This was originally posted on Medium. Moving here for linking convenience.
Jon Stein is the CEO and founder of Betterment. He was interviewed by Patrick O’Shaughnessy on the Invest Like the Best podcast. Here are three things I learned.
1/ Good nudging
Stein explains that Betterment has a Tax Impact Preview for trades.
“We try to help you make good decisions around that (trades). One of the things we’ve done is when you’re about to make a transaction we’ll tell you what are the likely taxes you’ll pay on that transaction.”
This doesn’t stop you from trading on the Betterment platform, but nudges you to not. It works too! Stein said that 75% of people with costly tax implications choose not to go through with the trade, and trading more frequently leads to lower returns.
Nudging is a good tool for this because, as Richard Thaler says, “because we are dealing with humans rather than Econs, they sometimes need help.” People are good at many things, calculating potential tax liability is not one of them.
The case for nudging — any nudging — begins with the idea some option has to be the default. Researcher Brian Wansink has found many ways to change how much and what you eat. Each way could start with a nudge; default to smaller plates, opaque food containers, not clearing dishes. Even the positions of food in a cafeteria line — high or low, front or back — affect what people put on their trays.
Stein has nicked nudges from cafeterias to save people money.
2/ Talk to customers & scratch your own itch
“I was consulting to banks and helping them with product management, risk management, investment portfolio policy and things like this. I found through that experience working with banks that I learned a lot, and I loved it, but I found so often they were building products without talking to their customers.”
When I wrote my book about failed startups one of the most common pitfalls was a failure to understand customers. Founders were intimidated, they talked only to friends and family, or they didn’t ask the right questions. Stein saw this not at a startup, but big bank.
A lot of companies start because a founder wants to scratch their own itch. Sometimes this works.
Naval Ravikant created AngelList. Barbara Corcoran’s Shark Tank investments are itch scratchers. Mark Zuckerberg started Facebook. Etc.
Scratching your own itch is a fine way to start.
But, it’s not enough. You have to get to this moment:
The founders that I wrote about failed to get to this point. They asked friends who were just being nice. They asked people who said, “well it’s not for me but it would be great for my brother-in-law.” In reality, it’s bad for everyone.
Building something without talking to your customers is like a blind pig looking for nuts.
3/ Slow down when things are confusing, complex, or unrelated to the past
Rather than a quote about Betterment freezing trading during Brexit, here’s the WSJ post:
Here’s what Stein had to say about the matter:
“We always look at volatility and thoughtfully trade for our customers. We always pause trading at the open because markets and ETFs tend be quite volatile. We watch out for things like that, and Brexit was one of those days when all the major banks paused trading because they were worried about volatility.”
I actually like this strategy because it’s aligned with what Daniel Kahneman has said.
“When things get really big and you’re really not sure, slow down.”
There are situations (with limited volatility) where humans can keep up with changes in the system. Sports is one, fire fighting another, piloting a third. We can succeed in those domains with lots of practice, feedback, and consistent condition.
If we don’t develop an adequate skill — fighter pilots call it being “behind the plane” — the system will eat us up.
When Gene Kranz worked at NASA during the Mercury, Gemini, and Apollo missions he wrote that they trained rigorously because, “During a mission countdown or even a flight test, so many things would be happening so fast that you did not have any time for second thoughts or arguments. You wanted the debate behind you.”
Kranz’s rockets went from 5 miles/minute to 5 miles/second. Trading data is 30,000 times faster. It’s too fast. Acknowledge your limits and slow down.
To recap:
- Decision options always have some sort of default. Arrange the default in such a way to nudge someone into what’s best for them.
- You can start by building something to scratch your own itch but eventually you have to get people to pay you. The sooner you prove/disprove this, the better you’ll be.
- In similar situations (“Same as the gym back in Hickory”) practice, checklists, and feedback work. When things move fast, the situation may have changed and you don’t know it. In that case, slow down.
Update 11/14/16: The first version of this post misspelled Jon as John.
The Waiter's Pad 
Perhaps the third point should be restated as “buy time” when crazy happens. Otherwise, we just got OODA-loop beat by our opponent, and slowing down just means we lose more. c.f. Meditations On Violence by Rory Miller.
LikeLike
I like that. How do you buy time though? Move faster through the loop?
LikeLike
Depends on what we’re doing. Like you said, there’s training. Other options include:
From Go, there’s the idea of “sente”, which is throw something back that must be dealt with by our opponent.
Boyd would say throw something the opponent hasn’t seen, increase their time to orient.
In other scenarios, one can often create a temporary patch that affords time to reason.
In trading, it’s not like high-volatility days are unexpected. Since we know they exist, we should have a plan for detecting them and profiting from them. While ignoring the market is certainly one response, it seems unlikely to be the optimal.
I don’t know what your media queue looks like, however Meditations On Violence by Miller and Deep Survival by Gonzales should probably be near the top as their works generalize to many “buy time” situations.
LikeLike
[…] Jon Stein told Patrick O’Shaugnessy that the Betterment service will display potential tax costs associated with trades. This extra bit of information stops three out of four people. […]
LikeLike