Tuesday, March 21, 2017

Behavioural Science and Irish Public Policy: Some readings and resources

In the last budget, the Irish Government Economic Evaluation Service (IGEES) released a document summarising the use of behavioural research in Irish government department. This followed a previous report they released summarising behavioural economics more generally (both of these documents are available here). Below gives a list of the different projects covered in the IGEES report, along with some relevant international examples in the given headings. The recent OECD report provides many more examples. The material below is intended to provide a source for discussion and reference in Ireland. Thanks to Sarah Breathnach for work on this. We are currently holding regular network events in Ireland to explore the potential of this research in the Irish context (see here for details and mailing list). 

1. Irish Project: Increasing Filings from Late Income Tax Returns. Department: Office of the Revenue Commissioners
International Example: Unpaid Car Tax, Behavioural Insights Team UK.
The Behavioural Insights Team (BIT) worked with the DVLA to test the efficacy of different messages upon individuals who fail to tax their vehicles. The BIT tested the original DVLA letter against a new letter with simpler, harder-hitting messages (such as ‘Pay Your Tax or Lose Your [Make of Vehicle]’) and an image of the owners vehicle. It is thought that the image would attract the attention of recipients and make the idea of losing their vehicle more salient. This image also involes a strong element of personalization, as drivers are shown an image of their own car. The results revealed that overall, sending letters to non-payers of road tax with a picture of the offending vehicle on the letter increased payment rates from 40 percent to 49 percent.

Service, O et al,. (2014) Behavioural Insights TeamEAST: Four simple ways to apply behavioural insights. Available from: http://www.behaviouralinsights.co.uk/wp-content/uploads/2015/07/BIT-Publication- EAST_FA_WEB.pdf
General Efforts to increase tax compliance:

Example US: Eliciting taxpayer preferences increases tax compliance. Lamberton, De Neve and Norton (2014)

While tax aversion has many causes, we suggest that a meaningful portion of that aversion can be understood — and addressed — by considering two psychological characteristics of the tax process. First, tax aversion is created by the decoupling of tax payments and the public goods obtained in return. This disconnect places distance between payment and benefit, and therefore decreases taxpayers’ perceptions of the tax-funded benefits they receive. Second, taxpayer frustration also results from the lack of agency or sense of influence over tax spending — which stands in sharp contrast to one’s control over private spending — and may prompt some degree of psychological reactance (Brehm, 1966). Academics and policymakers have begun to address the first source of frustration, aiming to better inform citizens on where tax money is being spent. To recouple payment and benefit, policymakers now widely publicize information on the allocation of tax dollars across expenditure categories (White House, 2011) and are introducing “Personal Tax Statements” (UK Treasury, 2012). We propose that the annual tax filing process offers an unexploited opportunity for governments to increase taxpayer engagement and compliance. We explore this possibility in a set of experiments that allow taxpayers to express non-binding (advisory) preferences regarding the use of their tax dollars, assessing the effects of this taxpayer agency treatment on tax compliance as well as satisfaction with tax payment and attitudes towards taxation. We find that providing taxpayers with such “taxpayer agency” — eliciting tax spending preferences — significantly increases tax compliance (16%). Additionally, allowing taxpayers to signal their preferences on the distribution of government spending results in a 15% reduction in the stated take-up rate of a questionable tax loophole.  We also observe that agency operates, in part, by recoupling payment and benefit. Giving taxpayers a voice may act as a two-way “nudge,” transforming tax payment from a passive experience to a channel of communication between taxpayers and government.

Lamberton, C. P., De Neve, J. E., & Norton, M. I. (2014). Eliciting taxpayer preferences increases tax compliance. Available at SSRN 2365751.

Example UK: Increasing tax compliance using social norms and fairness.  Hallsworth, List, Metcalfe, and Vlaev (2014).

Hallsworth et al. (2014) look at the propensity of taxpayers to pay their outstanding taxes and how norms and fairness messages affect payments. The Authors describe two large scale field experiments in the UK conducted in 2011 and 2012 using a subject pool of more than 100,000 individuals in each of the experiments. In the first experiment, the authors test two fairness and three normative messages which are added to the standard reminder letter for late payment of taxes. The authors report the largest effect for a normative message (‘Nine out of ten people in the UK pay their tax on time. You are currently in the very small minority of people who have not paid us yet.’). People in this treatment group were 5.1% more likely to make a payment towards their debts within 23 days after receiving the reminder letter compared to the control group. The second experiment is implemented the same way and tests for the psychological distance of normative messages as well as the effect of descriptive and injunctive norms. General descriptive norms were shown to have a larger effect compared to general injunctive norms. The psychological distance or the normative messages was varied by describing the behaviour of others in the same region (‘The great majority of people in your local area pay their taxes on time.’) or in a similar situation (‘Most people with a debt like yours have paid it by now.’). More specific norms were shown to have a larger effect. The authors argue that because including normative messages in the reminder letters is relatively cheap, also small effects can be cost effective.

