Wednesday, February 28, 2024

Brutal...

 It seems as though a payment process is a (relatively) easy flow to sketch. There are many out there, but here is one:

http://mqs.gtpl.net/mqsubscribe/Help/Online_Payment_Flow.htm

Here's what's funny (not funny "haha" but funny "hmm") to me. This is one of the biggest money grabs in the world, almost $32 billion in 2021 alone. This explains why the space to process payments is so crowded and why there are so many companies acting as middlemen. In fact, the payment systems in the US are quite inefficient compared to those in many other developed countries.

But nobody wants to work with the consumer. Like... nobody.

I mean, I think this is true generally. We, as consumers, don't get the new new thing from companies because they think we are stupid, stubborn, unwilling to change, etc. etc. - and in many cases they are right. On the other hand the enterprise side of this world is loading up on new technology and the consumer has a government filled to the brim with octagenarians on his/her side. Rough stuff.

There are some basic ideas that could help the consumer quite a bit here. In particular, what if I want to change my bank or I get a new credit card. In all of this mess shouldn't I just be able to swap out the account and routing numbers for new ones and carry on with my life, payment rails intact? The answer seems to be "yes" but you can't do it. 

Why? 

Well, I suppose it's mostly because JPMorgan Chase won't build a tool to let you jump ship to Wells Fargo in a heartbeat. Admittedly you'd need to have an account at WFC anyway, so that gets in the way. But on the other other hand (back to the original hand?) you have to complete (basically) the same document every time you open a new account. You can't just say "see my existing account @ Bank Y instead, please."

SO if you want to open a new account you have to complete the same document again. If you need to change something you have to tell everyone one at a time. If something changes in your life and you don't tell these companies, then they never find out and keep at you with their data-driven marketing that is now more off-target then it was when your information was correct.

Brutal... 


The two questions of flow

 We find a lot of value in books that were written a long time ago but usually ignore works that are only a few decades old. To our peril...

I have found some interesting ideas in a bunch of books from the 80s and 90s that I think are not really put into practice as they "should" be in business.

Today I was listening to "Beyond the Goal" that is a follow on to (obvi) "The Goal" from 1984(ish).

Interesting concept:

1. A technology can be useful in improving life if and only if it solves a limitation...

and

2. Before we have a technology to address a limitation we worked around it and usually just considered it a "fact of life."

Here's the problem. When you are living in conditions and rules from #2 and you develop the tech in #1, if you keep operating in the world of #2 then the tech won't work - or at least it won't deliver the value it could.

You see this a lot in commentary that technology hasn't really helped the world improve or businesses to get efficient. But this is exactly the problem these two ideas describe. A new technology is developed and inserted into (usually large) companies who only implement some of it. THis is mostly because many of them are generating a lof of income/profit/cash and so they aren't going to risk that with a new tech. Meanwhile, the old rules and processes prevent the new tech from really changing how things are done. 

We're not stupid. We're just creatures of habit.

End of the Road...?

ZIRP (Zero Interest Rate Policy) created many distortions in the world. 

When it comes to technology and fintech companies, there are two big ones:

1. The biggest companies hired thousands of engineers and stuck them in the middle of the company to work on *meh* projects. Now those people are in the market carrying those names on their resume and it's hard to tell just how good they are (anymore...).

2. Business models that cannot survive were perpetuated. Some examples are fundamental to the business and others are a broader problem with the concept. You can see this across technology but in fintech you can:

- cancel someone's subscriptions until you've done all of them and then...

- negotiate down someone's bill until you've done all of them and then...

- get rid of nuisance fees until you've gotten rid of all of them and then...

- eliminate trading fees on stocks and then...

- reduce investment fees to almost nothing and then...

... and then... you're done.

The argument is supposedly that the underlying user data that you get has a lot of value in and of itself. I find it unlikely that this is true for most of the companies created (and some acquired) for the ideas above (and more). 

Sunday, February 25, 2024

Venn is the New Black

 Simple isn't always the answer but it is usually the best place to start.

