Scott Alexander is Wrong About Business and Economics

But that's ok, so is pretty much everyone

The brilliant Scott Alexander of Astral Codex Ten and previously Star Slate Codex writes here on basically whether Jeff Bezos being a billionaire is fair. Arnold Kling, economist and former web business man, writes that Scott is way over simplifying how easy it is to make a super successful company, pointing out in the process that economists don’t know a damned thing about running businesses.

Scott:

I think “Internet retail giant which dominates the market through economies of scale” is a natural niche which Jeff Bezos won the race to fill. When I say “natural niche”, I don’t want to discount Bezos’ accomplishment - I certainly didn’t notice that niche in 1994, and even if I had, I wouldn’t have had the business acumen to fill it effectively. I just mean that, probably sometime between 1994 and today, someone with business acumen would have noticed that niche and filled it successfully. Maybe not quite as successfully as Bezos. But successfully.

Arnold:

On economic issues, Scott comes across as informed as any economics Ph.D. In this case, however, that is not a compliment. I have long thought that an internship in business should be a requirement for getting a Ph.D in economics. Most economists are naive to the point of cluelessness about how business works in practice, and unfortunately Scott suffers from the same lack of insight.

I strongly disagree that all Bezos had to do was notice a niche. That is like saying that all Dwight Eisenhower had to do to make D-Day a success was pick Normandy as the landing spot.

When Bezos was building his business, I noted that in order to compete with Wal-Mart, he would need to match their logistical system. All they had to do to compete with him was build a web site. And yet he won! It took incredible skill to do what he did. I don’t have space here to list all of the management practices and company capabilities he needed to develop. I give zero credence to the suggestion that if Bezos had never existed, someone else would have built Amazon.

I generally fall more into agreement with Arnold here than Scott, although I would point out that there are many Amazon-like companies out there, both abroad and in the USA. Not nearly so successful as Amazon, but pretty similar in role. So, no, if Bezos doesn’t make Amazon in 1994 you don’t get the Amazon we know today, but there still would be something fairly similar. Scott is right about that, although whether “fairly similar” means “comes anywhere near close in terms of generating social surplus” is a big question. For comparison, K-Mart was a lot like Walmart, but Walmart was leaps and bounds better. A world where Walmart never existed and K-Mart is the next best thing is a rather worse world.

I want to point out two different flaws in Scott’s argument, and indeed the arguments of most people who argue over the “fairness” of entrepreneurs and inventors getting lots of money simply because they got there first¹. There are two main considerations.

1: Amazon had to be good in its niche, and continue to be good for a long time.

I still remember when Amazon was a little niche site for buying books. My father was a very early adopter of Amazon, partially because all the big book stores were over an hour drive away, and partially because if you have unusual interests finding new books is tough in general. Amazon solved both those problems.

The trouble for Amazon is that after demonstrating that those were problems worth solving (via becoming and remaining a viable business for a bit) other companies know those problems are worth solving and compete. Barnes and Noble could jump right in and compete, not to mention every other person who wanted to set up an online book store. Bezos and company had a little bit of a head start, having already solved some of the growing pains problems competitors have to deal with, but that is no guarantee of success. Amazon continuing to be the number one online retailer of books (and CDs, and movies) required being better than the competition in important ways day after day, year after year.

Recognizing the niche, therefore, isn’t enough, contrary to Scott’s example:

(as an intuition pump, consider an archaeologist who spends years narrowing down the location of a pirate treasure to a beach in Florida, then digs there and finds it. Whatever the finders-keepers law, we would feel like in some sense she had earned the treasure. But consider another world where God announces on 1/1/2023 “THERE IS A TREASURE ON THIS BEACH IN FLORIDA!”, and everyone rushes there, and whoever ran the fastest got it. Again, regardless of the actual finders-keepers law, we would feel like that person hadn’t really earned the treasure, just beaten the next person by a couple of seconds. Or maybe there are some weak qualifications - you have to be a strong person who owns a shovel - but there are many strong people with shovels and it’s mostly luck that one of them was closer than another. )

In reality, growing Amazon was less like showing up and digging up treasure one day, and more like finding and digging up lots of different treasures over the course of many years. It isn’t a lucky windfall based on one good idea or recognizing one niche, but rather the result of repetitive hard work that results in many good decisions and choices keeping you ahead of the competition. Scott’s intuition pump works in exactly the opposite direction of how he intended, because he believes founding Amazon is closer to just showing up on the beach as opposed to spending years of work.

That is just the early days of Amazon. The fact that Amazon extended into essentially all retail itself, a second hand/private third party retailer marketplace, a web streaming service, and most profitably a Amazon Web Services hosting many other websites, demonstrates that Amazon (Bezos more or less) effectively started a number of other successful businesses over time. Many more good decisions made, competencies developed, niches recognized and filled, competition out performed, etc. Going back to Scott’s pump, Bezos is a billionaire because he spent years finding treasure over and over again.

So, yes, if Bezos didn’t exist there would likely be a less good but kind of ok Gammazon, or a series of other companies that sort of fill the same niches. Some might even be better at their particular niche, and some might be much worse. The reason Amazon succeeded as well as it did was not, however, recognizing the niche once, or making a good/lucky decision once, but making a very long string of good decisions over and over. Claiming that to be just luck is vastly under valuing the work that goes into running a business.