Hallsworth, M., List, J., Metcalfe, R., & Vlaev, I. (2014). The behavioralist as tax collector: Using natural field experiments to enhance tax compliance (No. w20007). National Bureau of Economic Research.

Example UK: Social norms on tax debt payments. The HMRC Behaviour Change Team (2012).

The HMRC Behaviour Change Team are engaged in a series of randomised controlled trials. The first of these tested the influence of social norms on tax debt payments. 17 Letters were sent to 140,000 taxpayers, and took four forms. The first was the standard letter with no mention of social norms (the control group), the remaining three all contained the statement ‘9 out of 10 people pay their tax on time’ in the context of either Britain as a whole, for the taxpayers’ postcode, or taxpayers’ home town. The interventions appeared to be successful: 67.5 per cent made payments in the control group; 72.5 per cent for national social norms; 79 per cent for postcode social norms and, finally, 83 per cent for home town social norms. It is estimated in the report that the difference between the standard letter and the highest performing home town social norm letter (15 per cent) could advance £160 million of tax debts to HMRC over a six-week period.

Cabinet Office (2012). Applying behavioural insights to reduce fraud error and debt.

Example UK: The impact of personalised text messaging on fine repayments, BIT (2012).

The Courts Service and the Behavioural Insights Team wanted to test whether or not sending text messages to people who had failed to pay their court fines would encourage them to pay prior to a bailiff being sent to their homes. In the initial trial, individuals were randomly allocated to five different groups. Some were sent no text message (control group), while others (intervention groups) were sent either a standard reminder text or a more personalised message (including the name of the recipient, the amount owed, or both). The trial showed that text message prompts can be highly effective, 28.2% of the standard text recipients paid the fine whereas 33% of the recipients of the standard text with personalisation paid the fine.

Behavioural Insights Team (2012). Test, Learn, Adapt.  Available from https://www.stir.ac.uk/media/schools/management/documents/economics/Nudge%20Database%201.2.pdf

Example US: Taxpayer response to an increased probability of audit. Slemrod, Blumenthal  and Christian (2001).

Slemrod et al. (2001) describe one of the first field experiments in tax compliance. The authors explore the effect of differences in perceived audit rates by sending a letter to a group of taxpayers in Minnesota. In 1995 a group of 1724 randomly selected Minnesota taxpayers were informed by letter that the returns they were about to file would be ‘closely examined’. Compared to a control group that did not receive this letter, low and middle-income taxpayers in the treatment group on average increased tax payments compared to the previous year, which was interpret as indicating the presence of noncompliance. The effect was much stronger for those with more opportunity to evade; in fact, the difference in differences is not statistically significant for those who do not have self-employment or farm income, and do not pay estimated tax. Surprisingly, however, the reported tax liability of the high income treatment group fell sharply relative to the control group.

Slemrod J., Blumenthal M. and Christian C. (2001): ‘Taxpayer response to an increased probability of audit: evidence from a controlled experiment in Minnesota’, Journal of Public Economics, Vol. 79, No. 3, p. 455-483

Example Sweden: Tax compliance and loss aversion. Engström, Nordblom, Ohlsson and Persson (2015).

Engström et al., (2015) find that in Sweden, taxpayers are more aggressive about claiming deductions when they owe additional tax at the time of filing than when they expect a refund - a phenomenon that is consistent with the predictions of prospect theory. An obvious policy implication is that a tax collection strategy that relies on over withholding followed by refunds at the time of tax filing may increase tax compliance and total taxes paid. The authors therefore argue the practical applications of prospect theory to tax compliance. Interventions that recognize individuals’ aversion to loss may be particularly useful in increasing tax compliance.

Engström, P., Nordblom, K., Ohlsson, H., & Persson, A. (2015). Tax compliance and loss aversion. American Economic Journal: Economic Policy7(4), 132-164.

Example Austria: Tax compliance within the context of gain and loss situations. Kirchler & Maciejovsky, (2001)

This study on 60 self-employed and 59 business entrepreneurs investigates self-reported tax evasion within the context of taxpayers' previous expected tax payments at filing time (payment or refund), their asset position (expected versus current asset position), and their tax category (self-employed versus business entrepreneurs). It examines whether different tax categories have an impact on the habitual decision-making framework and hence influence the reference point employed in the decision process. The results indicate that for the self-employed in the study an unexpected surprise payment leads to low tax compliance, whereas an unexpected surprise refund leads to high tax compliance. Thus, their self-reported tax compliance can be best described by the current asset position. By contrast, the reference point business entrepreneurs employ in making tax decisions can be best described as dictated by their expected asset position. As a result, expected payments are associated with low and expected refunds are associated with high tax compliance on the part of these individuals. In addition, the findings of the study showed that knowledge of the legal principles of Austrian tax law is correlated with tax morality, and that women are less compliant than men. Attitudes towards the tax system and the perceived justice of the system are not correlated significantly with self-reported tax evasion.

