Session 2: Financial wellbeing

Madelaine Magi-Prowse
Hello and welcome to the second session of BETA's BI Connect 2024 virtual series, showcasing work across the behavioural insights community. Thank you very much for joining us today.

My name is Madelaine Magi-Prowse and I'm a director in the Behavioural Economics team of the Australian Government, better known as BETA. I'd like to begin by acknowledging the traditional owners of the lands on which we are meeting today across Australia. For me in Meanjin, Brisbane, this is the Yuggera and Turrbal people. I pay my respects to the elders past and present, and I extend that respect to First Nations people here today.

As a brief introduction, BETA sits within the Department of the Prime Minister and Cabinet. We work across government to apply behavioural science to a range of policies and programs. Our mission is to improve the lives of Australians by generating and applying evidence from the behavioural and social sciences to find solutions to complex policy problems. A core part of our mission involves building capability. This event is one of many initiatives that we run to share knowledge and build awareness of behavioural insights. You'll see the role of that BI can have in supporting the development of government services, policies and programs. If you're feeling inspired by today's presentations, which I'm sure you will be, please visit our website where you can see further information about BETA’s projects. There are also various tools and resources that can help you learn more about applying behavioural insights to a project of your own.

I would now like to invite Lisa Elliston, First Assistant Secretary in the Department of the Prime Minister and Cabinet, to offer some comments on the importance of today's session on financial well-being. Thanks, Lisa, over to you.

Lisa Elliston
Thanks, Maddie. Good morning, everyone and welcome to the second session of BI Connect 2024.
I'm joining the meeting from Ngunnawal Country and I also acknowledge and honour the deep connection to land of the Traditional Custodians on the lands that we're all joining the meeting from today. And I also acknowledge and welcome Aboriginal and Torres Strait Islander People joining the call.

I'm really looking forward to today's session on financial well-being. The rising cost of living is a significant issue for many Australians. We've seen work across all levels of government to help address these pressures. At the Commonwealth level, the Australian government has delivered initiatives to reduce the cost of childcare and energy bills, and many states have also provided energy bill relief as well. BETA has been working with policy teams across government to support reforms aimed at relieving the cost of living pressures. We've been working with Treasury to help banking customers get a better deal on their home loans, with the Australian Energy Regulator to help energy customers find the best energy plans for their homes, and the ACCC as well to help customers find the best rate when transferring international funds. Behavioural insights play an important role in supporting and empowering Australians to navigate complex markets and financial decisions. Today we'll see a showcase of research where behavioural insights have been used to better develop systems and tools. Last week when the Secretary of PM&C, Professor Glyn Davis, opened the conference, he encouraged us to think deeply about how the knowledge you gain here can be used to prove the lives of Australians. Please take advantage of the professionals who have gathered here today and dig deeper by asking questions and reflecting on the BI approaches that are presented. I'm pleased to open the BI Connect 2024 session on financial well-being. Thanks.

Madelaine Magi-Prowse
Thank you Lisa for those opening remarks. As Lisa mentioned, today's session focuses on financial well-being. Financial stress is impacting many Australian households and we've selected 3 presentations today which are using behavioural science to help make things easier and reduce the burden for Australians as they navigate the current financial climate. The format of today session will be 3 presentations followed by an opportunity to ask questions, so I encourage you to add your questions to the Q&A function which can be found along the top window of your screen.

Our first presentation is by Doctor Bethany Jones or BJ who is the Research Director here in BETA. She will be followed by Doctor Iseult Cremen from the NSW Government Government's Behavioural Insights Unit. And finally, we will hear from Professor Michael Hiscox who is currently at Harvard University. Our first presenter BJ has worked in research for more than 20 years across a diverse range of settings, including private and community mental health, hospitals, universities and government agencies. Welcome BJ, over to you.

Dr Bethany Jones (BJ)
Thanks Maddy. Thanks for the introduction. So I'm BJ as many mentioned and today I'm going to talk to you about a project that Lisa mentioned in passing. And we did hit here at BETA with the Australian Competition and Consumer Commission, which is the ACCC and we're looking at ways to make online foreign exchange calculators more transparent. So this started back into 2019 when the ACCC started looking at foreign exchange and they found that if consumers in Australia were buying 200 U.S. dollars, they could save between up to $40 if they picked the cheapest provider over the most expensive. So that's a really huge difference. So people don't always want the cheapest. Maybe they want the fastest or maybe they want to stay with a provider that they trust. But the key point here is about the transparency and making it easier for consumers to choose. So they want to know like how much is that faster transaction going to cost or what's the premium on staying with that trusted provider? Allowing that transparency makes allows people the freedom to make choices that are right for them. So in 2023, Australians were paying around 5% of their transaction costs in fees and charges. But you want to guess what the G20 target is? It's actually 1%. So we're paying a lot more than we need to for our international money transfers. So we have seen a lot of improvement in recent years, but there's clearly still a long way to go.

So part of that improvement is because in 2019, ACCC created their best practice guidance and it's recommended that providers offer online calculators to allow consumers to estimate the price of a transaction before they're committed to it. So here are some examples of the calculators that are out in the wild at the moment. So they're better than nothing, but you can still see they're still pretty hard to compare. They have a lot of different information. It's all presented really differently and when ACCC looked at this, they thought, you know, we can do better. So they started doing some consumer research to come up with ways that they could be improved. And that's where we came in. They partnered up, partnered up with us to actually test what works. So we did a survey and we embedded like 2 experiments and a bunch of user experience work. We tested a whole bunch of changes to the calculators. But I don't have time to talk about it all today, so I'm going to focus in on a really, really small part. But that was actually the most punchy and that actually did inform policy.

So what do we do? This is one of the tasks from the survey that we actually asked our participants to do. So we asked people to compare mock calculators and pick which one represented the best value. So this is trying to mimic when somebody might go online and choose 4 providers and put in the amount that they want to send and see which one they want to choose. So when we said best value, what we were asking was for people to judge which one delivered the most money overseas or the lowest cost. So I might just pause here for a few seconds because you're probably trying to figure out which one's the best value. I'll just stop talking for like 10 seconds so you can just have a little think through.

I'm afraid you only get 10 seconds. Our participants hand longer than that. They also did this five times with five different sets of four calculators. So if they selected the calculator that represented the best value, then they got that task right. But if they selected one of the other calculators or if they chose, I don't know, then they got it wrong. So I'm not actually going to give you the answer to this right now, but I will actually go through it shortly. But first of all, we're just going to have a like, break this down a little more and see what we changed.

So this represents the calculators that you just saw and this is the business as usual situation. We have $250 that we want to send from Australia. There's a $20 fee. Hopefully you can see the little plus sign that I've circled there in red. That means that the transfer fee is added on to the amount you want to send. So we can see the total that the person's paying here is now $270.00. So we're still converting the original 250. We have an exchange rate that is, Yeah, I'm sure you're all familiar with that. And sometimes at the other end, the fee deducted by the correspondent bank, Sometimes at the other end, the fee at the bank will charge a fee as well. And finally, at the end, we get the amount that the person in the US gets, which is $173.58 US cents. So you probably found that task pretty hard. And you might have noticed that the fees were sometimes added on and sometimes subtracted. And they were all different amounts and exchange rates varied. So it's not actually an easy thing to do. 
So one of the things we tested was what if we just consistently subtracted fees? So in one of our groups, they saw the fees in every calculator, in every task was always subtracted. So here's an example of that. We have the same calculator, the same $20 fee, but it's subtracted from the amount that the person wants to send. So the total cost is still 250, but they're only converting 230 because the provider has taken that fee out. So now the amount the recipient gets is down to $159.69. So do you think this would make a big difference? We thought it would make a bit of a difference, but let's see what really happens. So in the Business as Usual group, before I do that, actually just a reminder that the outcome we're measuring here is the percentage of tasks that people got correct in each group. So each person completed 5 tasks. There were lots of people, so this is an overall percentage in each group for the answers that were correct. So in the Business as Usual group, we can see this was just under 47% correct, which actually isn't too bad. But when we look at when we subtracted the fees, on average of the group got 84.6% correct. Now I think there's probably a lot of researchers on the line. So if I said to you, I'm going to do a survey and I'm going to get people to do 20 mental arithmetic tasks and I reckon they're going to get it close to 90% correct. You should would probably ask me to step away from research because you'd think it was a big bonkers. And in fact, when I saw this, I thought this was wrong, and I spent a slightly embarrassing amount of time trying to figure out where the error in my code was. But this isn't actually an error, it is what happened. So the next thing I did was trying to figure out why we found such a big effect.

