(2024-01-10) Cowen Patrick Mckenzie On Navigating Complex Systems

Tyler Cowen: Patrick McKenzie on Navigating Complex Systems. Tyler sat down with Patrick to discuss signature fields on the back of credit cards, whether bank tellers or waitstaff are more trustworthy, the gremlins behind spurious credit card declines, how debt collection and maple syrup heists should change your model of the world, Twitter’s continued success as the message bus for government and civil society, crypto vs traditional money transfers, the intended desolation of bank parking lots, why he moved to Japan and how it affected his ambition, why Tether hasn’t collapsed, the internet as a Great Work, how he’s experiencing reverse culture shock after returning to the US, what he’ll learn about next, and more.

Here is Patrick’s own description of himself, and I quote: “The broad through line of my work is systems thinking applied to businesses. I think the social organization of the internet and its impact on the world are underestimated by almost everyone, including Silicon Valley. Everything important to you in the world sits atop several infrastructure layers. I am a storyteller for some of those layers and the people who build them.”

On money security and banking

COWEN: Why is it you still have to sign the back of your credit card?
MCKENZIE: Like many answers, the true thing is that that is a bit of a misconception. You actually don’t have to sign the back of your credit card in the legal realism sense, in that you can fail to sign the back of your credit card and go throughout your life untroubled by that fact. Most people who do that will be untroubled.

The original reason for signing the back of the credit card was about solemnization, that you are making a solemn, legally binding promise to the bank when you receive your card that you are good for charges made using that payment instrument.

The vast majority of clerks won’t even check signatures these days, which people are sometimes annoyed by.

COWEN: And you’re giving them your signature, right, by signing the back of the card? If someone stole it, they now have your card and your signature.
MCKENZIE: Yes. A lot of what we would call pre-modern thinking about security in terms of payments was not — I won’t say not all that well thought about. There are choices that were made back then that were good in the era they were made, that are disastrously bad right now.

For example, if I had, let’s say, a bank account, and I have a secret controlling that bank account, and that secret would allow me to take all of the money out of that bank account, you might think, “How should I protect the secret?” The answer you would come up with in 2023 is not “I want to print it in machine-readable type on every check that I give to everyone.”

In fact, that is what we did back in the day. Many, many tens of billions of dollars of check fraud happened over the years because of that issue, and we just took the punches and developed a vast shadow infrastructure to defang this fundamental, irreducible security hole in the heart of the checking system.

COWEN: Whom do you trust more, your bank teller or the waitress to whom you hand your credit card?
MCKENZIE: A bank teller by a country mile, although it is ultimately irrelevant.

But it is ultimately irrelevant because in the case you’re taken advantage of by either a bank teller or a waitress, we have an extremely effective public-private regulatory state machine which will make you whole for that. It will be annoying, but you won’t actually lose money.

COWEN: If I try to use my credit card and it’s a false decline, that is, they won’t take it when they should, what’s the most likely reason for that?
MCKENZIE: Gremlins. It sounds like a joke, but serious people . . . I used to work in the payments industry, Stripe.com. I’m still an adviser there. Everything I say is in my own opinion. Stripe, as a corporation, does not actually endorse the existence of fiscal gremlins. I’m not sure I don’t. Many people in the payments industry have spent a lot of time looking at the true source of false declines. Eventually, the best you can do is reduce it to some error term where you don’t know specifically what system is at fault.

The credit card ecosystem, while it might present to you as a single pane of glass that you interact with just like you interact with a single pane of glass on your iPhone, is actually a stitched-together patchwork collections of systems that have existed for the last several decades. Somewhere in that stitched-together patchwork, someone had a hiccup. Then your credit card gets what we call a spurious decline.

The rate of spurious declines is something of a trade secret, but it is much higher than any rational human being would expect is a tolerable rate of spurious false activity within a system. That that has persisted for so many decades is an interesting collective-action problem because it’s a stitched-together patchwork of systems.

There is limited visibility into what is actually causing the problem. Credit cards are an enormously lucrative business, and for the longest time, all players were like, “Yes gremlins. I don’t know what to tell you.”

A thing that is not exactly a trade secret is that it is not the fact that problems are not solvable in the world. You actually can move the false decline rate by having the right people in the right places actually work on the false decline rate.