Kirchler, E., & Maciejovsky, B. (2001). Tax compliance within the context of gain & loss situations, expected and current asset position, & profession. Journal of Economic Psychology. 22(2), 173–194

Example Estonia: Changing the perception of taxes (behaviourally-informed initiative). Tax and Customs Board (2011/ongoing).

According to Lillemets (2009), it is important to increase public awareness of the need to pay taxes because this is a means of increasing the public’s readiness to pay taxes voluntarily. Public research carried out in 2009 which covered attitudes of the population towards payment of taxes, indicated that awareness of the services people receive from the state is relatively low: 26% of respondents did not know what kind of services they receive from the state, while around half of those (11%) said that they get nothing from the state. The same survey also indicated that voluntary tax payment could be increased by higher trust towards the state, increased awareness of the use of tax income and good relationships between the state and the citizen. In an effort to modify tax behaviours and shift the focus from paying taxes as a burden to something which contributes to public good, the Estonian Tax and Customs Board (Ministry of Finance) launched an information campaign to raise awareness of how taxpayers’ money is being used by the state. The information campaign was implemented in two parts. The first part of the campaign was conducted in nine Estonian cities during January and February 2010. The main message used in the campaign was: “Unpaid taxes will leave a mark. You like highways in order, ambulances, efficient work of rescue workers and the police. So do we.” In addition, a thank you message was attached to rescue cars and ambulance cars in Tallinn saying that these cars had been bought with taxpayers’ money. The second part of the campaign was conducted in eight cities across Estonia during October 2011 and used radio commercials and outdoor ads. The messages were socially relevant and related to the Estonian context. Overall, the campaign was regarded as successful in that it was relatively well noticed and the average score was good compared to other state and educational campaigns. Also the follow-up campaign was effective in confirming the campaign message and making people think about why we pay taxes. It also helped to create an understanding that paying taxes helps the state to function and provide social guarantees to people.

Tax and Customs Board (2011), ‘Information campaign “Unpaid taxes will leave a mark” continues’. Available from http://www.eurofound.europa.eu/observatories/emcc/case-studies/tackling-undeclared-work-in-europe/information-campaign-on-tax-compliance-estonia.

Example: Increasing tax compliance through positive rewards. Brockmann, Genschel, & Seelkopf, (2016)

Can governments increase tax compliance by rewarding honest taxpayers? We conduct a controlled laboratory experiment comparing tax compliance under a ‘deterrence’ baseline to tax compliance under two ‘reward’ treatments: a ‘donation’ treatment giving taxpayers a say in the spending purposes of their payments, and a ‘lucky’ treatment giving taxpayers the (highly unlikely) chance of winning a lottery. The reward treatments significantly affect tax behavior but not in a straight forward way. While female participants alter their behavior as expected and comply somewhat more, men strongly react in the opposite way: They evade a much higher percentage of taxes than under the baseline. Apparently, there is no one-size fits-all approach to boost tax compliance

Brockmann, H., Genschel, P., & Seelkopf, L. (2016). Happy taxation: increasing tax compliance through positive rewards?. Journal of Public Policy, 1-26.

Example Denmark: Simplification of the tax system & compliance. Danish Tax Authority

The Danish Tax Authority implemented an initiative to avoid tax evasion among young citizens (often due to the perceived excessive complexity of administrative procedures). This consisted of the creation of a simplified platform for tax payment targeting youth, featuring a human-centred design based on behavioural research. Changes increased the use of the tax guidelines by 7% and tax revenues from the target group are expected to increase by 20% compared to previous years.

Example Austria: Prepopulated tax forms: Increasing compliance through simplification. Federal Ministry of Finance (2014)

The Austrian Federal Ministry of Finance offers citizens the FinanzOnline, - "one-click link to the Austrian tax administration." This service allows, for example, citizens to file their tax return electronically using a pre-populated form. This is an example of a nudge to encourage individuals to file their taxes electronically and comply with tax return. In other words, this is achieved through the simplification of the tax return process (in line with the need to decrease information overload) and the reduction of the effort to comply. Moreover, the Ministry's website indicates that "96% rate the Finance Ministry´s application, which has received multiple international awards, as 'very good'" – an example of the use of social norms and framing for the promotion of FinanzOnline as a secure and quality service. There is also a mobile phone signature app, which saw a substantial increase in users from 106,754 in 2013 to 176,721 in 2014. Other countries across Europe including, Spain and France also use pre-populated fiscal and non-fiscal declarations, while Italy and Hungary are in the process of becoming more taxpayer friendly through the use of behaviorally inspired simplifications of tax administrative procedures.

Federal Ministry of Finance (2014). The Austrian Tax and Customs Administration Annual Report 2014 Overview. Available from https://english.bmf.gv.at/services/publications/BMF-BR-Annual_Report_Tax_Customs_2014.pdf?5jcadl.

Meta-Analysis: A Meta-Analysis of Tax Compliance Experiments. Blackwell (2007).