Sorry, I'm seeing some chat there. Hopefully everyone can hear me.

OK, So what I did was dig into what people were doing where they got sent a stray, like were they selecting the calculator with the lowest fees or the best exchange rates?

Or were they looking at the ones that sent the most to the recipient or the ones with the lowest correspondent back fees? So we thought if we could work through this, we could unpick how people were thinking about it and give us insights into why this effect was so big. So I'll step you through it.

So this is the task you were just doing. So as you can see, we have the $250, we've got the transfer fees that vary between $10 and like $1.99. Oh, sorry, $20 and $1.99 and the 1st 2 are added and the last two are subtracted, which means everything else kind of changes. Now what we found that is across all groups and all calculators and all tasks, people pretty much just picked the calculator with the highest amount received by the recipient. In this case, it's calculated 2 at $184.03. So you might have picked that one too. You probably did, but you also probably noticed it's also the most expensive transaction there. So is it really the best value? The only way to really check it out is to actually do the maths. So we need to do like a unit cost. So this is a bit like the supermarket. If you want to know the cost of your pistachios, you've got a 300 gram bag and 125g bag. So you have to look at the 100 gram pricing to see whether that special on the 125g bag is really worth it, right? We need to do the same thing here. So we need to figure out how many US cents you get for every Australian dollar that you spend. And we can do that by just doing a simple calculation. So we'll go ahead and do that. And what we're doing here is dividing the amount our recipient gets by the amount that we spent, and we're looking for the result that gives us the highest number, the most US cents for every dollar. And in this case, it's actually calculator 3. So if you look at calculator 3, it's not the highest amount in a recipient gets, it's not the cheapest fee, it's not the best exchange rate. It's kind of not the individual best of anything, but it is actually the best overall. So it's really hard for people because there's no like ‘lay down misère’, just, you know, this one simple trick and you can find the best calculator. It actually just doesn't exist. You just got to do the maths as an aside, because most people did select the most in the recipient gets box is their correct answer. Well, as a selected answer in this business as usual group, that was actually correct for two out of the four out of the five tasks, which means that most people got 40%. So if you think about the result, the overall average for that group was 47%. So suddenly that's not a lot looking quite so good, right? Because 40% of that is just because people were using this heuristic rather than actually calculating it out. So when we subtract fees, why is it so much better?

So this is the same transaction or same activity, but for the people that had all their fees subtracted, the people in that group. So we have the same amount to send and the same transfer fees, but now it's always subtracted. So the total you pay is 250 and everything else has changed slightly again. So again, in this group, most people selected the calculator that had the most money delivered to the recipient. In this case it's calculator 3. So it costs $250 and we can check this again with the same maths. So we do that calculation again, our unit pricing. So we take the amount the recipient gets and we divide that by the amount that we actually spent. But the eagle eyed among you, hopefully it's clear on the screen.

You might notice now that the amount we're dividing by is always the same. It's always $250 because now we've made the total you pay always the same. If somebody wants to send 250, it always costs 250. So we can kind of ignore that, right? Like it's just a constant. So all we need to worry about now is the first number and find the biggest number there. And that number comes from the recipient gets box. So now we've created a world where if you select the calculator with the highest amount since the recipient, it is actually the best value. So we've changed the experience of the calculators to actually fit people's mental model rather than asking people to update their mental model or do something different to fit how the calculators are proceeding information So in this group, in the surface subtracted group where someone was always using this heuristic and choosing the calculator with the highest amount in the recipient gets, they would get 100%. And we did get a lot of people in this group who's got 100%, which is why the average is so high. Oh, that was the best one. I should have highlighted that.

So what do we do with this? Of course, we've published a report because we're BETA and that's what we do.  So you should check it out on our website, behaviouraleconomics.pmc.gov.au. This piece of work is actually much bigger than just the bit I've presented today. There's heaps of cool stuff in there, so do check it out. The ACCC also did a much larger piece of work than just these trials. They did a lot of market research and like surveys and interview work. It's really cool bit of work. So if you are interested in sort of competition in this area, it's definitely worth checking out those stuff. Most importantly and excitingly for us, the ACCC did update their best practice guidance to incorporate the findings from this trial. And it now says not only should they have calculators online so people can estimate the cost, but they should subtract the fees so people can compare more easily. 
So the ACC estimates that this will save Australians $200 million every year when it's fully implemented.

So my take home message is that if you can, you should change the world to suit people rather than change people to suit the world, which is kind of what we're trying to do in BI, right? We're trying to make the world a bit more human friendly. So the little bit of the world that you're in charge of might be quite small, like just subtracting a fee rather than adding it on an online calculator. But it can really help your service users a lot and make their experience a lot better. So that's all for me. Do send through your questions. But for now, back to you Maddie.

Madelaine Magi-Prowse
Thank you BJ, what an amazing outcome and I really love your take away there at the end. We change the world to suit people. I'm sure the audience is keen to know more about this work. So please continue to submit questions for our presenters. We've already had a few great ones come through.

We'll now move on to hear from Doctor Iseult Cremen. So Doctor Iseult Cremen is manager at the Behavioural Insights Unit in the NSW Government and will be speaking with us today about women's financial well-being in terms of retirement income. Her recent research has focused on gender, employment and building public sector capability and behavioural insights. Thank you for joining us. Iseult, over to you.

Dr Iseult Cremen
Hi everyone and my name is Iseult Cremen and I work at the NSW Behavioural Insights Unit at the Department of Customer Service. And a tough fact to follow. That was a great presentation, BJ.