I had this long-running assumption that we’re doing about as good as we can with rockets. The laws of physics are a thing. We’ve dumped billions of dollars into this problem, and the smartest literal rocket scientists in the world at it. Then Elon Musk came around and said, “Actually, I think we can cheat the costs of getting things up to orbit by, oh, three orders of magnitude.”

On debt collection, maple syrup heists, and other evils

COWEN: In a country of near full employment, why would anyone work in the debt collection sector? It’s ugly work.

MCKENZIE: The wage is certainly not so high.

A typical debt collector is a high school graduate. They make plus or minus $15 an hour. Perhaps that has shifted a little bit as inflation has hit hard the last couple years, but broadly speaking, it’s that. You have a pay scale that’s not too different from McDonald’s for a much, much worse working environment, et cetera, et cetera. One reason — and I think this is over-advanced in the world but is partially explanatory in this case — is that the existence of — I’m about to use the word “evil,” and I don’t want to use the word “evil” to —

There are some people who enjoy inflicting suffering on others. If you enjoy inflicting suffering, debt collection is a wonderful gig for you. Some people who enjoy inflicting suffering go into the police, for example. It’s not the fact that all members of the police enjoy inflicting suffering, but that is a major source of utility for at least some people who get that job.

Another thing for debt collection is that in the places where debt collectors actually live, the economy as a whole can be near full employment, but in rural New York, that might not be great.

And indeed, the turnover rate for debt collectors is astounding. It’s on the order of — I hate making up numbers. This sounds like I know what I’m doing. It is higher than it is for fast food restaurants, which is already close to 100 percent a year.

COWEN: How did the maple syrup heists change your overall view of the world and of humanity? What’s the model update?

I often assume that large, well-regarded, professionalized industries have a rate of crime associated with them which rounds to zero

I would assume agriculture is relatively heavily regulated.

But it turns out that the amount of supply-chain fraud in maple syrup is actually quite higher than zero.

So, it is possible to steal millions of dollars of maple syrup and sell it on the black market, which blew my mind.

COWEN: Do you ever encounter sectors where there’s less fraud than what you had been expecting? Or is there just more evil all around, and you were naive and now you’re wiser?

one place where we detect much less fraud than I would expect us to detect is in insider trading and other forms of, obviously, illegal market manipulation...

am I getting a biased look at the world? Or are we only detecting the dumbest possible criminals?

For a while and continuing, I thought that crypto was drawing a lot of the fraud out of the more traditional sectors of the economy

But I don’t think that is fully explanatory.

COWEN: As we’re speaking in October of 2023, Elon Musk has suggested that he will take X — formerly Twitter — and turn it into some kind of universal payments app. What’s the chance he can succeed with this? If you’re bearish on it, what’s the fundamental problem? ((2023-07-29) The Everything App Why Elon Musk Wants X To Be A Wechat For The West)

I understand, from a product management perspective, how you could believe that that would be fantastically valuable if you got it to work, because the super apps (everything app) that exist in large portions of Asia are fantastically valuable businesses

The fundamental reason why I am extremely bearish on it happening for Twitter — and I would put the percentage at below 1 percent, far below 1 percent — is that this is very, very hard. It feels unlikely to me that the organization that is Twitter is resourced or competent enough to make this happen.

I will say, I’m an enormous fan of Twitter. I think it is actually descriptively one of the most important products that exists on the internet because it has not solved, but ameliorated a lot of cross-organization coordination problems that happen in very important places

It is broadly under-appreciated how Twitter is the message bus, the sub rosa coordination mechanism for the United States federal government, for every counterparty that the United States federal government or any agency or individual faces, for the media, for everything.

COWEN: If I go to Western Union and send money to Mexico, why are the fees and transfer costs so high? It doesn’t seem like it should be that hard.

MCKENZIE: This is an enormously frustrating thing for me because the crypto industry has been saying, “Why are Western Union fees so high?” Yet, very few people in the crypto industry seem to do the college undergraduate–level thing that one would do to answer that question, which is, why don’t you read their annual reports and see what they spend all the money on?

What they spend all the money on is retail points of presence by the customer in both countries, so 50 percent of all revenue taken in by Western Union goes to have a physical point of presence (PoP) by you sending it and by — without loss of generality — abuela in Mexico.