Blackwell conducts a meta-analysis of twenty laboratory experiments, which were carried out between 1987 and 2006. Blackwell attempts to synthesize this literature in a meta-analysis to draw conclusions regarding the determinants of tax compliance. Specifically, the author examines the impacts of traditional economic determinants of tax compliance: the tax rate, the penalty rate, and the probability of audit. In addition, the author examines the effect of a public good “return” to taxes paid. The author finds strong evidence that increasing the penalty rate, the probability of audit and the marginal per capita return to the public good lead to higher compliance, but finds no statistically significant effect of the tax rate on compliance.

Blackwell C. (2007): ‘A Meta-Analysis of Tax Compliance Experiments’, International Center for Public Policy Working Paper Series at Andrew Young School of Policy Studies, Georgia State University

Additional Resources:

Weber, T. O., Fooken, J., & Herrmann, B. (2014). Behavioural economics and taxation (No. 41). Directorate General Taxation and Customs Union, European Commission.

Madrian, B. C. (2014). Applying insights from behavioral economics to policy design (No. w20318). National Bureau of Economic Research.

The British Psychological Society. Behaviour change: Tax & compliance (2016).  Available from http://www.bps.org.uk/system/files/Public%20files/tax.pdf

Lourenço J.S., Ciriolo E., Almeida S.R., & Troussard X. (2016). Behavioural Insights Applied to Policy European Report 2016. Available from

Leicester, A., Levell, P., & Rasul, I. (2012). Tax and benefit policy: insights from behavioural economics. Institute for Fiscal studies, Commentary C125.

2. Irish Projects: Increasing Responses to the SME Survey (2013) & PAYE Survey (2015)
Department: Office of the Revenue Commissioners.
& Encouraging Responses to the National Health Experience Survey.
Department of Health

Example US: Different Domain Similar idea: The impact of alternative incentive schemes on completion of health risk assessment surveys. Haisley et al., (2012).

Purpose: The biggest challenge for corporate wellness initiatives is low rates of employee participation. We test whether a behavioral economic approach to incentive design (i.e., a lottery) is more effective than a direct economic payment of equivalent monetary value (i.e., a grocery gift certificate) in encouraging employees to complete health risk assessments (HRAs).
Design: Employees were assigned to one of three arms. Assignment to a treatment arm versus the non-treatment arm was determined by management. Assignment to an arm among those eligible for treatment was randomized by office.
Patients: A total of 1299 employees across 14 offices participated.
Intervention: All employees were eligible to receive $25 for completing the HRA. Those in the lottery condition were assigned to teams of four to eight people and, conditional on HRA completion, were entered into a lottery with a prize of $100 (expected value, $25) and a bonus value of an additional $25 if 80% of team members participated. Those in the grocery gift certificate condition who completed an HRA received a $25 grocery gift certificate. Those in the comparison condition received no additional incentive.
Measures: HRA completion rates.
Results: HRA completion rates were significantly higher among participations in the lottery incentive condition (64%) than in both the grocery gift certificate condition (44%) and the comparison condition (40%). Effects were larger for lower-income employees, as indicated by a significant interaction between income and the lottery incentive.
Conclusion: Lottery incentives that incorporate regret aversion and social pressure can provide higher impact for the same amount of money as simple economic incentives.

Haisley E, Volpp KG, Pellathy T, Loewenstein G. (2012).  Promoting Completion of Health Risk Assessments with Lottery Incentives. American Journal of Health Promotion.

Example US: Total cost/response rate trade-offs in mail survey research: Impact of follow-up mailings and monetary incentives. Larson & Chow (2003).

This article reports results of a mail survey experiment in which several response-inducement methods were manipulated. The experiment assesses the impact of follow-up mailings and monetary incentives on total cost/response rate trade-offs. Experimental findings lead to a number of recommendations for researchers and managers who conduct mail surveys. First, follow-up mailings and monetary incentives should be used to maximize response rate. Second, given a limited budget for survey administration, follow-up mailings are preferred over monetary incentives. Third, if there is limited time for survey administration, monetary incentives may be preferred over follow-up mailings. Finally, follow-up mailings have the added benefit of enabling nonresponse bias estimation.
Larson, P. D., & Chow, G. (2003). Total cost/response rate trade-offs in mail survey research: impact of follow-up mailings and monetary incentives. Industrial Marketing Management32(7), 533-537.

Example Norway: Improving response rate and quality of survey data with a scratch lottery ticket incentive. Olsen, Abelsen & Olsen (2012)

Background: The quality of data collected in survey research is usually indicated by the response rate; the representativeness of the sample, and; the rate of completed questions (item-response). In attempting to improve a generally declining response rate in surveys considerable efforts are being made through follow-up mailings and various types of incentives. This study examines effects of including a scratch lottery ticket in the invitation letter to a survey.
Method: Questionnaires concerning oral health were mailed to a random sample of 2,400 adults. A systematically selected half of the sample (1,200 adults) received a questionnaire including a scratch lottery ticket. One reminder without the incentive was sent.
Results: The incentive increased the response rate and improved representativeness by reaching more respondents with lower education. Furthermore, it reduced item nonresponse. The initial incentive had no effect on the propensity to respond after the reminder.
Conclusion: When attempting to improve survey data, three issues become important: response rate, representativeness, and item-response. This study shows that including a scratch lottery ticket in the invitation letter performs well on all the three.