If you draw a Venn diagram using three circles:

Financial Services - Bank, investment, insurance, payment, tax, and lending products

Merchants - Goods and services

Consumers - Personal identifying information, contact methods, address, family members and structure, contacts, employment, and balance sheet

--------

Financial Services + Merchants - Financing

Financial Services + Consumers - Account opening, KYC, suitability, qualification, maintenance, and documentation

Merchants + Consumers - Logistics, payment type, relationship, billing, documentation, purchase venue, and warranty.

--------

Financial Services + Merchants + Consumers - So... what is in the center?

YOU are in the center, which extends your consumer circle. You are the only one with all the information in your head. You are the only one who knows when something changes (but you may not know if that change matters). You are the one who has to do something about anything that happens.

Today you can hire a dashboard to aggregate most of this information. But if you are like most people - literally if you consider the usage statistics for these kinds of platforms - these products will create more work for you and will likely relegate your accounts to registered but inactive (read: not valuable). This is one step away from the incumbent models.

On the other hand, my sense is that we are not ready to take three steps into the future, where everything is automated. In the same way that driverless cars may someday bask in the sun, fully automated consumer finance platforms may emerge. But, we are not ready for either yet.

So, I ask you what is two steps into the future?

My answer.

...

Coming soon!


What Do We Really Know About Technology?

 One of the reasons I'm interest in a fintech company is that we don't really know how impactful technology "should have been."

Wut?

People who do no think that innovation in established businesses is important point to the lack of measurable and significant impact across industries. But I don't think you can lay this at the feet of the tech.

The responsibility here lies at the feet of management in two primary ways. 

1. I've sat across from the CFO of two of the largest banks in the United States. Both before COVID hit. Each of them regurgitated the old MBA line that they use capital to buyback stock when it is viewed as a good use of capital (using the cost of capital as a guideline). Then, COVID hits and nobody is ready to even consider implementing remote work. How can a bank like JPMorgan, with an $8 billion+ technology budget miss this? Because the mone is being spent to directly benefit shareholders, rather than making employees more efficient and effective which would impact shareholders but indirectly.

2. Tech is never fully implemented. The reverberation would have been much more significant in terms of jobs and other issues. INstead enough money is spent to "adopt" new technology without disrupting the business as completely as it could. Is this a bad decision? Not if you are printing billions of dollars a quarter. But it does blunt the potential impact of technology on the business.

What should have the impact from technology advances over the 20 years have been? I don't think we'll ever know.

Technology for Me but Not for Thee

 Not only is the consumer consistently underrepresented in existing markets but it is also starting to happen (again) in new technology markets.

I get it.

Nobody wants to build for the consumer because he is unwilling to change, uninformed, dispersed, unreliable, etc. 

And sure, we (you and me) as consumers are responsible for changing that reputation if we want to get "the good stuff."

That is why everything I'm seeing is about building AI and LLMs to help existing enterprises do their jobs better. I heard this a while ago (maybe 2-3 years) and almost nothing has proven to be more accurate:

Incumbents don't adopt technology to change their business model. They adopt it to make their distribution channels (existing and perhaps new) cheaper.

So yes, there are a ton of AI-driven assistance being built. Trust me, there are more than you can count. As I expect(ed) they are falling flat because they were a rush to build "something I can sell to Google." There has not yet been a great consumer-focused company that disrupts the enterprise and gives the power-to-the-people (TM). 

This is the opportunity no matter the industry (almost). This is the opportunity in fintech. Not just to use technology to change the landscape but to change the business models that we confront. In the case of fintech, to truly be able to own, understand, and manipulate our data to match and change each of our unique circumstances.

Saturday, February 24, 2024

DevOps

 As a non-technical person this is a difficult topic to approach. While it's not realistic to expect the ability to code to (ever) be on par with engineers and other people who have made this their career, it is important to have an understanding of how these ideas fit with each other. 