Another reason why the fees are so high is regulation. This one is, we as a society could choose to do things better and have not chosen these. I wish we would. Due to anti–money laundering and know-your-customer (KYC) requirements, the amount of compliance drag on small transfers of value in the world is very, very large.

if you can do small transfers of value easily without proving your identity, then you could do large transfers of value by breaking it up into small transfers of value that are unsurveilled, and then that would enable crime and terrorism and tax evasion.

I’m a longtime crypto skeptic, and one of the things that I think I see eye to eye on with the crypto folks is that AML and KYC impose enormous costs on society.

we should probably work on changing the laws and regulations and/or changing how we implement them in products that are actually valuable.

COWEN: Why are bank parking lots usually so empty?

MCKENZIE: Oh, this is a straightforward result from queuing theory,

But the thing that the bank really wants to optimize for is new account opening. (Acquisition)

it turns out that gratuitously over-provisioning the parking space almost all of the time maximizes for not losing new account opening at a few very limited windows per year. Those very limited windows can make or break the branch’s ability to contribute to their larger financial institution.

The number of accounts a bank actually opens per year in terms of checking accounts per branch — that is a number that one can have access to in various places. Depending on the bank and the locality, et cetera, it’s between 200 and 500. So, if you turn away a single customer or two customers in a row to open accounts because you had three parking spaces where you could have had seven, you’ve done a very outsized amount of damage to your financial institution

COWEN: I should infer that I shouldn’t keep much money in my checking account

To ask you the Willie Sutton question, why do people try to rob banks?

MCKENZIE: It largely is a low-return activity.

The median haul that a bank robber gets is about $8,000. It is the worst possible way to earn $8,000 because a career bank thief — the other word for that is long-term guest of the United States government. It is impossible to get away with over an extended period of time.

credit card fraud is so amazingly more lucrative than robbing banks.

you do see casual fraud by unsophisticated, let’s say, non-professional criminals, high school students

therefore, the people that actually do stick up banks are overwhelmingly drug addicted. They’re disproportionately struggling with a mental illness, et cetera, et cetera. It is not that big a risk.

That, by the way, is part of a few reasons why banks are not optimized to defeat people doing bank robbery, which is surprising to many nonspecialists. If you join a bank, one of the first things that they say is, “In the event of a bank robbery, don’t be a hero. This does not matter to us at all.

MCKENZIE: I moved to Japan almost right after university. I suppose I would’ve been 20. The original reason — and I wish I had a copy of this article. I should post it on my wall somewhere. I had gone to Washington University in St. Louis to study computer science. I was worried because the Wall Street Journal said the dot-com bust means that engineering employment will cease in the United States of America. All future engineers will be hired in China or India at a fraction of US prevailing wages. (offshoring)

Clearly, the successful engineers in the future will be playing a Venn diagram game where they do one hard thing that is engineering and one hard thing that is speaking a foreign language.

I thought speaking foreign languages is hard because it’s the only class in high school that I did not get an A in for showing up — Spanish

So, I made a spreadsheet

Sort by Column H descending, and Japan was at the top of the list.

I thought, “Okay, I’m going to learn Japanese, and then I’m going to go to Japan after university for just a few years to get better at business Japanese. Then I’m going to come back to America, get a job at Microsoft in a Japan-facing role

It turns out that there was a great deal of rigor there, chasing a conclusion that had absolutely no basis in fact. It is not the case that the tech industry stopped hiring Americans in the last 20 years, but it also led to what was ultimately one of the most important decisions in my life. (decision-making)

Also, I try to tell sophisticated people that other people who they have a high degree of regard for — some sophisticated people, for whatever reason, have a high degree of regard for me — even people you have a high degree of regard for sometimes make very large decisions based on very small amounts of data or very small amounts of consideration with regards to that decision, and therefore you should update your world model about that.

COWEN: You’re doing a startup. How is that evaluated in the Japanese marriage market? (dating market)

MCKENZIE: Extremely poorly historically

Both in the United States and in Japan, one of the most influential works of fiction ever is The Social Network. I bang the drum on this a lot, that The Social Network raised the status of entrepreneurship for proto-entrepreneurs worldwide

Unfortunately, that gets expunged out of the public narrative because you don’t want to tell serious people, “Oh, yes, I totally started doing a startup versus going to Google because The Social Network gave me social permission to do so.” But that is a real thing.