Olsen, F., Abelsen, B., & Olsen, J. A. (2012). Improving response rate and quality of survey data with a scratch lottery ticket incentive. BMC medical research methodology12(1), 52.

Example: The Influence of Advance Letters on Response in Telephone Surveys: A Meta-Analysis. De Leeuw et al. (2007).

Recently, the leading position of telephone surveys as the major mode of data collection has been challenged. Telephone surveys suffer from a growing nonresponse, partly due to the general nonresponse trend for all surveys and partly due to changes in society and technology influencing contactability and willingness to answer. One way to counteract the increasing nonresponse is the use of an advance letter. In mail and face-to-face surveys, advance letters have been proven effective. Based on the proven effectiveness in face-to-face and mail surveys, survey handbooks advise the use of advance letters in telephone surveys. This study reviews the evidence for this advice and presents a quantitative summary of empirical studies on the effectiveness of advance letters in raising the response rate for telephone surveys. The major conclusion is that advance letters are also an effective tool in telephone surveys, with an average increase in response rate (RR1) from 58 percent (no letter) to 66 percent (advance letter), and an average increase in cooperation rate (COOP1) from 64 percent (no letter) to 75 percent (advance letter). It seems advanced contact generally improves response rates and, often just as importantly, in a cost effective manner.

De Leeuw, E., Callegaro, M., Hox, J., Korendijk, E., & Lensvelt-Mulders, G. (2007). The influence of advance letters on response in telephone surveys a meta-analysis. Public opinion quarterly71(3), 413-443.

Example: The Role of Topic Interest in Survey Participation Decisions. Groves, Presser, & Dipko (2004).

While a low survey response rate may indicate that the risk of nonresponse error is high, we know little about when nonresponse causes such error and when nonresponse is ignorable. Leverage-salience theory of survey participation suggests that when the survey topic is a factor in the decision to participate, noncooperation will cause nonresponse error. We test three hypotheses derived from the theory: (1) those faced with a survey request on a topic of interest to them cooperate at higher rates than do those less interested in the topic; (2) this tendency for the “interested” to cooperate more readily is diminished when monetary incentives are offered; and (3) the impact of interest on cooperation has non-ignorability implications for key statistics. The data come from a three-factor experiment examining the impact on cooperation with surveys on (a) five different topics, using (b) samples from five different populations that have known attributes related to the topics, with (c) two different incentive conditions. The findings suggest that emphasising the importance of a topic will encourage those already interested in the research subject to take part. However, for those uninterested placing too much emphasis on the topic importance may put them off. Contacting a sampled individual or household in advance is an opportunity to make a contact aware of the sponsor of the research as well as aware of the importance, or salience, of the research. The findings suggest that survey commissioners need to be careful in balancing the message put across on advanced letters and in other contacts. Negatively affecting the participation of those who are already unlikely to take part in research may increase non-response bias.

Groves, R. M., Presser, S., & Dipko, S. (2004). The role of topic interest in survey participation decisions. Public Opinion Quarterly, 68(1), 2-31.
Example US: Improving response rates through better design. Clarkberg & Einarson (2008).

In an effort to address declining response rates on our campus, we considered the role of survey instrument design and how it can contribute positively or negatively to the experience of completing a web-based survey of student engagement. Existing research on survey design emphasizes that potential respondents evaluate signals from the survey instrument itself in deciding whether to respond. We focused on the following four considerations: 1) Survey length. The perceived burden of responding to a survey is tied directly to its overall length as well as the number of questions that appear per web screen. 2). Survey content. Surveys enjoy higher response rates when their contents are seen as relevant to the respondents’ own experiences and values. To enhance the salience of our survey, we asked our interview participants what they thought of the questions we had developed and what they wanted to tell us in a survey. 3) Visual appeal. Research suggests the visual design of a web-based survey affects response rates. Our study found that visual design elements – such as the borders and spacing used in questions, and the number of questions presented per page – elicited strong reactions from our pretesters. 4) Delivery of survey results. Providing respondents with survey results can help build rapport with respondents. In this project, we incorporated instant results into the survey. In this way, we immediately rewarded survey respondents with a handful of findings at key points. Consistent with past research (e.g., Couper et al., 2001; Dillman, 2000), our study found that differences in survey length and visual design seem to have accounted for significant differences in the proportion of students responding to the original and revised versions of our survey, as well as differences in the proportion of survey questions completed