The Dev Ops handbook, while a bit dated is a good start. There is a strong desire to highlight that IT and other knowledge-based work is just as subject to the operational procedures governing manufacturing. It still sounds reasonable even if it is par for the course among technology companies today.

I can say with confidence that these powerful ideas have not found their way noticeably into any of the large banks I have worked for in my career. So, they have not gotten everywhere they could be useful... yet...

Friday, February 23, 2024

Generations

 I think it's easy to focus on youth. Millenials, Gen Z, and Alphas are always targets of new technology. 

But, when it comes to financial services the generation with most of the wealth is the Boomers, who are old (65+).

There is a growing focus on "eldertech" that I think should be incorporated into new ideas built for the future.

https://eldertech.org/senior-apps-10-non-technical-tips/


Where to start...?

 The problem with dropping a new company or product focused on financial services is that you are dealing with important personal information. It seems difficult to imagine a company coming out of nowhere and asking people for their account numbers, SS#, and other information that is important to keep secure. The vast majority of people don't have the ability to evaluate the security any new organization puts in place either so, like the rest of finance, it is a trust game.

It seems like there are two ways to address this:

1. FInd a way to establish trust immediately. This is probably best done through a partnership with a large and well-known institution that is already trusted. This sounds like a good idea unless it is anathema to the idea that you want to build.

2. Start small. Super small. Start relevant to your niche but not based on the information above. Something that is not as key and that won't be as difficult to get someone to share with you. 

Point #2 seems more interesting to me and there is no shortage of interesting data points around people's financial life. The question is what is the right place to start?

The Problem with Dashboards

 The personal finance dashboard has become a well-recognized tool since Mint appeared in the late 00s. That said, it is not the powerhouse that I think many expected it to be. My view on this is informed mostly on the information coming from Mint and the new competitors in the space.

Mint boasted ~15 million users in its marketing materials in 2023. All indications were that the company really had 3-4 million active users (people who accessed their accounts on at least a semi-regular basis). That is a big gap! Many other budgeting and personal finance apps were built in Mint's wake but few ever achieved this scale. Mint was shutdown by Intuit in early 2024 and so there has been a scramble to collect those customers by new and existing players in the space.

January data on SimilarWeb indicates a surge in visits to Mint competitors with Monarch Money, a company founded in 2018-2019 by Mint's former head of product as one of the winners with 3.9 million visitors.

Here the problem...

1. Information is not knowledge - A dashboard full of data is more convenient than a bunch of different platforms that need to be visited one at a time. But, a read-only database of information is not necessarily actionable not is it really any easier to use given the lack of context generally associated with the information presented.

2. DIY - Most budget and personal finance apps fall short for the truly motivated who DIY this with excel. The general flexibility and usability of (still) the best business software created is far superior when the goal is to aggregate data and manipulate it within a closed environment.

3. Work - This is a lot of work. In essence, though there may have been some growth in the overall number of customers using personal budgeting and finance apps over the last 10-15 years since Mint was launched, I don't think anyone would argue that we've seen a significant adoption of these platforms. They are too much work, which is why there was such a huge gap between the number of users at Mint and the active users. I had an account at Mint but I never used it. Once in a while I would go back in to look at what it said and the mere idea of trying to update made me close it back down. I think that for the wealthy, the HNW families as they're called in the finance world, these dashboards are just another set of to-dos that are ultimately not particularly rewarding. 

And why should an interest in personal finance to high financial literacy be the bar? Most people want to do their jobs or sit on social media, or be with their family, etc. The idea that you cannot be financially efficient and successful without being educated and engaged seems like a bar that is too high for many, particularly when the idea is that you have to spend at least a few hours every month slogging through this information even when aggregated.

Dashboards are one step forward but one step too few to be really useful. The user statistics bear that out I think.