Today I'm happy to share kind of a very classic example of a nudge trial with a large employer thinking about how we can support women's engagement with their superannuation decisions. And if there's anybody international on the line that is kind of our pension system here in Australia. We got some really promising and interesting findings around how employers can do more to support women to engage with their decision making for retirement. But.  guess moreover, what I'm really interested in sharing with you today are some thought provoking findings, which will be very relevant for BI practitioners. Because really on the one hand, what I'm about to share with you is a very familiar and perhaps, dare I say even unremarkable e-mail based nudge trial. But on the other hand, I'm really excited to share, I think as this project was something for us that really challenged our ideas and thinking around defaults. So like the BI powerhouse that we all know and love very well, it also tested our assumptions about how we can support gender equity in financial outcomes. It did allow us to explore kind of the applied practice of BI in a more upstream policy context. And I think the other thing that you'll notice as we go through is we've really intentionally tried to focus on scalability here. So I guess the novelty or innovation wasn't because of the complicated nudge mechanism that we used, but really because of the simplicity and scalability of the intervention. And then I'm going to finish with some questions for you all on the use of BI tools and trying to kind of build the scalability in from the start of your interventions. 
So I'm going to start with a nod to defaults because we know that defaults are kind of a powerhouse in the behavioural insights world. And there are so many different elements of how behavioural insights can contribute to public policy, in financial well-being, but in relation to retirement outcomes. We thought it would be good to explore the power of defaults. So many of us will know that defaults are consistently shown to be one of our most effective tools for changing behaviour. So many different analysis looking at the comparing the effects of different types of behavioural science tools often show defaults coming out on the top. They're the ones with the biggest effect sizes, the ones that lead to the greatest changes. And in particular, one common example that's often spoken about is how changing defaults has been shown to dramatically change retirement outcomes in in countries that use default systems. And this includes both participation rates and amounts of retirement saving schemes. And I think here in Australia we're particularly lucky to have an incredible default system. So if you think about defaults at play within the Australian system, there are a number of ways at which it's interacting. So I guess the system as a whole very successfully leverages defaults through employers. So if you were over 18, typically if you do absolutely nothing, your current employer will pay at the moment 11.5% of your earnings into your superannuation fund. But when we take even a step beyond this, which fund? So employers often nominate a default fund and often we see that people go with the, I know I'm guilty of this, I went with the default fund that my first job offered and then I've never switched since. And sometimes people do switch to their next employer's default option when changing jobs, but often it's not a decision that we're making actively. And then even within the funds themselves, the defaults continue. So people are defaulted into a very balanced investment portfolio, the default insurance options and we go with the flow of these preset options and we're happy. And really this is wonderful. This is the beauty of behavioural insights. Many countries, as I said, don't have such a scheme and then they require a lot of effort on behalf of an individual to set up their pension savings. So within this wonderful system of defaults, I think what we were particularly interested in for our project was really looking at how they actually might be interacting with inclusion and equity outcomes and how changes in the prompting environment might interact with default to lead to slightly better outcomes for some people. So I'll take a little step further into that on the question of inclusion. 
So for this project we were specifically asked to do work on women's retirement income. So A Women in NSW report who are project partners in this piece of work, found that on average in NSW, women are retiring with 42% less superannuation than men. So that's a big gap and it's the gender pay gap essentially compounding over time. So really there are lots of structural factors that lead to this gap in superannuation, including those many factors that contribute to the gender pay gap. So things like imbalances and caring responsibilities, differences in pay, workforce participation and other individual factors. And at the Behavioural Insights Unit, we have previously done work looking specifically at reducing the gender pay gap and boosting the rates of women in leadership. But here we really wanted to look kind of at this complex upstream challenge and really trying to disentangle as a state government who isn't responsible for the policy of superannuation, what levers do we have here and can they actually make a difference? So where we landed to explore gender differences was it was to explore gender differences in engagement with superannuation. And when I say engagement, what I mean is making an active choice within your fund. So not just going with the flow of preset options, but actually going in, changing maybe the fund that you are choosing or the settings within the fund or making additional contributions. So we're calling all of those behaviours engagement. So why are we interested in engagement? Well, we did conduct a survey of about 1000 members of the general public here in NSW and what we found was that women are more concerned than men about their superannuation, which is perhaps unsurprising given that they do tend on average to retire with less. But they also reported not being engaged with their superannuation a lot more frequently. So things like managing actively and making decisions. We also saw that other research from the University of Sydney has previously shown that women are less likely to make any change or deviate from the default settings and they're super compared to men, but also that those who were less engaged in making active decisions actually ended up with poor financial outcomes. So our goal here was to boost the engagement or active choices within superannuation decisions within this beautiful default system to support women to have better financial outcomes in retirement. And, but again, what are our levers here? So we did kind of explore the system, Superfunds is one, but we decided to focus on employers, not only because we are a very large employer, the NSW government, the largest employer in the southern hemisphere, but also because employers are making contributions on behalf of employees. And really the workplace systems can create an environment that acts as an enabler or a barrier to engagement. So what we decided to do was a trial at a large organisation to see if employer nudges could help to boost engagement with super decision making.

So I'll just give you a quick overview of the trial, this standard randomised control trial diagram which you might recognise. We chose an organisation with about 12,000 employees, which was about 60% women, included frontline staff and kind of corporate staff and so on. So quite a nice profile and we randomly allocated them into receiving one of three groups. So we had two groups that received an e-mail prompt or a nudge. The first group were calling the act now e-mail prompt. And I'll show you what this looks like in a minute, but essentially it was trying to encourage people to engage with workplace salary sacrificing. So in this employer they have a scheme where people can make additional voluntary contributions pre tax. So there is a tax benefit to doing it. And so that was the act now group. In a second group, we did the exact same message, but we also had the offer of some free advice. And in the third group, we had no prompt within the e-mail. And in terms of what we were actually measuring and comparing across these three groups, we looked at if there was any change to Vol voluntary contributions through salary sacrificing data.

So we actually had payroll data within the organisation, so we could actually see if the e-mail being sent, the e-mail led to a change in your interaction. We also wanted to look at other engagement behaviours. So not just salary sacrificing, but things like making a contribution in a different way that we couldn't see through our behavioural data and seeking out information and we're doing some sort of other interaction with your phone like changing your insurance and so on. So how we measured that was we did a large survey in the same organisation and we got a very large response rate to a survey, 1200 employees, which we were very happy with. And so that was the trial. And in terms of what the nudge actually was, this is a simple e-mail. The first group got just the top section here, which I call the act now group. In terms of some of the behavioural insights that were in there, we threw a couple of concepts.

We did frame it as investing in your future, some of the preliminary research we did some discussion groups and interviews. And what we found was that for young people, they were particularly hard to engage in superannuation as it's a long way off - we all know about present bias, but we did see some hypotheses and other behavioural research that if you frame it as an investment, it's more engaging for younger staff. We simplified the process so tried to really chunk it down into three simple steps. We tried to make the benefits salient, so just showing that even a small amount can really add up over time. And we also wanted to provide a kind of an anchor or a starting point because a lot of people said that it was just overwhelming. They didn't know what to do. But we're an employer. We're not someone who can provide financial advice. So what we just did was share the average of cross all staff of what was being contributed by others. 
And then we had the second group. So this is kind of where the gender element comes in because when we did our discussion groups and interviews with women in the organisation, what they said was that they were really overwhelmed by the complexity of the decision and they didn't know where to get support. So we didn't hear that as much from the men in the organisation. They seemed to have, you know, social networks and they seemed to go online and really kind of be comfortable with making this decision where women said that they would like more support. So what we did was we had the second group where we offered some free advice and this was a 30 minute free session where people could just quickly book in and very low stakes to try to frame it as quite friendly. And I think as I mentioned at the start, this looks like maybe a salient e-mail, but in actual fact, this was presented in a circular newsletter where there were fourteen other items in the newsletter. So really this was quite buried. And when we actually looked at some of the data, the newsletter itself was read by maybe one in four recipients in the organisation. So it's pretty low impact stuff here. And I think this was because we were designing and keeping scalability in mind when we were designing the intervention, we thought, who is the most authoritative messenger? How can we make this the most salient? And we got some fantastic advice from our director, which is that's not how organisations communicate. And if you want this to be something that goes out every year, you have to rethink about what an organisation is actually like to pick up.