Clarkberg, M., & Einarson, M. (2008). Improving response rates through better design: Rethinking a web-based survey instrument. New York: Institutional Research and Planning, Cornell University. Available from: https://www.dpb.cornell.edu/documents/1000421.pdf

Example US: Using Norm-Based Appeals to Increase Response Rates. Misra, Stokols & Marino (2013)

Background: In an earlier experiment Misra, Stokols, & Marino (2012) found that participants who received a descriptive normative prompt in the message requesting them to complete an online survey were more likely to comply with the request compared to participants who did not receive any normative prompts.
Purpose: Building on that earlier study, the present field experiment compared the separate and additive effects of descriptive and injunctive norm- based persuasive messages on response rates of online surveys. We also investigate the influence of email reminders on response rates.
Intervention: Participants in an interdisciplinary conference were assigned to one of four groups. The three experimental groups received one of the following messages asking them to complete an online survey that highlighted: (1) a descriptive social norm indicating typical response rates among attendees of prior similar conferences; (2) an injunctive norm appealing them to join fellow participants in completing the survey; or (3) both social norms. The control group received a generic request to complete the online survey without any norm based appeals.
Findings: Participants receiving a message highlighting the descriptive social norm when asked to complete an online survey were more likely to comply with the request compared to all the other groups. Additionally, one and two email reminders were found to be effective in improving response rates of online surveys.

Misra, S., Stokols, D., & Marino, A. H. (2013). Descriptive, but not Injunctive, Normative Appeals Increase Response Rates In Web-based Surveys. Journal of MultiDisciplinary Evaluation, 9(21), 1-10.

Additional Resources:

The Scottish Government (2011). Methods in Survey Design to Improve Response Rates: A Review of the Empirical Evidence. Available from: http://www.gov.scot/Resource/Doc/933/0098777.pdf

Improving Participation in Research (2012). The social research newsletter from Ipsos MORI Scotland. Available from:

3. Irish Project: Increasing Attendance at Group Information Sessions. Department of Social Protection.

Example US: The Power of Prompts: Using Behavioral Insights to Encourage People to Participate in Paycheck Plus Demonstration meetings. OPRE Report (2015).

The Paycheck Plus Demonstration is evaluating whether offering single New Yorkers a generous earnings supplement on top of the existing earned income tax credit improves their economic well-being and encourages employment. The report presents findings from two behavioral interventions designed to increase the number of participants who attended a meeting about the Paycheck Plus program. The Behavioral Interventions to Advance Self-Sufficiency (BIAS) team designed two types of postcards: one meant to reflect a typical message that program operators would have produced in the absence of a BIAS intervention (the “standard” version), and one that incorporated concepts from behavioral economics, including implementation prompting, loss aversion, and prominent deadlines (the “behavioral” version). Additionally, half of the participants were sent text messages. The study contained four research groups. Group 1 was sent only behavioral postcards, and Group 2 was sent behavioral postcards and behavioral text messages. Group 3 was sent only standard postcards, and Group 4 was sent standard postcards and standard text messages. Behavioral messaging led to a 7 percentage point increase in meeting attendance (over a base of 18.5 percent), compared with standard messaging. The strongest outcomes were found for the group that was sent both behavioral postcards and behavioral text messages, which improved response by 12 percentage points when compared with the lightest touch, sending only standard postcards. Behavioral messaging led to a large, statistically significant increase in the percentage of participants who attended the meeting.

US Office of Planning, Research and Evaluation. Available from: https://www.acf.hhs.gov/sites/default/files/opre/bias_paycheck_plus_2015_acf_b508_2.pdf

Example US: Framing The Message: Using Behavioral Economics to Engage TANF Recipients. OPRE Report (2016)

In late 2013, Los Angeles County began scheduling appointments with needy families in receipt of temporary assistance who were formerly exempt from engaging with the welfare-to-work program in order to bring them into the program and engage them in activities. Despite extensive communication attempts only about half of the participants who received a reengagement notice between September 2013 and January 2014 attended the scheduled mandatory reengagement appointment. The report presents findings from an intervention designed to increase the number of Temporary Assistance for Needy Families (TANF) recipients who “reengaged” in the welfare-to-work program. Two different notices that employed behavioral techniques were designed; one highlighted the losses participants might experience by not attending the appointment and the other highlighted the benefits they might gain by attending. Participants were randomly assigned to one of three groups: (1) a program group that received the gain-framed notice; (2) a program group that received the loss-framed notice; or (3) a control group, which did not receive additional materials. The test found that receiving an additional behavioral message increased the percentage of program group members who engaged in the program within 30 days of their scheduled appointment by a statistically significant 3.6 percentage points. This increase was largely driven by the loss notice, which increased engagement at 30 days by 4.4 percentage points, while the gain notice, when compared with the control condition, did not produce a statistically significant impact at 30 days.

US Office of Planning, Research and Evaluation. Available from: http://www.mdrc.org/sites/default/files/Framing_the_Message_ES.pdf

Example US: Cutting Through Complexity: Using Behavioral Science to Improve Indiana’s Child Care Subsidy Program. OPRE Report (2016).

