism of these interventions is unknown, where effects could be explained by a pro-social desire to support one’s community, conditional cooperation, desire to conform, or fear of reprisals. We conduct a field experiment in local government taxation in the United Kingdom which tests the effectiveness of a social (descriptive) norm against a control condition and against messages that highlight the enforcement process. The social norm outperforms enforcement salience, suggesting that this explanation, although more powerful than the control, does not fully explain compliance effects. This study further provides evidence that social norm type interventions can be effective at the subnational level, a context where previous work has shown they may produce null effects.

No abstract

Background Cervical screening saves 4500 lives in England annually and is off ered nationally to women aged 24-64 years. A successful screening programme relies among other factors on uptake by the at-risk population. National coverage fell from 80•3% in 2005 to 77•2% in 2015. How health information is framed, in terms of potential gains (eg, lives saved) versus losses (eg, lives lost) of taking part in screening, can aff ect uptake. This study aimed to investigate the eff ect of framing in invitation letters for cervical screening on attendance in a low uptake urban area of London.Methods In a randomised controlled trial, two simplifi ed invitation letters were compared with the existing invitation letter (control). Women living in the trial area, who were due for screening between Oct 21 and Dec 16, 2015, were randomised to receiving a gain framed message, a loss framed message, or the existing invitation letter (1:1:1, by computer-generated random number). Participants but not investigators were masked to allocation. Primary outcome was attendance for cervical screening at 15 and 32 weeks. Association between framing and attendance was measured with logistic regression. Analysis was by intention to treat. This trial is registered with ISRCTN, number ISRCTN93653543.Findings 6223, 6223, and 6202 women were allocated to the control, gain framed, and loss framed groups, respectively. Mean age was 35•4 years (SD 9•9). Median number of invitations previously received was 4 (IQR 2-6) and past attendances was 1 (0-4). At 15 weeks no signifi cant diff erence in uptake between the groups was found. Attendance at 32 weeks was 32•7% (n=2032) in the control compared with 32•3% (2011) in the gain framed group and 33•5% (2079) in the loss framed group. Logistic regression showed a signifi cant increase in uptake in the loss framed group (adjusted odds ratio 1•1, 95% CI 1•0-1•21; p=0•04). No adverse events were observed.Interpretation Overall, the use of loss framing in invitation letters for cervical screening can increase the uptake of cervical screening at little or no extra cost. This study highlights one example of how information in invitation letters can be made more salient. Further research to explore these techniques is required. How changes in content might aff ect uptake is diffi cult to predict, so further trials in diverse populations are needed to ensure that new interventions are eff ective and not detrimental to uptake.Funding Department of Health. ContributorsSHu contributed to concept and designed the intervention, performed data analysis and interpretation, and drafted the abstract. DK contributed to intervention design and critically reviewed data analysis and abstract. SHo contributed to design intervention, trial set-up, and critically reviewed the abstract. FT critically reviewed the data analysis and interpretation of results and abstract drafting. IA contributed to designed interventions and trial setup. IV critically reviewed data analysis and interpretation and critically reviewed the abstract. AD ...