So that said, what did we find? Again, I would like to take a leaf out of BJ’s book and I would like to give you 10 seconds just maybe to jot down one or two bullet points. on what are your expectations, what do you think that we found from the research and what impact did it have on behaviours? All right, so first of all, what did we find? We did actually find that the simple act now prompt, which encouraged people to engage by either making a contribution or seeking more information, led to a significant increase in people making a change to their super salary sacrificing arrangement at work. This was over the trial period, which was three months, and you can see that in the control group, only 2.7% of employees in total made any change. So this is again speaking to the incredible power of defaults. Most people are not engaging with these decisions. It's very low. However, our simple nudge buried in a circular newsletter read by very few people actually did lead to a significant increase in people engaging or making a change. And you're probably wondering, what do I mean by engagement here? So what do they actually do? So this is just a breakdown of what the changes were. And most often we saw that people went in to start a new salary sacrificing arrangement. That was the most common behaviour. But we also saw people increasing the amount, people who were already contributing, increasing the amount, stopping or decreasing the amount. We're setting up a future arrangement and I think with the context of the cost of living crisis, this is something that we were really important as using as our outcome measure. It wasn't just people saving work because we didn't think that that was the right decision. For some people, the right decision is actually stopping or reducing the amount of contributions. So this was really promising and encouraging to us. We also wanted to know why, like for the people who didn't start or start engaging at all, why not? And unsurprisingly in the current context, the most commonly cited reason was affordability, that half the people said that. But also other respondents quite commonly reported administrative burden or sludge. So we're guessing that these are the ones that didn't receive our e-mail. But they didn't understand the process, they didn't have time to figure it out and they had difficulty. So there's things that organisations can probably do to make sure they streamline the process. So that's the act now condition.

Then what did we see with the offer free financial advice and in particular looking at the gender effect. So we actually surprisingly saw that when we split the finding out findings out for gender. You can see women on the left and men on the right here. For women, the ones who were offered the free financial advice were significantly less likely to make a change to their salary sacrificing arrangement at work. And this was a significant effect for women only, but not among men. And we saw this same pattern for the self reported engagement behaviours that we measured through our survey. So I'm sure some of you are nodding along and thinking of course, and others are thinking why did we get this result? And like, why would offering free advice lead to lower engagement among women? Well, I think we had a number of hypotheses. And firstly, we thought, well, we're asking them to do an extra behaviour here and potentially they are going to do their research. But as I said, when we looked at the survey and we actually looked at different advice seeking behaviours like going to websites or speaking to others and actively seeking advice or chatting, we saw that they're in fact less likely to be doing any information seeking behaviours as well. So something about this offer of advice seems to be putting women off just doing anything. So our hypothesis really is that for women who tend to be more time poorer, offering advice might imply that you do need to take this time consuming extra step before you can take action. So we thought that there's something about it's increasing the perceived complexity of the decision, implying you need to get an expert advice, it's implying that you need to take more time before you can act and possibly it's affecting confidence. So in our survey, what we saw was that people's women's trusted source of advice where their partners, family and friends, maybe there's something that we're doing here implying that that's not enough and you need to get financial help. So it might be giving them a bit of a confidence knock.

And finally, we in our survey, we measured lots of things, but I'm just going to pull out some final interesting ones. We also gave employees a simple test of financial literacy. You might be familiar with this, the Big 5 questionnaire. I know that national surveys like the HILDA survey do measure this and it understands things like people's understanding of interest, inflation, risk, diversification. And we did see that there was a gender gap in financial literacy in the organisation. And this is similar to that reported by national household surveys in Australia. And we saw this across the younger staff, the middle-aged staff and the older staff. So really across all age groups. And, and so why is this important or why does it matter? Well, we tried to see if this was correlated with those self report behaviours and what we saw was that people with higher financial literacy did have greater engagement with those super behaviours, but the information seeking behaviours in particular. So they weren't more likely to make a contribution change, for example, but they were more likely to be going out and getting informed, which makes sense because it doesn't cost money to go and do a bit of research, but it does cost money to make contribution changes. So there's still probably things at play there where money is a financial barrier.

So what are the kind of key insights that emerge from our findings and key opportunities? Well, the first one, unsurprisingly, the superannuation guarantee is such an effective default, which is really, really wonderful when we think about it. But maybe some employees might need an extra encouragement to engage more often. So we think there's an opportunity to kind of prompt people to regularly review their superannuation decisions because we've seen that such a simple prompt can actually lead to more people engaging and making changes. And also there's an opportunity for employers to make voluntary contributions easier and more salient. Secondly, the gender effect. So women might be like less likely to engage with salary sacrificing if you imply that they need more time and more advice. So what we thought there was an opportunity here to do was to recommend simple specific actions to women at different life and career stages. So taking out the onus of them to have to go and do more research, but just providing tailored heuristics or rules of thumb at different life stages. And also remember to test your assumptions on what works to support women. It's not always just give them more advice and more help. Thirdly, we did think that removing kind of unnecessary frictions in the workplace or what we call sludge is a big opportunity for employers because it is still remaining a barrier. And we saw that come through in our survey, especially for women. And finally, we did see that people with higher financial literacy were engaging more. So there is an opportunity to boost financial literacy among women of all ages. And in terms of the practical things that we're doing, this is like hot off the press research. So we have got some initiatives under way. We're working with super funds to do that simple, specific tailored actions to life stage and work through those because again, as I said, we can't give financial advice, but we are trying to get really practical steps that don't require a lot of additional research. We're also developing an employer guide that can be shared out across NSW and beyond on what employers can do to streamline and support engagement within the workplace, in particular for women. 
So as I said at the start, on the one hand, it is a very standard e-mail based notes trial, but on the other hand, it was something that for us challenged our ideas around default from the default systems, tested our assumptions about gender and also applied or we intend - I thought it was something that was really interesting because we intentionally focused on scalability and usability at the start. So hopefully that means that it can be scaled up quite broadly and beyond just our behavioural intervention. So thank you so much. I've got two kind of questions to leave you with and please feel free. I've put my e-mail address down at the bottom. And if you're interested in these concepts, the first is starting with scaling in mind. Is it worth testing initiatives that may have a smaller effect size if they do give us maybe a more real realistic understanding of effects at scale? And my second question for you is about defaults. So how can we complement our fantastic, amazing defaults with other behavioural interventions, thinking about how to make them potentially a bit more inclusive? All right, I'm going to stop sharing now. And thank you so much for your time today.

Madelaine Magi-Prowse
Thank you, Iseult. That was a really interesting presentation. It's great to see an employee using BI in that way, even if it is a simple, relatively simple nudge or prompt. It seems like a really meaningful impact for a group of customers or women who really need to have those good financial outcomes in retirement. Just a reminder for the audience that you can submit questions online and perhaps even answer Iseult’s questions that she's just posed to you. 
We'll now move on to our final presentation. Which is with Professor Michael Hiscox, who is the Clarence Dillon Professor of International Affairs in the Department of Government at Harvard University. Michael is the founding director of the Sustainability, Transparency, Accountability Research Lab, the STAR Lab, and a faculty member of the Behavioural Insights Group at Harvard Centre for Public Leadership. In 2016, he was also instrumental in establishing BETA. Welcome Michael, over to you.

