Behavioural economics is a hot topic these days and you’ve probably come across its application in a variety of everyday situations; from choice architecture prompting you to select healthier food in the canteen, to altered defaults which help you to save for your retirement, to social norms encouraging you to reduce your energy bills. In this article we look at various research findings and insights around charitable giving, and list five different ways in which a charity might be able to increase donations.
We’ll start with the UK government’s Behavioural Insight Team (BIT) which has been trialling some interesting schemes to make charitable giving easier in the workplace, using behavioural economics. Together with the Charities Aid Foundation, BIT recently published a new report on ‘Applying behavioural insights to charitable giving’3 which looks at the impact of several different trials.
1. Donate more tomorrow
BIT have been working with the Home Retail Group (which owns Argos and Homebase) and the Charities Trust to trial a new way of encouraging charitable giving in the workplace by using a system known as auto-escalation.
This is a clever adaptation of a famous US scheme called Save More Tomorrow, which encourages people to save more for retirement by asking them to commit to saving more when they get their next pay rise, thereby reducing the pain of what is known by behavioural economists as ‘loss aversion’.
Saving more ‘today’ means we’re sharply aware of the resulting reduction in our take home income, but by committing to save more ‘tomorrow’ – as soon as we begin to earn more – we experience no loss in take-home income, making saving easier. Auto-escalation has now become extremely successful in the US and many retirement savings institutions offer it.
BIT adapted this model, and instead of committing to increasing retirement savings, employees were asked to commit to auto-escalation of workplace donations by 3% each year.
Initially take-up of the scheme was low, with only 10% of new donors opting in. So in October 2012, they changed the default, making enrolment to the scheme the automatic setting, but allowing employees to opt-out if they preferred. This has had a significant impact and the take-up rate rose dramatically to 49%. BIT estimate that this could raise an additional £3 million per annum for charities if launched across all payroll schemes.
2. Converting legacy desire into action
Legacies are an important source of income for charities, and they also make financial sense since inheritance tax is reduced for people who leave more than 10% of their estate to a charitable cause. However, BIT noted that there was a disconnect between people’s attitudes to legacies – around 35% of people want to leave money to a charity – and their actual behaviour – only 7% of wills actually contain a charitable bequest.
So BIT looked at using a trigger to prompt more people into leaving money to charity in their wills. They used social norms as that trigger, encouraging people to nominate a charity in their will by reminding them that many other people had also left charitable legacies.
Collaborating with Co-operative Legal Services and Remember a Charity, the trial ran on 1,000 customers who were given different messages by will writers:
- Will writer A [Baseline]: Customers not specifically asked to donate.
- Will writer B [Plain ask]: “Would you like to leave any money to charity in your will?”
- Will writer C [Social norms]: “Many of our customers like to leave money to charity in their will. Are there any causes you are passionate about?”
BIT found that using social norms increased the percentage of will customers leaving a legacy to a charity to over 15% from around 5% and also doubled the size of contributions. The social norm group raised twice as much as the Baseline group (see graph below) and in total raised £990,000 from 1,000 individuals, which represents an increase of £825,000 above the baseline group.
3. Anchoring low to build donation rates
Something that causes a bit of a dilemma for charities is the issue of small requests. Asking for a small donation can often lead to a small donation, and a correspondingly low level of funding raised. Yet asking for larger donations can generate the standard excuses: ‘I don’t have that kind of money right now,’ or ‘I already give to many charities,’ with the additional problem that because they’ve been asked to make a large donation people feel that it is not acceptable to give a smaller amount instead, leading to low response rates.
This tiny addition not only led to an increased rate of contribution of 50% vs 29% – it also (notably) made no difference in the size of contribution.
Behavioural scientists Robert Cialdini and David Schroeder tackled this problem in a fascinating field experiment and found a simple solution. They recruited a team of researchers who posed as fundraisers in a door-to-door campaign across a middle-income suburban housing area. They tested two scripts: a control group were told ‘I am collecting money for the American Cancer Society. Would you be willing to help by giving a donation?’, and an intervention group were told the same, but with the additional line ‘Even a penny will help.’
This tiny addition not only led to an increased rate of contribution of 50% vs 29% – it also (notably) made no difference in the size of contribution. So more people were giving and were not giving any less. Cialdini and Schroeder believe that this was due to what they call a ‘legitimising effect’ – making it ok to give a little (altering what behavioural economists call the ‘subjective norm’ – what we think we should really be doing according to our perceptions of societal norms) rather than directly asking for small amounts.4
Another strategy for raising donation amounts uses the behavioural economic concept of anchors. When making a choice, we are often affected by the context in which we make that choice and the other options available. We anchor to extremes and tend to pick the middle option as a compromise – a phenomenon called ‘Extremeness Aversion’.
For example, Oxfam present three pre-set options for people who want to make a one-off donation online. It changes from year to year – currently it’s set at £18, £50, £100 on the website. £100 probably seems a bit high for the average person, even for a one-off donation, but £18 feels quite low and is perhaps a deliberately odd amount. So donors are subtly directed to opt for the middle option of £50. (The circle is already helpfully marked at £50 too.) Donors are further encouraged to see the £50 level as desirable by its also being suggested in the ‘Own amount’ box and so they are given another helpful nudge in that direction.
4. “I’ll match you” – If I give, they give
Over time, charities and non-profit institutions have come to rely on a simple rule of thumb for fundraising – that of matching gifts, where a $100 donation from an individual to a non-profit organisation is matched by another $100 from another donor such as an employer. Oxfam’s donation page above is
Looking at this technique through the eyes of behavioural science, we might speculate that its success could be due to reciprocity and commitment effects. It is motivating if we know that someone else has pledged to match whatever we donate. However, it has often been thought that the higher the ‘matched’ donation, the more effective it is at getting people to donate.