Cost to Serve - The Numbers

 Taking some of the constituents of the XLF (financial services ETF) produces the following breakdown of operating expenses or "non-interest expense" in the case of banks:

Expense                                            %

Compensation                                57%

Occupancy                                       8%

Technology and Equipment            11%

Marketing                                         4%

Professional Services                       8%

Other                                            Remaining

These statistics are interesting and reveal the structure of the costs that these companies have built to deliver their services to customers. Obviously these numbers are skewed away from retail customers in some ways because they include the numbers behind commercial and investment banking. It would be reasonable to expect that the pure retail side of the a bank would not have quite as high a compensation expense and might have a much larger RE footprint to boost the occupancy expense. Some of the smaller banks bear this out, though there are few publicly traded pure-play retail banks.

Most of the more interesting fintech companies (e.g. Chime) that might show different results are still private and therefore not available for analysis. These companies specifically targeted incumbent cost structures in their strategy so it's likely that the mix is different and more heavily focused on technology and away from occupancy. 


Wednesday, February 21, 2024

Pricing - How Much?

 It seems like a good idea to keep in mind the target customer when we consider the way and amount we want to price.

Let's say that we want to price this service at $150.00 / month ($1800 annually). If you are already paying $18,000 per year in fees and taxes for your financial life then you are going to increase your annual cost by 10%.

But...!

You are probably saving yourself more. 

Let's say that you think that an hour of your time is worth $500 (how you get there is unique - maybe that's your job or you just sort of have this idea in your hear). If you need 3 hours a month to manage and react to all of the different aspects of your financial life, then you are spending $1500/month.

If a company can offer you a 66% reduction in time, in this case $1,000, how much would you pay for the service?

To be clear, you are still spending $500 a month on your financial life. You are also still spending $18,000 annually on your financial life for fees and the like. 

But if you pay $150/month for this service, you are keeping $850 each month of the value that the company has created for you. So, you are paying $1800 more in fees but you are saving $10,200 annually of indirect costs and you can go further by spending that time with you family or whatever it is that adds even more juice to this equation.

Party on Wayne!

Party on Garth!


Pricing - Structure and Concept

 The correct way to price any service has to be (i) in context of the industry, (ii) competitor aware, and (iii) appropriate for the service/product.

1. Context - As stated earlier, the estimate for banking Cost-to-Serve is $550/yr/client, which is $45.83. This is not charged directly to the customer but is something that is felt in either lower returns or higher fees on products. Companies do, after all, just pass costs through to their customers.  

2. Competitors - There are a lot of competitors in various parts of the financial services industry when it comes to dealing with people. Some of them are free (or ad supported) while others have a periodic fee. TO go beyond the cliche "if it's free you're the product" I also think that if you are a customer and you aren't paying for something then you don't have a seat at the table. You are not in a position to be well represented when you are willingly using a product or service that you don't pay for.

3. Appropriate - Financial services must be bound by the realities of the first two points above. On the other hand, there has to be a strategy of its own. I recently read Game Changer on pricing and until I find more nuance or a better theory, this is the framework I'm going to use.

a) Most important driver - Value and competition (excluding costs)

b) Industry Structure - B2C has many buyers with a mostly shared set of problems - there are only so many when it comes to financial life. Suppliers are more constrained but barriers to entry are low and there are new companies targeting slivers of these markets all the time.

c) Best model(s) - Value (how much is created above competitors, what is it worth, and how is it split between the company and customers - see: Apple) and Choice (Building a set of features that help to further segment the market - see Salesforce).

Sunday, February 18, 2024

Cost to Serve as Opportunity

McKinsey estimates that in the United States, it costs ~$550 per year to service each customer or $45.83 per month. I have found this data to be approximately correct in reviewing the financial statements of several banks in the US that are publicly traded (JPM, Citi, BofA, WFC, USBank, Alpine Bank, Morgan Stanley).

These numbers can be divided between compensation costs and other costs, which are mostly technology, marketing, and occupancy (real estate). You can't get all of this for the retail-only business lines but you can make some assumptions and get some estimates that are useful guideposts.

That is an interesting number from the perspective of a new venture. This is because ultimately incumbent service is neither very good or very cheap. It is built of legacy systems that are held together by masses of people whose only job is to make sure these systems keep talking to each other. 