Professor Michael Hiscox
Great. Thanks Maddie and thanks to everyone. It's a bit hard to follow BJ and Iseult. Those were great presentations. I thought mine will be, I could do a little bit, you know, something a little bit different today and just sort of talk more generally about financial decisions and behavioural issues and financial well-being. I'm actually in Sydney by the way. I should, I should point that out that at the moment I'm in Sydney for the week and part of the work I'm here for is, is fun financial decision behavioural work with, with BETA, my old team and with the Commonwealth Bank. And I'll talk a little bit about some of the work with the Commonwealth Bank today. OK, See if I can actually advance this slide.
OK. So what I thought I'd do is just a little bit of an overview and then I'll talk about one specific trial that is not quite hot off the presence like assaults, but goes back a couple of years. But we're now reporting the full results from this trial in an academic paper. Now, I thought I'd kind of sort of set the scene a little bit in a couple of ways. And one is, you know, we all know this, I guess looking at the body of work and behavioural, applied behavioural work over the years, including a lot of the work done by BETA and the team in New South Wales, including this recent trial at Iseult presented. And with the focus on under saving and saving for retirement in particular. There's a lot of work now on saving for unexpected emergencies, you know, addressing the problem that roughly 1/3 of Australians, roughly 1/3 of Americans report that they don't have to borrow to deal with unexpected expense. We're looking at overborrowing. We've done a lot of work both at BETA and in work that I've done with my team on credit card debt and repayment of credit card debt, avoiding delinquency and going into collections. There's a lot of work now by the CFPB in the US on buy now, pay later schemes and of course the old work on payday loans. A lot of work addressing those kinds of issues, work on overspending, budgeting, and there's that sort of dovetails with work on financial literacy and the ability to manage finances. BETA has done some great work on problem gambling. This is, you know, an emerging, I would even say exploding problem in the United States with the expansion of online gambling as a legal enterprise in multiple states in the US. BETA has led the way on work on home energy efficiency upgrades. And we did some work when I was there earlier on identifying the most efficient appliances for your home and appliance, you know, efficiency ratings. So there's a lot of work there that is really financial as well as environmental. I'm going to talk about work on excess accessing government benefits. At BETA, we did some work on maintaining access to benefits with reminders about reporting in income as a part when you're receiving unemployment benefits. There's a lot of work in the US now on accessing government benefits, and I'll talk about a bit of new work I'm doing there. And increasingly the work on fraud victimisation. I know ACCC, We talked with them when I was at BETA about helping people to avoid, you know, classic financial frauds. I'm going to talk about new work on that in a moment. Ok, So a range of stuff I haven't even mentioned like BJ talked about, you know, you know, we could call that overspending like problem or difficult sort of forms of financial decisions that are a bit more niche like avoiding fees on from currency transactions. We could talk about tax issues as well as there's a bunch of stuff, but these are the headline ones.

When we think about core behavioural issues and when I teach about this in in my classes, we really talk a lot about cognitive capacity, cognitive overload, cognitive effort that's required to deal with complex financial decisions. How to reduce complexity is obviously you know, part of the behavioural intervention toolkit, including the use of defaults or simple recommendations. We know that some of the implications of this is that include that people delay or even just outright avoid altogether difficult complex financial decisions. And you know, a main one that we're all focused on at the moment, including a BETA sort of home renovation decisions. Like these are big financial decisions. They could save you a lot of money, but of course they are very complex and require a lot of sustained cognitive effort and people seldom have the energy or the time to devote to this, right. And so this has implications for how we intervene, not just to try and simplify these kinds of decisions for people to give them tools to sort of navigate the complexity, but also when we ask people to address these questions. So timing is everything here. So, you know, on a Wednesday at whatever time it is now about noon in November, that's probably not just a good time to approach people to talk about home renovation, right? So we have to think about when people might have the resources. A lot of what we do in financial decision making with behavioural work is to focus on self-control, and this is especially true when it comes to problem spending and the borrowing that's associated with that. We talked about problem gambling before too with the work of BETA. So here, a lot of the focus is on how we deviate from rationality as people, not just in our inability to deal with things in a cognitive sense and the complexity of the issues and how we might try to avoid those. But how our emotional state matters for whether or not we can exercise the kind of emotional regulation that's required to rein in, you know, the, the, the power of the self that wants to spend and feel good now rather than do the disciplined thing. So there's a range of work that's sort of focuses on hot and cold emotional states and things like gambling and overspending as well.

The one I'm going to talk about as sort of setting the scene for the discussion is the new work on cognitive decline with ageing. And you know, this I think is an emerging and important problem. And we have some new tools to try to identify problems early and, and provide interventions. And so that's some of the work that we're doing. You know, I'm still, I'm still working with BETA, which is the joy of my life still. But back at Harvard, I have a set of researchers that that we that have formed what we call the STAR Lab, which is to work not just with governments, but to focus on big companies and increasingly banks that are willing to try behavioural interventions to address things like financial well-being and social and environmental outcomes associated with good decisions that improve well-being, but also reduce carbon emissions or address important social effects of, you know, associated with financial disadvantage, domestic abuse, a whole range of issues. So, so we've worked with a bunch of partners applying behavioural insights and always trying to move them towards a randomised control trial to test these things. So exactly the game plan for BETA, but only in in partnering with companies to do this at scale in a range of areas of financial decisions and other types of decisions as well. We're not consultants. We work purely for access to data and partnerships that produce trials that we can publish in academic journals. You'll notice a few of the names on the on the list here, including David Laibson and Sendhil Mullainathan, who've been out and visited us here to work with BETA at different times. John Beshears and I are doing a bunch of stuff together at the moment in the financial area as, as and working with Leslie John as well. So a bunch of names there that have all been working with different partners of ours. Among the partners that we've done the most work with of late, you'll notice on the top rung there the Commonwealth Bank, Truist Financial, which is now I think the 8th largest retail bank in the US and Bank of Ireland. As soon as I don't know if you have any ties back to the Bank of Ireland, but I'm on a call with them each week and we have a bunch of interesting work going on in Ireland as well. I thought I would just dig down and talk a little bit about one trial and then set the scene for the discussion. 
Our work with Commonwealth Bank goes back to 2016 in fact, and it's focused mainly on financial well-being. It's evolved in some interesting ways to focus as well on energy efficiency and, and issues such as, you know, a vulnerability to fraud and ageing. But the initial work really focused on the sort of core areas I talked about savings, borrowing and, and access, you know, financial management and access to government benefits. So we've done a range of trials sort of address these issues. The one that I thought I would just present here because I don't think I've ever presented it to this audience in Australia. And, and you know, the work with Commonwealth Bank, I think this may be the kind of flagship piece of work that we've done that I'm most proud of. This started back in 2017. We were talking about ways to help people in financial disadvantaged customers at the bank who are facing the real risk of going into collections, in particular. And so this has become potentially more relevant now with high cost of living and concerns about cost of living. And the issue came up where we, you know, there's a classic issue in the discussion of benefits in the US where the low take up rates of government benefits, often below 50% of those eligible are actually accessing government benefits, for most of the major benefits in the US. I work with the Veterans Affairs Administration and it's only about 50% of veterans actually get the health care benefits that all veterans are entitled to and they're entitled to a whole bunch of other benefits. And it's way less than 50% of veterans get access to those. And that's probably the most organised form of benefit in the US. So this is a, we know this is a problem. There's lots of behavioural reasons why people may not access government benefits for which they're eligible. It could be a lack of awareness, so attention could be the problem. People may not actually notice that they're eligible for these programs, that they exist and that they're eligible. It could be social stigma associated with benefits at a range of things.  We zeroed in on the complexity of actually applying for benefits and getting access and thought, let's just let's just try something. What if the bank identifies customers that are most likely eligible for government benefits and tries to basically simplify the process for them? So addresses awareness by sending them a notification about this, but also helps them with the first stage of application. You know, we can't go the whole, the whole complete process because it's a government process, but we could go a fairway and we thought we'd try this first with the low income household rebate in New South Wales because there we could more or less go all the way, we didn't need help from the government on that one. And it's a, it's a rebate that, so back in 2018 it was almost $300 a year for a qualifying resident of NSW. They had to be eligible for a concession card based upon their income. The takeover, it was relatively high. We talked with service NSW, it was, it was around 70% we thought. But that left a significant number of people who weren't getting this quite generous benefit. The process for applying was to call your electricity company, your utility and give them your details and they would process this as a credit to your account. But you know, to, to, to understand your eligibility and how to apply, you had to, you know, go to the NSW site, look it up and find out that this was the process. So we thought, let's try it with the bank. We'll identify a bunch of people who we think are eligible based upon information, their account about where they live, their age and their concession cards, which are often clear from the data and send them a notification in the smartphone. And so we tried this in 2018 and go real quick here Maddy.