The Indiana Office of Early Childhood and Out-of-School Learning (OECOSL) is the lead agency responsible for administering the state’s Child Care and Development Fund (CCDF), which provides child care subsidies to low-income parents who are working or in school. As part of the program, parents are required to verify their continued eligibility for child care subsidies at least every six months by submitting required documentation. Data reported by the OECOSL suggested that about 40 percent of parents missed their scheduled appointment date, about a third who attended any appointment had to attend multiple appointments to complete the process. As parents often struggle to comply with this requirement, the OECOSL aimed to use behavioral insights to encourage parents to attend their first scheduled appointment and to complete the process in one visit. In order to achieve this, the appointment materials were redesigned and a new reminder postcard was created. The redesigned materials incorporated the behavioral elements of simplification, personalization, loss aversion, and reminders. The findings of all interventions are discussed.

US Office of Planning, Research and Evaluation. Available from: https://www.acf.hhs.gov/sites/default/files/opre/bias_indiana_execsummary_acf_508.pdf

Example UK: Reducing Missed Hospital Appointments. Behavioural Insights Team (2015)

Missed hospital appointments are a major cause of inefficiency worldwide. Healthcare providers are increasingly using Short Message Service reminders to reduce ‘Did Not Attend’ (DNA) rates. Systematic reviews show that sending such reminders is effective, but there is no evidence on whether their impact is affected by their content. Accordingly, we undertook two randomised controlled trials that tested the impact of rephrasing appointment reminders on DNA rates in the United Kingdom. Participants were outpatients with a valid mobile telephone number and an outpatient appointment between November 2013 and January 2014 (Trial One, 10,111 participants) or March and May 2014 (Trial Two, 9,848 participants). Appointments were randomly allocated to one of four reminder messages, which were issued five days in advance. Message assignment was then compared against appointment outcomes (appointment attendance, DNA, cancellation by patient). In Trial One, a message including the cost of a missed appointment to the health system produced a DNA rate of 8.4%, compared to 11.1% for the existing message. Trial Two replicated this effect (DNA rate 8.2%), but also found that expressing the same concept in general terms was significantly less effective (DNA rate 9.9%). Moving from the existing reminder to the more effective costs message would result in 5,800 fewer missed appointments per year in the National Health Service Trust in question, at no additional cost.

Hallsworth, M., Berry, D., Sanders, M., Sallis, A., King, D., Vlaev, I., & Darzi, A. (2015). Stating Appointment Costs in SMS Reminders Reduces Missed Hospital Appointments: Findings from Two Randomised Controlled Trials. PloS one, 10(9), e0137306.

Example UK: Addressing Missed Hospital Appointments. NHS Bedfordshire (2013)

Missed appointments are a large source of inefficiency in the NHS: around 6 million appointments are wasted each year, at an estimated cost of £700–800 million (Martin et al. 2012). This initiative trialled a package of simple interventions, based on behavioural science, to increase patient attendance in a primary care setting. Reception staff at 2 primary care sites in NHS Bedfordshire were trained to implement 3 interventions. They were given time to reflect on the changes and ask any questions. Clinical staff were also briefed on the changes. The package of 3 interventions successfully reduced the number of appointments wasted by patients who did not attend (DNA) by 31.7% (124 appointments per month in total across the 2 sites). The interventions included: on the telephone, reception staff asking patients to repeat back verbally the day and time of the appointment they are given before completing the call in the primary care setting, providing patients with a card to write the details of their appointment themselves rather than a receptionist, nurse or doctor doing so replacing the poster highlighting the number of missed appointments with a poster that showed the much larger number of patients who do turn up on time. Reports 12 months after implementation show that a reduction in the DNA rate of about 30% has been maintained.

Example Australia: Increasing Cervical Screening Attendance with Behaviourally Informed Reminder Letters.

NSW Pap Test Register by the Cancer Institute NSW send over 300,000 every year. Less than 30% of women book an appointment within three months after receiving their letter. The aim of the study was to trial whether different reminder letters could increase the number of women who attend Pap tests, in order to increase early detection and cancer survival. Four new 27-month reminder letters were co-designed and trialled in an RCT against the existing letter. The outcome measure was whether women attend a Pap test within 3 months of being sent the letter. Letter 1 involved a gain framed message. Letter 2 involved a gain-framed message and an act now stamp and highlighted risks of not screening. Letter 3 used gain-framing and included a case study of cancer survivor. Finally, letter 4 used gain framing in combination with a commitment device. The results indicated that over 12 months, the Commitment Device letter could lead to an additional 7,500 Pap test appointments within three months of the reminders being received (32.2% of women sent the Commitment Device letter booked a Pap test within 3 months, compared to 29.7% in the control group).