After decades to properly integrate all of this, most banks have not yet done so. Even with the coming of AI it seems a job that is both unappealing and unlikely to be done. The sheer expense of this kind of effort alone is probably enough to dissuade most CEOs (note that JPM already expenses ~$8 billion on tech each year, which does not include the purchases that they capitalize.

If you can provide objective and platform-agnostic service that is better and hit this target, the one that these banks already pass onto their customers anyway, you could be in business.

Literally...

Financial Life (v1)



 This is a good example of financial life. This shouldn’t be stuck on an ad on X. It should be a tool, a lens through which you can see this offer and its details before you even lift a finger to start opening an account. 

It’s time for the era of fine print to end…


Financial Supply Chain - Asymmetry in Action

 For anyone interested in business, supply chains are an important concept. 

The traditional idea takes you to a large field in the middle of nowhere, to find a huge building churning out whatever. Taken to the next step puts you (probably) on a plane to many parts of the globalized world to find the raw materials that make up the inputs for the factory you just left.

Though it looks quite different, there is a supply chain for financial products. Instead of smokestack factories however, it is composed mostly of people sitting on high floors of buildings in major cities (e.g. lawyers, financiers, regulators, and more lawyers). 

Just like in traditional manufacturing, every link in that chain adds cost to the product that is being sold. 

Just like in traditional manufacturing, these products must be sold for a profit (calculated as cost+ or some other methodology).

If this is true, then the price that is charged for the product must be influenced by the cost. The difference in financial services is that there is a variable revenue stream in many cases. 

As an example, a mortgage costs $X to put together. If the bank wants to make a 10% margin, then it sells for $X*1.1. But the revenue that the bank makes is based (mostly) on Net Interest Margin (NIM), which is the difference between the rate they pay on deposits and the rate they can get from (in this case) this loan. The price of a mortgage is just all the projected future cashflows discounted back to today, using (again) interest rates to do the discounting. 

You can see that if the supply chain were cheaper (had fewer steps or was more efficient at each step), then the price of the product would be lower, which in turn would lower the price that the bank would need to sell it for. In essence the supply chain, just like in traditional manufacturing, influences the price of the product irrespective of market conditions. It is also reasonable to assume that this only hurts consumers - that is when the product "should be" cheaper, banks do not let the price fall whereas when the product "should be" more expensive they let the price drift to "market levels."

"Hey Siri, define 'asymmetry'"

Net Interest Margin at the Core

 It strikes me that one reason people have so many financial issues is that the definition of "financial life" is too narrow. There are a lot of topics and issues that are important in the traditional world there are others that are not and cannot be addressed by those companies. For example, banks will help you buy your car or house but play no role in maintaining them, keeping documentation in order, etc. after you do it. 

This makes sense. Banks are really "good" at one thing - they take short-term deposits and pay them X% then lend that money to other people at Y%, collecting the spread (eg Net Interest Margin). While they have built and developed other capabilities, those activities are not their core skill (see: revenue composition of any bank). 

So long as the loan making them Y% doesn't fail, their model works. Why would they add new costs and problems to deal with voluntarily?

A different kind of company can do it though...

Thursday, February 15, 2024

Cost -> Profit Center

 In the true goal of all companies today, transforming a cost line into a profit center is the ostensible goal of many management teams. 

This can be used both within a company that exists but also to help understand how to enter an industry. Interestingly, this goes beyond the idea of merely cutting the costs of the incumbents (see: Neobanks and fintech companies vs. incumbent financial institutions). It goes to the idea that a cost line is an available revenue line instead of just something to be cut.