This is a classic field trial where we had about 200,000 people. In this trial, we held out a control group of about 23, 24 thousand and then divided the rest into four groups. And we didn't know how to approach these people with a notification from their bank about a government benefit. We thought that would be an unusual kind of approach from a bank. And so the bank was quite worried about how that would go down. And so we thought, well, you know, let's just try a few different things. We had one message that that tried to talk about the trust associated like addressed the sort of trust issues. So, so we said let's see “as a concession card holder in NSW”, this treatment here, the second one, “you may be eligible”. So we tried to sort of try, this is why we're contacting you, we know that you're a concession card holder. We had others that sort of tried to address the present bias by flagging it only takes a few minutes to claim we had the most simple version, which was just no dollar amount at all of here, just get an electricity rebate. You may be eligible and then we had one that was the dollar amount as were the other two over here, but just you may be eligible. So simple, but with a dollar amount. You guys can take your guesses here, following the other presenters about which one you think was most effective in getting people to click ‘Tell Me More’ in their smartphone app. Once they clicked, if they clicked to get more information, we then sent them from these treatment groups onto into two different potential experimental conditions. One was they basically got a landing page that looked like the New South Wales, so service NSW explanation for the rebate and how to get it. So it was quite complex, multiple pages to identify your eligibility and then how to get the benefit. And then we created with the bank a more simplified page that had clicked the call numbers for your utility. So basically you it said call you utility, here's the number for your utility and click and we're put you straight through to the line where you could claim a benefit.

So first question was of which of these notifications worked the best. Here's the page. Sorry, if you click through and got the business as usual kind of approach. This was all the information that was on the NSW Service NSW site. So a lot of text here before you got more pages here about what to do, which was to call your energy retailer, but no help in doing that. Our simple version was this. So if you clicked and got into the condition, that was a simplified process. This is what you looked at, sorry and you got these ‘Click to call’ numbers. So we found I didn't don't have the findings here. I wanted to just be brief, but we found that First off, the message that was the simplest one, sorry, this one here had a click through rate that was about almost 20%. So most notifications in bank smartphone app, the click through rate is under 1%. So this one was just off the charts board. It set a historical record at the bank as the most popular in terms of click through rate notification that they've ever sent out to to customers. The others were a bit lower. They're around 10%. So still incredibly popular, but this simple one with the dollar amount was the one that that won the race. They all increased obviously got a lot of interest, got click throughs and then we had an increase in among those who were eligible and hadn't claimed the benefit of about 10%. So it was almost 100% of those who clicked through actually then claimed the benefit and they clicked to call their provider and got their benefit and compared to the similar their counterparts in the control group who continued not getting the benefit. 
So that was a lot of sort of proof of concept for the bank and that's then we got together with the bank and it said, well, why don't we roll this into a new digital app that we can then send out to our customers who are who we think are eligible for benefits more generally. That basically allow, you know, searches for users, their information to search for the benefits that we think most generous and, and it's most likely that they're eligible for. And then they can essentially get a warm hand off from the app to the application process at the federal government or the state government depending upon where they live. And so we built with them benefits Finder and then we rolled out a trial in which we launched that in a kind of a, a wave approach where we randomly assigned notifications about it to a group of, of customers in different post codes to begin with. And then we and then we after a short delay sent the notification out to everyone so we could look at the effects of the rollout of Benefits Finder as it was introduced in 2019 before COVID. And it's been a resounding success. We can roughly the intent to treat of hearing about the Benefits Finder in 2019 was about $100 a year. For those who actually got to Benefits Finder, they got about $500 a year, it looks like this still roughly you, you go into a benefits Finder in the Commonwealth Bank app and you put in some information and it generates some recommended things like the low income housing rebate. It's still up there that you can apply for and here's how to apply call now. So we're pretty proud of this in that sort of near the end of COVID, there'd been about 1.5 million claims that will commence for a government benefits through the Benefits Finder. We now estimate that it's over a billion dollars in, in benefits that have been generated for people based upon the use of Benefits Finder. It's been extended to small business. So there's a part of the app now that allows businesses to claim benefits. And so it's been, we think, a quite resounding behavioural success in terms of addressing attention and complexity by providing a way for people to see the benefits that they're likely eligible for and getting them to start that process and simplifying that process a little bit.

I'm aware of the time, Maddy, so I don't, I don't want to hold us up too much. I wanted to sort of just flag a few issues for discussion. We're now working with the Commonwealth Bank and the Bank of Ireland and Trueist on a range of issues that are sort of extensions of that earlier work, particularly applications of AI to identifying cognitive decline and vulnerability to fraud and a poor financial decision making. So I'll talk about that in a moment. Applications of AI to helping people access supports and that could be extended to government quite easily now and using large language models to help fill in forms or to have custom customer service agents provide more effective support for people who are applying for things. We're looking at those low interest loans that most of these banks have for rooftop solar and other home renovations that increase environmental, sorry, energy efficiency and also have good effects for the environment. That includes low interest loans for electric vehicles as well. And there's a whole new range of work that is associated with climate risk. You know, the risk of extreme weather, weather events and the fact that most of us don't have full knowledge about how vulnerable we are or our properties are to extreme weather events. So this work is, I think, the agenda in the financial space for us with these partners.

I want to just last thing is flag this, this recent paper by Carole Roan Gresenz and Jean Mitchell and co-authors at the Federal Reserve in New York. And this is really innovative and interesting. So this has started a wave of work with our partners and this is essentially they're using the data that they had from merging credit reporting, which includes delinquencies or late payments on credit cards and mortgages and how it affects credit scores in the US with Medicare data. So Medicare in the US is, you know, medical care for those over 65. And, and so, so there's data in there on diagnosis of dementia and Alzheimer's. And what they found by combining these data sets was that years before there was actually a diagnosis of Alzheimer's or severe dementia, They could see in the data on credit scores, late payments, on mortgages and credit cards. The evidence that this was happening. Essentially that all these kinds of decisions became poorer and hence credit outcomes be became worse in the years prior to the diagnosis. And that that deterioration accelerated right up to the point of the diagnosis. And so obviously this is this is alerting the financial institutions and government agencies to the fact that we can see in the data on behavioural data on financial decision making, evidence of memory disorders and cognitive decline. And so that raises a whole bunch of interesting practical and ethical issues for these financial institutions and for the agencies about what should be done to help protect individuals and their families from declining cognitive ability or capacity that is becoming evident in the data. And so we're talking not only about, you know, poorer decisions in terms of spending money, but also giving money away and susceptibility to fraud. And so that's a really interesting agenda for us all. And I will stop talking at that point, Maddy, and hand it back to you.