For example, when a $100 donation is matched by a $200 or even $300 sum from the charity itself (a 2:1 or 3:1 ratio, rather than a 1:1 ratio). Fundraising experts preach that you should ‘never underestimate the power of a challenge gift…and a richer challenge (2:1) greatly adds to the match’s attractiveness’.5
Behavioural economists Dean Karlan and John List, finding that advice on matching was based largely on anecdotal evidence, decided to test this rule of thumb in a set of field experiments. What they found countered some of the conventional wisdom on ‘matching’.
Testing different letters on 50,000 prior donors to a liberal political organisation in the US, they found that although matching did indeed have a significant impact on both response rate and amount donated, larger matching ratios had no additional impact. For 1:1 ratios, the probability that an individual might donate increased by 22% and the amount donated rose by 19%, but 2:1 or 3:1 had no further effect.6
So, using matching gifts to incentivise potential donors is a useful tool, but there is no need to go overboard.
5. Seeing the end in sight
Online funding/sponsorship websites such as Kiva in the US and JustGiving in the UK make a point of displaying not only the amount of money raised so far, but also how this equates to the desired target in percentage terms.
This allows donors to see themselves in the powerful position of tipping the balance
This allows donors to see themselves in the powerful position of tipping the balance – rounding up a current total to a heftier sum – and there’s no denying the satisfaction to be had from taking someone’s total raised so far from say, £455, to a mighty £500 and in so doing pushing the percentage ever closer to the 100% mark. Kiva add to the motivating narrative by featuring details about projects and loans which are close to meeting their target eg. showing that Josefina Del Carmen (see below) has already raised 75% of the money she needs to buy farming goods.
This technique of focusing on percentages is based on more research conducted by Karlan and List. They analysed what difference already having raised a percentage of the fundraising target made to the final sum raised, testing the theory that it would positively impact on the final sum, as part of a fundraising campaign at the University of Central Florida.
Their results were impressive – increasing the percentage already raised from 10% to 67% led to improved response rates: 3.7% vs 8.2% of the solicited individuals. And not only do more individuals respond, but the size of donations increases as well.
- at 10% of target achieved, the average donation was around $15;
- at 33% of target achieved, the average donation was $26; and
- at 67% of target achieved, the average donation was almost $40.
They also found that people gave fewer small donations when a large percentage had already been achieved, and large gifts (defined as over $20) were more frequent.
Why might posting the percentage of target achieved be effective in raising funds though? Looking at this using insights from behavioural science can help to cast light on some possible hypotheses.
Karlan and List speculated for example, that it was used by donors as a signal or rule of thumb for the deserving nature or quality of the aim. This may be due to two different reasons; in the case of the kind of fundraising operated by Kiva, it could be perceived as signalling a commitment to cause from the charity – in that they were willing to devote some of their own scarce resources to it; another reason might be that because private individuals had already made donations this could lead to herding and social norm effects which other potential donors might interpret as a shortcut indicator of the fundraiser’s credibility.
(Interestingly, JustGiving further prompt the social norm effect by including details of the most generous donor to date in their breakdown of individual sponsorship progression, and this can be a compelling message. It’s not donor matching, but it sends a subtle hint about levels of donation nonetheless.)
There may be other things at work though too – reaching goals is initially hard work and donor motivation can be low near the beginning as the prospect of reaching the end goal seems very distant. When we are closer to the end, motivation picks up, so donating to a cause close to its target can be more appealing. If donors can see how their contribution is helping to chunk the progress of the target total it encourages strategic and, perhaps therefore, more generous giving. So our warm feelings of generosity are rewarded more when we can add £10 to £480 of a £500 target for example, than an initial £10 to kickstart the entire £500 target.
We also tend to prefer gains today rather than having to wait for future gains (something behavioural economists call time inconsistency), and knowing that a donation today could result in meeting the target very soon is more motivating. So if you need to raise some money, think about communicating how much has already been raised and how close you are to the desired target, before you go out asking for donations.
Many of us want to give (or want to give more) to charity, but we often fail to deliver on our intentions. As with so many of our plans, there can be a gap between what we intend to do and what we actually end up doing and charitable giving is one of the plans that we can easily let slip; it needs some trigger action.
These simple behavioural tools such as motivating people to give by telling them that others are giving and how much, or manipulating the anchors and reference points for donations, or giving donors the satisfaction of feeling as if their donation has made a real difference, or playing to the way we discount the future are all powerful ways to increase charitable giving and goal attainment very easily, with few additional costs to bear.
These insights are also relevant to nearly all forms of fundraising – from attaining venture capital funding or departmental budgets or maybe even The Behavioural Architects’ fundraising for social projects!
1. UK Giving 2012, National Council for Voluntary Organisations and Charities Aid Foundation
2. Doing Well and Doing Good ‘The Economist’ July 29th 2004
3. Behavioural Insights Team ‘Applying behavioural insights to charitable giving’, 2013
4. Cialdini, R., Schroeder, D., ‘Increasing compliance by legitimizing paltry contributions: When even a penny helps’ Journal of Personality and Social Psychology, 1976, Vol. 34, No. 4, 599-604
5. Dove, K.E., ‘Conducting a Successful Capital Campaign’, 2000 Second edition, San Francisco Jossey-Bass, 510pp
6. Karlan, D., List, J. ‘”Does Price Matter in Charitable Giving? Evidence from a Large-Scale Natural Field Experiment,” American Economic Review, American Economic Association, vol. 97(5), pages 1774-1793, December 2007
Photo: dominoes nudged by Mazzzur on Shutterstock.com