One might as well claim that an old west gunfighter that wins 100 showdowns in a row isn’t actually a good gunfighter, but just got lucky enough to win the first duel. Sure, losing that first duel wouldn’t be good for your career, but you also had to win the next 99…

2: Scott entirely ignores the process of how people decide to exchange money for goods and services

As Scott writes:

The typical neoliberal defense of self-made billionaires goes: entrepreneurs and other businesspeople create a lot of value. EG an entrepreneur who invents/produces/markets a better car has helped people get where they’re going faster, more safely, with less pollution, etc. People value that some amount, represented by them being willing to spend money on the car. The entrepreneur should get to keep some of that value, both because it’s only fair, and because it incentivizes people to keep creating value in the future.

…

The problem with the neoliberal argument is that it gives the first person to fill a niche credit for the niche’s entire existence, not just for filling it earlier than it otherwise would have been filled. Just because Jeff Bezos solved Internet retail two years earlier than the person who would have done it if he was never born, he gets to collect rent on all transactions forever, while that other guy gets nothing.

I don’t know who exactly put forth that argument, but note that it assumes it is right and proper for a third party to redistribute wealth differently than what the principle parties agreed was acceptable. I.e. it assumes governments owns everything and gets to choose how much everyone should have, as the government sees fit. Each of us then must justify why they should be entitled to their current possessions.

But why should anyone have to justify how much money they have if all their money came from voluntary transactions?

I expect that this notion comes from the regrettable human tendency to see the state of the world as a snapshot of time, disconnected from the past and future. “Why should Bezos have lots of money when I don’t?” only needs justification if we don’t know how he got the money in the first place. Asking someone to prove they deserve what they have is an intuition rooted in “Prove you didn’t steal that.” In a world where the upper bounds of what one person or group can produce are very low that makes sense. Not so much today, when even small groups and individuals can serve billions of customers and make a profit of a dollar or so each.

If we assume instead that individuals own property and may dispose of it as they see fit, so long as no one is stealing or defrauding, we shouldn’t find any particular distribution outcome demands justification. Using coercion to change the distribution requires justification in this case.

In other words, if someone wants to justify taking money from Bezos and redistributing it for greater justice overall, one would have to demonstrate where he stole from someone, where a transaction was not voluntarily agreed to by the other party.

Do you think Bezos needs to redistribute to Amazon employees? Show where employees did not agree to do a job for a given amount of money.

Do you think Bezos owes the government more money? Show where he did not pay taxes properly in accordance to the tax code at the time.

Do you think Bezos needs to redistribute to customers? Show where customers did not voluntarily choose to pay for a good or service².

That’s it. If every step of the process was done with the agreement of both parties, it is nonsense to claim that the process as a whole was unjust. You could argue that the agreement was come by via fraudulent representation, but you would have to show that³.

If you can’t show either involuntary coercion or fraud, you don’t get to comment on the validity of the outcome as something to be corrected by coercive action. You could argue whether it would be nice of Bezos to spread some wealth around differently, but you can't argue that Bezos having the wealth is somehow unjust or improper⁴ and it needs to be taken from him.

Scott’s argument assumes that no one has the inherent right to anything, no matter how they got it, and all property is possessed only at the pleasure of the mob (government) who can decide to rearrange it as it sees fit, whenever it sees fit.

There’s a word for that sort of behavior. It isn’t a nice one that governments like to have applied to them.

---

Now, none of that says that Amazon is good! I am pretty well certain that Amazon and Bezos have some skeletons in the closet. If nothing else, it is almost impossible not to commit a few felonies just by accident as an individual, must less as a giant corporation. Some probably weren’t by accident.

Be that as it may, unless one demonstrates that Bezos’ income was ill gotten, arguing over whether or not he deserves it is entirely wrong headed. He deserves it because at every step people agreed to the exchange, customers and employees. Every individual decided “Amazon deserves my money for this good,” or “Amazon deserves my labor for this money,” instead of some other competitor. Every individual decided that themselves.

Humanity has a lot of experience with systems where desert was decided by groups instead of individuals, and those systems did not work out well at all.

1. I won’t even bring up the fact that “fairness” is entirely meaningless in this context. The fact that arguing fairness as though it is some objective thing marks you as a sniveling child still crying to mommy to fix things the way you would like is the subject of another essay.

2. This might be the easiest target, by the way.

3. Possibly you could make a euvoluntary exchange claim, that trades were voluntary only in the same way that being charged 10 million for a bottle of water while dying in the desert is voluntary, but that strikes me as a very hard claim to make regarding books and other retail items. Amazon is quite far from a monopoly position in everything other than perhaps AWS, and even then that’s a tough argument.

See (or hear) Mike Munger on euvoluntary exchange.

4. In other words, if every step accords with commutative justice, you can’t use coercion to correct it. Just saying Bezos should use his money differently falls under the loose, vague and indeterminant realm of distributive justice, and it is unjust to use coercion to correct perceived deficiencies there. I really need to get around to writing that explanatory essay…