Madelaine Magi-Prowse
Thank you, Michael. Those were some really interesting insights across such a range of relevant areas and fantastic results for the Benefit Finder, a great example of how to use BI and leverage the data and customer relationships that a bank has to get better outcomes for customers. And I will just add that that last point around credit scores and dementia is so interesting. Another really great example of what we can see in the data and how we could leverage this to improve people's outcomes. And then those interesting questions about what this actually means and the responsibilities that we have and we can see these connections. So hopefully there are some more questions. I have many questions to ask, but I don't want to hog the floor.

So that brings us to the end of the talks for today. We now have some time to ask our presenter some questions and we have approximately 10 minutes or so for this. I can see we already have some posts in the Q&A function. Feel free to continue adding to those. I'm sure that last presentation has generated lots more questions. I will start with BJ while more questions are being added. So there are quite a few questions from the audience which are asking what will happen next with this work, BJ and have the changes been implemented yet?

Dr Bethany Jones (BJ)
Thanks, Maddy, and thanks everyone for your questions. I don't know about whether the changes have been implemented by providers yet. The updated guidance that the ACCC released was actually only last month. So I think it's probably safe to say it hasn't been implemented by providers yet. It will probably take them a little while to rework those systems, but I think we could be pretty confident that it will be implemented. So this is not legally binding or mandatory or anything. It's just advice from the ACCC, But when they did their first round of guidance, it was fully up, taken up by all like 15 of the biggest providers, which you know, pretty much covers the entirety of the market in Australia. So they'll probably do it again. And we also had a huge amount of interest from the sector in this work, like really, really a lot that we're very, very interested. There's a bunch of disruptors in this field that are trying to like rapidly get market share. So they're really looking for any way that they can offer a better service to their customers. So like I mentioned, there's actually a whole bunch of extra work here and there's a lot of findings that they're interested in implementing on top of just the fee subtracted. So yeah, and watch this space. I think it might take them, you know, maybe up to a year to actually get a get implemented online, but hopefully we should see that coming through.

Madelaine Magi-Prowse
Thank you. Another question from the audience, do you think any other financial calculators may benefit from a similar review, noting that the findings are not completely generalisable, but anything that would be drawn across?

Dr Bethany Jones (BJ)
Yeah, this is a really interesting one too. As you mentioned, it isn't really generalisable. This is a quite a niche problem. But what I think is generalisable is anywhere where the fees are sort of coming on afterwards. So drip pricing for like flights or you know, concert tickets and stuff like that. I know ACCC has cracked down on this a lot anyway, but we still see it floating around out there. And you know, Taylor Swift isn't interchangeable with The Cure or anything. Like if you want to go see Taylor Swift, you're going to see Taylor Swift, right? But flights are a bit more interchangeable. Maybe you're trying to find the best deal anywhere where the fees kind of come on afterwards, which is the problem that we saw here, right? People are saying 250 bucks, but then there's a fee coming on afterwards. It's not part of your mental model. It makes it hard to do that comparison. So I think if you're working in a field where fees are sort of being added on after and it's maybe not as transparent as it could be, this is where you might want to start thinking about these things and how we can be more upfront about the costs and it can help people make clearer decisions.

Madelaine Magi-Prowse
Thank you, BJ, we also have one more for you. How did you identify the heuristics that people are using to make their choice?

Dr Bethany Jones (BJ)
This is a great question in in a nutshell with like hundreds of lines of ugly hacky code. Basically I'm just like digging around in the data going what the hell’s going on? So basically what we did is took our best guess at what people might be doing and then we just pulled apart all the calculators and all the responses and sorted them out and literally looked for patterns. Luckily for us, the pattern of people choosing the calculator with the most which the recipient was a really, really strong effect. It was about 50% or slightly more across the entire experiment of people just selected that calculator. So it was actually really obvious. So it was kind of easy to spot, but we also checked whether people were selecting like the lowest fees. There wasn't really anything going on there. Or like with people focusing on the exchange rate, we couldn't really see obvious patterns there. So yeah, lucky for us, there was a really like massive trend that was easy to spot.

Madelaine Magi-Prowse
Thanks, BJ, really interesting. I might move to you, Iseult. We have some questions coming through and thank you for the ones that you've already answered in the Q&A. I was wondering, do you think about how this might be relevant for culturally and linguistically diverse women or staff in particular and also thinking about women with caring responsibilities versus those without. So those are different considerations.

Dr Iseult Cremen
Yeah, great question and thank you. And I think in terms of the second part of the question, women with caring responsibilities versus those without, we, I guess we kind of can look at elements of this question in two ways. So first, when we looked at the payroll data, so that's the data that we could see whether people were engaging or not. And before we did our trial, we tried to look into what people were doing in terms of making voluntary contributions. And overall, we found that about 12 to 13 per staff, percent of staff make voluntary contributions. But when we dug into this further, what we found was that staff who were on parental leave were much less likely to do any sort of extra contributions. And these were much more likely to be women than men. So I think there was four times as likely to be on parental leave if you're a woman than a man. And then I guess amongst those who were salary sacrificing, what we saw was that those on parental leave tended if they were contributing, it was a much smaller mean amount And so there was less savings definitely. We also did some interviews and focus groups with women and men in the organisation. And I guess something that might be a bit relevant here was, you know, you can actually share your superannuation with your partner, particularly when they're on leave. This is a an option that lots of super funds have and I guess it is to like for, for multiple reasons. But when we spoke to people about this in the organisation who were had either were going on parental leave or leaving parental leave, which is a specific group, we asked really nobody, like very few people had heard of it. For those who had, they said that they didn't uptake this option. And when we spoke to super funds, they said that this was really, really not taken up quite a lot. And we were really trying to understand why because it makes a lot of sense. But really what we thought was that people often see the kind of the cost of parenting as a joint expense. So often they have joint accounts where they will share, like for family expenses, but they saw superannuation as an individual savings account for some reason. So it was something that had followed you from your career and it was something that you were building up quite individually. And I think what we saw then was that when women went on parental leave, sometimes they said that they felt like it was selfish to keep contributing to their super, whereas in the current term, they wanted to prioritise their family or their children. So it was just a theme that came out. And we also then saw with some of the older women that some women were looking back at the gaps in their career that they did take for caring responsibilities or childcare. And looking back felt it was really unfair that they now were the consequences of on their super balance, especially if they had been through separation or divorce. So I think there was some interesting nuances around caring, but I think specifically we didn't pull that apart in the data because of the trial, yeah/

And then the other question you asked us about any differences with culturally and linguistically diverse. And again, I can talk about in our survey, we actually found that staff who were culturally linguistically diverse in the organisation were had the same or higher levels of financial adversity across the organisation, which was really interesting. We still saw a gender gap. It was a little bit smaller, but we actually saw that they were less likely to be engaging in any of the superannuation behaviours. So despite this higher financial literacy, they were not engaging in, you know, information seeking and making contributions or changing contributions or other types of fund interactions. And I guess when we thought about the why behind this, we thought it might make sense because if you're coming potentially from a country like myself, who, who doesn't have a superannuation setting and the people that you're speaking to most often are your partners, your parents, your friends. It might be something that you're as engaged with and you have to kind of learn from the start as opposed to something that's always been part of the more context. So that might be one reason, but I'm interested if other people have any other hypotheses there. And yeah, I hope that answers the question.

Madelaine Magi-Prowse
Thank you. Yes, really interesting points of view there as well. I think just on the women and caring responsibilities, another challenge could potentially be that it's quite could be an uncomfortable conversation to have with your partner to have those contributions. And also the reason why you want to do that could be an uncomfortable one is kind of preparing for yourself and protecting yourself, which is something you may not want to be thinking about really because it's potentially separation that leads to that. Just one more question for you. This is a bit more of a general one, but are you able to share if you or NSW BIU you, I have many insights or evidence on where defaults are most effective.