I said, “I’m an entrepreneur.” Preschooler, mind you, said, “Omotta tōri, kare wa mushokuda,” which means, “Just as I thought, he is unemployed.” That is the form of a too-perfect anecdote to encapsulate a broader societal zeitgeist with respect to entrepreneurship.

COWEN: You once cited the following idea on your blog, and I quote, “Most people want to become wealthy so they can consume social status. Japanese employers believe this is inefficient and simply award social status directly.” That’s not you writing, but it’s you citing. Can you explain?

Now, the thing that you are trying to do if you are a salaryman is not to get the promotion so that you earn more money, because you will not earn appreciably more money. Indeed, you will not earn appreciably more money than the false course of just taking social promotions through your entire life. The thing you are trying to do is get a higher . . . both true and sounds too good to be true. (status)

Tanaka San had a couple of weeks of seniority versus me in my role at the company. In the default ordering of the social hierarchy at a Japanese company, he would be above me on this chart that is pasted on the wall. It’s pasted on the wall for an actual business purpose because, if someone’s out of the office that day, you write they are absent or they’re at a customer site or et cetera, et cetera. But the order matters, and everyone knows the order matters.

Tiny little anecdote that shows much more about the shape of the world. A huge amount of the effort involved in the culture that is Japanese salaryman is convincing the people that matter to you... that you have made it if you are ambitious, et cetera

COWEN: How is it that your ambitions grow steadily in this environment? You weren’t beaten down,

MCKENZIE: Interestingly, from having the perspective of being around people who are incredibly ambitious the last couple of years, and whose ambitions had positive social purpose, one of the great regrets of my life is that, in my twenties, I had so little ambition and was so beaten down, and didn’t take more aggressive actions to exit out of the situation I found myself in.

I remember, after I quit the salaryman job

I worked at jobs that are described by that for about six years, burned out terribly, made a socially appropriate excuse to quit, which is the one thing that you, as a salaryman, are not allowed to do.

I remember April 1, 2010, thinking I could get on a plane right now with nothing but the clothes on my back, close the Japanese chapter of my life entirely

But I’m not willing to call it quits on this Japan adventure yet. I do enjoy living here. Working here is terrible. I will try to carve out some semblance of living from my tiny internet software publishing operation that I was running at the time and see what happens. Then indeed, the thing that happened three months later is that I met the young lady who eventually became my wife, and so we were off to the races.

COWEN: Wasn’t it efficient, in some regard, that you had so many years out there in the wilderness, so to speak?

there is a level of energy and health that I had access to in my 20s that I no longer have access to. And so, I feel some counterfactual regret as to things that I actually spent that energy and health on, which was running the World of Warcraft III guild and being a Japanese salaryman for many years.

I do not think that six years of Japanese salaryman optimized for me learning all things. I think I could have extracted most of the signal from that experience in six weeks to six months and then left, but felt myself socially constrained from doing so.

I will say one thing about the experience, is that having a lot of free time on your hands in the evenings and no outlet for English was a great push in the right direction of writing more than most people do in their 20s. The compounding effects of writing more are responsible for both a lot of the good stuff that has happened to me and a lot of the good stuff that I’ve been able to accomplish in the world.

COWEN: Crypto questions. Why hasn’t Tether collapsed yet?

MCKENZIE: I pre-commit that if it is not a fraud, I should have enormous egg on my face.

From one sense, I can claim victory. It’s a fraud. I was saying it was a fraud before it was cool. The government now acknowledges that they did funny things at some point in the past. But I think it should also collapse. It is not just a tinkering-around-the-margins fraud. There is a huge amount of actual criminal activity there.

One sort of naive answer is, do we expect frauds to collapse quickly? Probably, no. Bernie Madoff was active for 17 or 18 years, and the only thing that actually brought his fraud to light — well, various cognoscenti knew about it — the thing that made it undeniable was being unable to meet redemption requests in the wake of the global financial crisis.