To wit, McKinsey:

https://www.mckinsey.com/~/media/mckinsey/industries/financial%20services/our%20insights/reshaping%20retail%20banks%20enhancing%20banking%20for%20the%20next%20digital%20age/reshaping-retail-banks-enhancing-banking-for-the-next-digital-age-full-final-v1.pdf?shouldIndex=false

If this is correct, then by 2020 estimates there is a $550 cost per retail banking customer available to be another company's revenue. Today, this is $550 spent with inefficiencies, poor service, obsolete technology, etc. etc. It would seem that a much better version could be offered out of the box without the need for all of the capex that would make its way into this line if an incumbent tried to do it...



Starting Up - Banking Identity... (pt 2)

 I remember someone telling me once that the government, or congress, writes legislation. But that document has to be turned into rules and regs by the various agencies and groups that deal with the policy. Turns out, that is true for this question as well.

The last item for identification is "other relevant information," which is broad and vague to allow for breadth. It turns out that the site for the BSA/AML manual to direct examinations 

https://bsaaml.ffiec.gov/manual

both goes into more details and send the reader to The Code of Federal Regulations

https://www.ecfr.gov/current/title-31/subtitle-B/chapter-X/part-1020#1020.220

It turns out that there really isn't a more detailed requirement beyond the Big 4...

1. Name 

2. Address

3. Date of Birth

4. Identification Number (e.g. SS#)

But you can imagine that a bank would want to get information beyond this for anyone working or making their money from something less than up the middle of Broadway (e.g. pawn shop).

Wednesday, February 14, 2024

Starting Up - Banking Identity...

 It looks to me like the PATRIOT Act requires that banks both verify the identity of account holders and maintain a record of identifying information:


SEC. 326. VERIFICATION OF IDENTIFICATION.

(a) IN GENERAL.—Section 5318 of title 31, United States Code,

as amended by this title, is amended by adding at the end the

following:

‘‘(l) IDENTIFICATION AND VERIFICATION OF ACCOUNTHOLDERS.—

‘‘(1) IN GENERAL.—Subject to the requirements of this sub-

section, the Secretary of the Treasury shall prescribe regula-

tions setting forth the minimum standards for financial institu-

tions and their customers regarding the identity of the customer

that shall apply in connection with the opening of an account

at a financial institution.

‘‘(2) MINIMUM REQUIREMENTS.—The regulations shall, at

a minimum, require financial institutions to implement, and

customers (after being given adequate notice) to comply with,

reasonable procedures for—

‘‘(A) verifying the identity of any person seeking to

open an account to the extent reasonable and practicable;

‘‘(B) maintaining records of the information used to

verify a person’s identity, including name, address, and

other identifying information; and

‘‘(C) consulting lists of known or suspected terrorists

or terrorist organizations provided to the financial institu-

tion by any government agency to determine whether a

person seeking to open an account appears on any such list.


There is no way, nor really should there be, to get around these regulations given their focus on terrorism and the hangover that remains since 9/11. At the same time, it isn’t specific as to what they have to have - is this something the industry just figured out over time? That is, if the regulators have not determined exactly what the banks are required to have then is it up to them? If that is the case, then how do they figure out what to do other than by copying each other? If this is the case then which organization ultimately determined the required information? 


The other question is what it means to “maintain records.” I am sure that this is interpreted as records that they own and protect to make sure that they can tie things together. That gets troublesome when all the information ends up in the hands of the front office, who are moving around a lot (particularly in retail). Had this information not been available at WFC then the rogue employees could never have opened new accounts on behalf of the customers. This was true when I was at JPM - I could see the SS# of my clients even though ostensibly there was no need for this. I never used it to verify the identity of a caller.


The other reason that this data exists is because they want to use it to market new products to people using the data-driven models they’ve built over the last 20 years. This is both a business-line and cross-selling effort - credit card databases are potentially quite useful for identifying people with large spending habits, which you might think would correlate with wealth. The big banks certainly think so.


I guess it isn’t important to start by disrupting this marketing machine the banks have built. The fight seems larger than the prize. But the first step, which can be an easier way to communicate the information… that should be doable.


Types of Websites

 According to Wix, there are 27 types of websites: https://www.wix.com/blog/types-of-websites Types of websites eCommerce website Business ...