Dr Iseult Cremen
Thanks Maddy. Yeah, really good point about difficult conversations in terms of where defaults are most effective. Potentially Michael or BJ might have a perspective on this as well. I think BETA have done some great work on harnessing the power of defaults. I think it tends to be the literature suggests in situations where decisions are really complex or where there is like a lots of different choices that you could make and potentially in situations where there is a lot of administrative burden. So super things like superannuation and financial decision making's defaults tend to be really powerful. I know that there was other work done by the Commonwealth Bank, I think maybe last year or the year before showing that again in credit card choices, women were more likely to go with the default credit card that was offered based on like whatever settings. I can't quite remember the details of the trial. But yeah, I think it is those complex decisions where there's lots of choices and potentially some administrative burden. But yeah, BJ or Michael, I'm not sure if you have any additional insights there, no.

Madelaine Magi-Prowse
Thank you, Iseult. Yeah, really interesting. I'll now move over to Michael. We have a couple of questions for you. The first one's probably starting quite broad and I think relates to those points you're bringing up at the end about work looking forward. But what do you see as the next significant Australian financial policy that needs attention from behavioural scientists?

Professor Michael Hiscox
Yeah, well, Iseult, I think you're absolutely right. You nailed it on the defaults and sort of those most those very complex ones with lots of choices. And also, I think if you know, in situations in which delaying the decision seems to be costly, people want to get on and do the thing, the next thing, then the default is just so powerful. Maddy, on that question. That's a big question. I was thinking about that while we're waiting. And I think, you know, often I, when I teach the class, I say like, what are the big, the big things that we're all facing as policy challenges and climate change is the existential threat. And you know, a lot of the behavioural change that's required to effectively address climate, you know, I think involve financial decisions. So we've talked about home renovation, adopting solar in particular and heat bumps and just reducing carbon footprint associated with household energy use is huge. And it could be like if we got all homes there to, you know, to zero carbon in Australia would be knocking off about 20% of total emissions, right. So that that, you know, there's a big debate about structural reform versus individual behavioural change, but just getting the behavioural interventions right to make behavioural change, financial decisions that involve that kind of behaviour embody that behavioural change on environment. I think is maybe the most urgent one in terms of the existential threat. But I'm also focused a little bit now on aging and the problems of healthcare and the associated healthcare costs to the system and to individuals associated with aging and medical care in retirement dealing with Alzheimer's. And, and so I think just like that is a tremendous costs for the economy and for individuals, the costs associated with healthcare in the latter years of life. And that is a classic of behavioural things like what are the things that people don't want to think about, you know, getting old, dying, having to move out of your home to a retirement community or a healthcare facility. Awful things to think about. People don't think about them. And what happens is then a crisis situation that you know, in old age. And that is extremely expensive in terms of the healthcare that's associated with, with, you know, rather than doing the preventive health interventions early, it's doing the emergency work when the situation becomes horrible in later years. And that's the same with not planning for moving out of your home and, and, and having other people help you make decisions in those later years. So I think as a behavioural challenge for policy, the ageing population and the healthcare costs associated with it, that's a big one.

Madelaine Magi-Prowse
Thank you. It's a really good point and I'd like that you've drawn out. It's not just direct financial policy, but financial decision making is something that it goes across all these different sectors and that can have really big impacts on us. So it's really interesting perspective and also kind of going back to BJ’s point of meeting people where they're at and, you know, shaping, shaping these things to really suit how people behave, not forcing them to do something really unnatural. We do have another question from the audience, which goes to why a bank or another private firm of any type would engage with this largely pro social work. So the Benefits Finder and what are your thoughts on that?

Professor Michael Hiscox
Yeah, it's a great question. I get this question a lot with all the work of Star Lab is. You know, what are these companies doing this kind of thing for? And so there's a, there's a lot, there's a lot of interesting stuff here from a political economy point of view. In my former life, I was a political economist. And, and so, you know, you see this, the, the tension for a big companies dealing with the, the shadow of regulation in, in these areas, right? So, so often the government will identify a critical area that has to do with cost of living, for example, or energy efficiency and or sustainability in some form or the potential for sort of, you know, safeguarding citizens with more regulation on how financial institutions operate. So it could be about like what you get in your credit card statement or how credit card is regulated, that kind of borrowing is regulated. And so there's this push and pull right where the industry obviously with tech at the moment, like with content moderation and, and how you know how the tech industry is being treated in terms of what they, what they publish and allow published and are they responsible for the consequences. So that threat of regulation induces a lot of voluntary activity on the part of private players in terms of like demonstrating that they can provide safe products or address the social and environmental consequences of their operations by doing something that you know, addresses that issue in in some way that, you know, satisfies the government and, and avoids what they see as more onerous or, or costly regulation. So there's that push and pull where I think we see that in lots of sectors and we see it in finance. There's also more in a more edifying or promising way. There is the idea that, you know, we don't know in a lot of these areas what more sustainable, more responsible business behaviour achieves. And you know, this is not to go all the way to kind of shared value and all, you know, the kind of the win-win mentality. But the idea that a company may actually achieve business value from doing what quote unquote, could be the ‘right thing’. Helping customers to address decarbonisation and becoming more energy efficient, for example, or accessing government benefits so that their financial health improves. That may pay off for a company in terms of customer loyalty, customer recommendations. The improved financial health of customers is good for the bank in terms of avoiding the costs for the Bank of collections and managing customers who are in arrears. And so, you know, this is a way in essentially, essentially I think for a lot of these companies to explore innovation that may actually be valuable from a pure business point of view as well as, you know, addressing something that customers care about. So I think that's the little area in which we're trying to work at the moment.

Madelaine Magi-Prowse
Thank you. Really interesting. Both the regulation and then also those longer term benefits that that the private sector can gain from some of these things. I think we have time for one more question, so I'll just flip it in. Are there any thoughts about why the simple prompt was the most effective in your trial?

Professor Michael Hiscox
Yeah, yeah, I think that's a good question. You know, we like, we've done this in a bunch of trials. We did this at BETA, a bunch of trials where we're just a little bit too clever for ourselves, you know, where we just like, let's have six different messages and we'll like have lots of version. We'll have, you know, this framing and that framing. And since so many times we just say like basically it all turned out to be the same in this one, they're all the same except that one. And the distinguishing thing I think was that it had the dollar value, which is a salience prime as like, OK, that's a lot of money. And so that pulls your attention in and then very few other words. So, so the cognitive load then it's like very, very low. And so I think that's why that's my hunch as to why it's just like very coherent and simple with something that like is compelling, which is the dollar number. So I think that like that's a general lesson for me in in the future work has been like if we can give a dollar, you know, kind of a magnitude number like this is important - pay attention and then very few other kind of complicating pieces of information. I think that that's a winning message.

Madelaine Magi-Prowse
Thank you. Yes, so we shouldn't overthink it too much as behavioural scientists. You might wrap up the questions there. Thank you very much, BJ, Iseult and Michael for your presentation. It's really fascinating insights from all three in really different areas that you're tackling. Thank you to everyone for joining us today. A recording of today's session will be up on our website soon and we will be sending out a survey later today to get your feedback on the session.

Our last session, if I can just do a plug for next week. So our last session of the conference will be on Tuesday, the 26th of November and the topic will be working at the margins. So please join us to explore how behavioural scientists are working at the margins to better understand hard to reach cohorts and design effective solutions. So another really important topic. We look forward to seeing you then. Thank you.