The thing that we would expect the collapse to happen with — my dominant narrative — is some form of government action against their reserves that were custodied in some financial institution in New York, followed by the public disclosure of that action

COWEN: Circa late October 2023, Bitcoin is priced as high as $33,000. Why is that?

It’s been crushed — what — five, six times since its origin, and it keeps on coming back. Is it a proven asset class?

I was saying that Bitcoin was overpriced at $17. I still believe that to be true in a lot of senses, meaning if counterfactually it was $17 today, I would say it was overpriced at $17.

But some sense of epistemic humility — I have been watching that ticker for over a decade now, and the ticker and market participants keep voting against my view, and question why — is it because of manipulation? Surely, yes, but that is not a full explanation.

Bitcoin as itself is clearly not the best product that crypto has come up with. Yet the market says it is, by a comfortable margin, the most valuable product that crypto has come up with, and I don’t know how to square that circle.

If the human race does not exist in a hundred years, it is almost definitionally because of some knock-on consequence of the invention of the internet. If the human race ceases to exist, and the internet was necessary in the causal path to get to there, then I would regret the internet. In every other case, I would not regret the internet.

I have an ebullience and love for it in a way that people in our social class in the United States are aggressively socialized out of having ebullience and true love for anything. I think that the internet is the capital G, capital W, Great Work of the human race in a lot of respects, that it is magical.

I think, in aggregate, the internet is obviously more important than penicillin by many orders of magnitude. ((2024-09-30) ZviM AI And The Technological Richter Scale)

COWEN: When you ran VaccinateCA, which helped people in California get access to COVID-19 vaccines, what is it you learned about talent?

There’s a thing that I would have put a high probability on prior to VaccinateCA, that VaccinateCA increased my credence on, which is that relative to credentialed talent in institutions, like — throw a dart at the dartboard about it. Let’s say county health departments... Jay Rando, person on the internet: “Can we be better at important facets of pandemic preparedness given, hmm, two to six weeks versus a county health department that is prepared for exactly this problem for their entire professional career?” Yes. Straightforwardly, yes. [laughs]

It is extensively underrated how effective nonspecialists can quickly come to be, how spiky talent is distributed, and how the internet can act as a force for bringing together the right spikes for that in a very fast fashion and creating leverage good with it.

COWEN: If you had to explain or express, in its most basic, fundamental terms, why people are afraid to negotiate more and negotiate harder, what’s that explanation?

MCKENZIE: There is a societal script against being greedy

I wrote one blog post about salary negotiation back in 2012

Hundreds of thousands of people read that every year. Many of them choose to write me an email

What do I come to believe about this? One, I clearly under-optimized my selection of picking things to write about because that was the shining star in my portfolio for so long, yet I never attempted to take the knowledge that it was the shining star and do more stuff like that in the future.

Two, society was hideously under-optimized in that it took one guy of no special skill in this field, writing for one day on the topic, to unlock tens of millions of dollars of value.

Then, three, this implies that there are a lot of sophisticated actors who are making huge, huge mistakes that could probably be ameliorated for about as much effort as that blog post took to write, but who have not taken that effort.

COWEN: You’ve left Japan, you’ve returned to the United States, you chose Chicago. Why Chicago?

MCKENZIE: Straightforward reasons on this one. My parents live in Chicago, and my family grew up here

I expect in the future to, at some point, be an American living in Japan.

COWEN: What has most surprised you about returning to Chicago?

MCKENZIE: I am having more difficulty dealing with culture shock than my children who have never known a nation other than Japan, or my wife who has not materially spent time outside Japan. The shock is coming in little ways and big ways at the same time.

One of those things is, portion sizes are so much larger than they were when I was growing up

There are a lot of little things that Japan does right. I call it the will to have nice things. In the United States, many places lack the will to have nice things and suffer the consequences of that. These are just a million little paper-cut annoyances, like the delivery company deciding to deliver my bed to the middle of the front yard because they couldn’t get into the house. (competence? care?)

COWEN: Last question. What is it you plan to learn about next?

MCKENZIE: I have an answer to this question. It is more of the engineering reality of geothermal heat and electricity extraction between 6 and 10 kilometers below the surface of the earth, because I’m going to be doing a tiny bit of pro bono work on behalf of a nonprofit organization that is attempting to popularize that technology as part of the energy mix over the next couple of years.