Next in our series of practitioners is Daniel Egan, Director of Behavioral Finance and Investing at Betterment.com. Prior to joining Betterment he was the Behavioral Finance Specialist for Barclays Americas, and a founding member of the Barclays BeFi team in 2007. He earned his Master of Sciences in Decision Science from the London School of Economics and Political Science and his B.A. (Distinction) in Economics from Boston University, and has authored multiple publications related to behavioral economics as well as lecturing at New York University, London Business School, University College of London, and the London School of Economics. Betterment is an online financial adviser and investment manager which uses brilliant technology to ensure clients identify and achieve their financial goals. They use behavioral finance to ensure optimal saving, investing and behavior along the journey. As a part of that, Betterment is keen to collaborate with academic researchers in studying and understanding their clients’ behavior and it’s impact on achieving their goals.

Tell me about your work: how does decision making psychology fit in it? It permeates our service. We apply research findings regarding the psychology of saving, risk-taking, and investing and use it to guide how we communicate and display information to our clients. For instance trying to reduce myopic loss aversion when clients are assessing performance, and preventing narrow framing when making forward-looking risky decisions. We think about how defaults, inertia, loss aversion and hyperbolic discounting can be used to help clients make the right decisions, rather than trip them up. We even look at how social interactions and peer comparisons can be used to motivate better behavior.

And it informs how we make decisions internally. For example, when making hiring decisions we don’t let our opinions bias each others. We either agree a set of criteria to make a decision by, and do it algorithmically, or we gather opinions anonymously and independently. That way we have honest and independent feedback. We’re very careful to not let group-think run our meetings.

How did you first become interested in decision making psychology? As early as high school, it was one of the most interesting subjects to me. My father is a clinical psychologist, so since I was young I’ve I loved understanding how I made decisions, for better or worse, at all levels. From the tiny nuts and bolts of neurotransmitters to the very high order functions such as understanding probability weighting, discounting, and the genetic component of risk taking, I like psychology because it’s so applicable to my daily life.

I think I’m very lucky that psychology has grown into other fields which I’m interested in such as economics and finance, as I’ve gotten older. This means there are constantly new findings, the field is growing, and there is more and more to learn and apply.

What type of research do you find most interesting, useful or exciting? Research which shows how you can improve things, with a solid theoretical foundation. In general, I have “bias fatigue” – I know that humans aren’t perfect, and finding another “gotcha” bias is not too interesting, especially when it’s very similar to existing ones.

Diagnosing a problem is only step 1. I think step 2 – finding a solution, and showing why or how it can work, is far more interesting. That there is more and more JDM research aimed at debiasing, or using biases fruitfully, is very exciting. The “nudge” examples are obviously a good example here, and I’d like to see more work done on other ways to help people improve their decision making.

Do you see any challenges to the wider adoption of decision making psychology in your field? If things go well, behavioral finance will grow into a fairly standard role within most financial organizations, especially as data analysis and understanding how minor details can make a big difference becomes more mainstream.  I foresee two challenges to this happening.

The first is that some industry players pay lip-service to psychology and behavioral finance without systematically implementing and testing improvements. The hallmark of this is that the institution employs an “expert” (sometimes who doesn’t have a strong background in psychology), talks about biases and findings, and doesn’t produce any evidence of how they’ve improved client outcomes. In the long run it will appear that psychology in finance is vague promises based on gimmicks and parlor tricks. The lack of empirical and experimental evidence of the benefits of a behavioral and psychological expert is the biggest threat to the field in my opinion. The field must prove its value by not only criticizing, but improving things.

The second challenge (related to the first) is that very few psychology and JDM programs prepare graduates for commercial roles (there some exceptions to this, obviously) . Despite the fact that most companies would love to have someone tell them how to improve client satisfaction and client outcomes, and that these jobs give you the ability to actually improve things in the real world, few academic institutions encourage and equip psychology graduates for the commercial world. From companies’ point of view, this means there are few qualified individuals to trust to run experiments and keep both the company and the clients interests at heart.

How do you see the relationship between academic researchers and practitioners? Underdeveloped, but improving slowly.

There are tremendous gains to be had by academics and the commercial world working together. Companies have access to representative and relevant samples which it is sometimes impossible for researchers to access. They have more money to and resources to throw at solving specific defined problems. And they want to advertise how they help their clients. But the often don’t have the theoretic, statistical, or experimental knowledge to originate and execute a solution to these problems. They don’t know how to prove that they’re doing the right thing. Those are the skills a more academic background brings which are lacking.

Far too often we dwell on the zero-sum games and incentives of businesses – not wanting to use our knowledge to redistribute wealth from clients to businesses. In my experience, it’s a minority of commercial enterprises who want this – it’s a very short-term way to make money. Most often, they want to be the leader in their field, and show that they’re better than the competition. Focusing on exactly the projects which make the clients better off is a win-win.

What advice would you give to young researchers who might be interested in a career in your field? I’m very conscious, after writing out this list, it’s probably the things which people dread the most, but it’s honest.

• Take chances and network hard: Probably the most important – ask to meet and speak with people. Pursue an internship with a company you really like, admire, or think you could help improve. Talk to people about their products. Write into companies suggesting (very specifically) how they could improve things. Remember that if your odds are 1 in 20, you only need to do about 40 such meeting before you’re extremely likely to find a great role somewhere. 
• Get comfortable with economics & finance – Think about a company whose product, or even better, service you love. How and why do they get paid? What’s their cost structure? How could they offer a better service, and would you pay more for it? Get inside their head, and internalize their mission – that way you can think of ways to help. You can learn about the most important ideas in these subjects quickly, and they’ll be useful to you not just in work, but also your personal life. 
• Get comfortable with lots of statistical methods (not just experimental and survey design). You don’t need prove the central limit theorem – rather, you need to know when to use logistic regression versus linear, what statistical test to use etc. Be broad rather than narrow, have a good toolkit. Check out Regression Modelling Strategies by Frank Harrell – my bible on the topic.
• Get comfortable with programming. This is probably the one you least want to hear, but which has been most important to me. Being able to do analysis quickly, and effectively requires knowing the tools. No, Excel is not good enough. I highly recommend R for practical reasons. It is free, so you never have to convince someone to buy you a license. Many people use and contribute to it, which means it is well documented and supported. Getting good at data management and analysis in R is an investment – it requires some sacrifice up front, but will pay off high dividends later. Also, programming forces you to think in a very exact and clear manner, which is useful unto itself.

 Want to read more? Try these…

• Introduction to In The Wild
• Advertising guru Rory Sutherland on behavioural economics
• Decision making and advertising with Matthew Willcox
• Introduction to Outside The Matrix
• Data science at Google with Paul Litvak
• Interview with Richard Thaler