The new Financial Advisory Service - An AI bot for coaching and counseling paired with a rules-based expert systems for financial advice.

From Trust is mutual by Allison Schrager.  The subheading is How to to restore trust in the economy, society, and with your AI-powered financial planner.  She highlights some AI implications I had not considered.

On trust:

As with many problems, I believe the solutions for restoring trust can be found in financial markets. The financial industry constantly loses trust, yet people still entrust their life savings to markets. I take a Mertonian approach. Bob Merton argues that to trust a counterparty, we must believe they are competent and lack nefarious motives. However, in a large economy, we rarely know either of these things directly, so we must infer trust through other means. One approach is transparency—knowing everything about the transaction, such as reading the specs when buying a dishwasher. Another is verification, which is often more practical since transparency isn’t always possible, either because we're not expert enough to make sense of the data, or there’s just too much of it. Verification can come from third parties, like comparing returns to a benchmark, reading Amazon star-reviews, or consulting Consumer Reports.

She discusses the implications of AI on the financial advisor sector:

Good, fee-only advisors tend to focus on high-net-worth clients, but everyone needs good advice. In fact, those with less money arguably need better advice, as they have less margin for error. With AI and current robo-advisors, everyone can get pretty good portfolio construction, arguably as good as from a high-end advisor. But portfolio construction is only part of an advisor’s job. Advisors also act as therapists, having difficult conversations—like telling you that you can’t afford to keep subsidizing your 40-year-old son’s music career, helping with end-of-life financial decisions, or guiding you when your spouse dies, and you have to manage household finances for the first time.

She heads in a slightly different direction in the conclusions she draws because of her concern that these are not "conversations I’d want to have with a bot."  Fair enough.

But for those down market, which is most everyone most of their lives, with little in savings to draw the interest of financial advisors, a bot is likely a very viable alternative.  Indeed, because most people materially and embarrassingly mismanage their financial lives for long stretches of time, I suspect that the opportunity to interact with an AI capable bot might actually be an inducement to seek financial advisory services.  Interacting with a human being under these circumstances is tantamount to a visit to the confessional after a ling absence.  

My wife is a financial advisor (FA)who serves the top end of the market but, through moral convictions and religious faith, she actually has a surprisingly diverse portfolio of clients, from those with virtually no-savings to those with a great deal.  

Schrager is right on three points.  1) Most enterprises structure their financial incentives so as to induce financial advisors to do a lot of trades and sell a lot of fee-based services; 2) Most financial advisors are better sales people than they are either financially savvy or effective counsellors; and 3) the greatest value-add is indeed the counseling service rather than the financial advice.  

The actual financial advice is mostly driven by rules-based expert systems, the FA is just the human face, not the actual expert.  Indeed, most financial enterprises spend a lot of money constraining the action of FAs because most of them cannot come close to the performance of the rules-based expert system.

The opportunity that Schrager nods towards would arise from the anonymity and judgment-free interaction of the beginning saver (early or late in life) with an AI enabled bot which sits on top of the ruled-based expert system for recommending investments.  

It has been three or four years since I last researched AI as a counseling/therapy tool but even then, it was proving surprisingly effective, particularly in specific circumstances.  I would bet that an AI advisory bot now is pretty good at giving the advice and coaching towards it in a way that most FAs can't, won't, or are too expensive to do.

For most people the fundamentals to financial well-being are well established and documented.  They are the Bourgeoise Values of Dierdre McCloskey in combination with the Success Sequence.  

The success sequence is:

Finish high school on time.

Get a full-time job once you finish school and stay employed.

Get married before you have children and stay married.

Do these three things and you only ever have a 1-2% chance of being in poverty, a pretty phenomenally successful outcome were one to measure it as a social policy.  No other social policies, individually or in combination, come even close to such success.

Throw in the bourgeoise virtues described by McCloskey and you have a powerful one-two set of tools for a financially successful life.

Courage

Justice

Prudence

Temperance

Faith

Hope

Love 

Coaching and counseling on these two tools is the predicate to the accumulation of savings which, in turn, is a means of enjoying a comfortable life.  

I would wager an AI bot that coaches and counsels on the adoption and practice of these ten practices in conjunction with a ruled-based expert system for financial recommendations would be an enormous benefactor to the bottom 70% of society who are not well served by the financial system.  

For the provider, it would cheap (after design and establishment) and a great feeder.  Right now, the financial services industry seeks out those who have already succeeded and compete brutally for those few.  An AI bot (for behavioral coaching) linked with a rules-based expert system (for financial recommendations) could establish a relationship far up the chain and long before they could be served by expensive humans.  

The blot on this rosy vision is the reality that knowing what needs to be done (success sequence and bourgeoise virtues) is hardly the impediment.  We all know what we should be do.  We just don't do it.  Will AI bot coaching make a difference?  At the margin, I suspect so.  A wholesale difference?  I am less sanguine.