4. Irish Project: Tackling Climate Change, Reducing Carbon Emissions. Department of Expenditure and Reform

Example Netherlands: Reducing traffic congestion through choice architecture.
In The Netherlands, the Ministry of Infrastructure and Environment has implemented the "Optimizing Use" project. In this context, national and regional governments and businesses collaborated to improve road, waterway and railway accessibility to reduce traffic congestion in the busiest regions. The project comprised behavioural measures, such as increasing the number of bicycle shelters at stations. This is an example of a measure addressing a barrier to cycling (i.e. availability of sufficient secure cycle parking) through a change to the choice architecture. This is in clear contrast with monetary incentives to promote sustainable transport like granting "eco vouchers" for buying green products, such as a bicycle or train tickets. A follow-up programme is scheduled for 2017.
Platform Beter Benutten, “Beter Benutten: less congestion in 2014, shorter journey times in 2017.” [Online]. Available from: http://www.beterbenutten.nl/en.
Example Finland: Reducing carbon emissions through simplification (2013).
In 2013, SITRA and the Finish City of Jyväskylä launched the "Towards Resource Wisdom" project, which aims at developing an operating model for regional resource efficiency. In spring 2015, the model was piloted in Forssa, Lappeenranta and Turku and, in June 2015, a network was created to support Finnish cities in making their regional activities carbon neutral and waste free. A set of indicators was also developed to measure progress towards these goals. As part of the Resource Wisdom project, a series of pilots were conducted, one such example is the "Bus Leap Project" aimed at increasing the use of public transport and reducing carbon emissions and fuel consumption. The project taps on behavioural levers such as simplification – i.e. development of a route guidance system to assist residents with basic logistic information – and is testing whether introducing staggered working hours would have an effect on decreasing periods of high traffic.
“Resource wisdom indicators,” Sitra, 2015. [Online]. Available from: http://www.sitra.fi/en/artikkelit/resource- wisdom/resource-wisdom-indicators.
Example Spain: Can Defaults Save the Climate? (2012)
A study by the University of Las Palmas de Gran Canaria used an RCT to test the effect of defaults and framing in the context of a policy for mitigating CO2 emissions. Results showed that framing influenced travellers’ willingness to pay €10 extra for a flight ticket to mitigate their CO2 emissions. That is, 81% paid extra when the question was framed as a rejection (i.e. tick in this box if you would like to deduct the additional amount) vs. 62% when this was framed as an addition (i.e. tick in this box if you would like to include the additional amount). Note however that rejection was also the default option. The implications are discussed.
J. E. Araña and C. J. León, “Can Defaults Save the Climate? Evidence from a Field Experiment on Carbon Offsetting Programs,” Environ. Resour. Econ., vol. 54, no. 4, pp. 613–626, 2013.
Example US: Reducing energy consumption using a social norms intervention (2011)

This paper evaluates a large-scale pilot program run by a company called OPOWER, to test the effects of mailing home energy use reports to residential utility consumers. The reports include energy conservation information as well as social comparisons between a household’s energy use and that of its neighbours. Using data from a randomized natural field experiment of 80,000 treatment and control households in Minnesota, it is estimated that the monthly program reduces energy consumption by 1.9 to 2.0 percent relative to baseline.  To prevent a "boomerang effect" which might cause those better than the norm to regress, the Social Comparison Module adds injunctive messages labelling low- and moderate-consumption households as "Great" and "Good" and adds "smiley face" emoticons.

Allcott, H. (2011). "Social Norms and Energy Conservation." Journal of Public Economics 95(9-10): 1082-1095

Example US: Encouraging Sustainable Practices Beyond Here and Now: The Case of Programmable Thermostats for Low-Income Tenants (2016).

Research has shown that programmable thermostats can help families save on energy costs while keeping homes at comfortable and healthy temperatures. By setting back the temperature when the homes are not occupied or overnight, the occupants can save and keep homes comfortable. 40% of the thermostats in use in America are programmable. We designed a randomized controlled experiment to evaluate the effect of two behavioral interventions in the willingness to keep using thermostat schedules. The experiment had three experimental groups: G1-Control: no intervention; G2-Ability: custom-program the thermostats according to occupants’ schedules and comfort preferences, and leave a sticker to remind the families to go-back to use schedules and keep home comfortable; G3-Motiviation: In addition to G2-Ability, we obtained verbal and signed commitments to keep the schedule during the heating season. We found energy savings for both groups G2-Ability (2%) and G3-Motivation (1.5%). Based on these results we suggest that the activity of programming the thermostats according to familial habits and remind occupants to press run to use the schedules should be adopted as a measure that is a part of a portfolio of energy efficiency offers designed to help low income families to adopt new sustainable habits that can help them adapt to scenarios of increasing energy prices and climate change.

Abreu, J. (2016). Encouraging sustainable practices beyond here and now: The case of programmable thermostats for lowincome tenants. Presentation at the 2016 Behavior, Energy & Climate Change Conference. October 20-22, 2016, Baltimore, MD.

No comments: