Networks of Trust vs Markets

post by Henrik Karlsson (henrik-karlsson) · 2021-05-31T19:59:19.842Z · LW · GW · 22 comments

Contents

  Friends are cheap
  Safeguards and asymmetries
  A Coasian theory of networks
None
23 comments

Markets are trust reducers. When functional, they make sure that you mainly need to trust one thing: that others will be self-interested. And since, as Adam Smith pointed out, you can trust even butchers and bakers to be self-interested – markets greatly increase the number of people you can trade with.

"Markets!" my econ professor would proclaim in a thick Armenian accent. "If only they were a machine! People would love a machine where you can put in wheat and pick out a helicopter."

That is a marvelous machine.

But one should not forget what it is built for: enabling transactions with strangers. Being able to transact with strangers doesn't mean transacting with strangers is in your self-interest. Not always. Not, I have noticed, as often as I had thought.

Often relying on non-market transactions is more efficient than markets. What is a non-market transaction? It is a set of different things: neighbors providing each other mutual aid, or friends trying to outcompete each other in generosity. Gift economies. Iroquois longhouses. Transactions that take place within your family, your circle of friends, and the people they can refer you to – your network of trust.

Last year, two incidents made me sit down and calculate just how much markets costs me compared to my network of trust. There are a bunch of things that markets can do, but my network can't: assemble the raw material to a computer, ship me a banana. But in areas where I could rely on trust, it turned out to be surprisingly cost-effective.

Friends are cheap

A few weeks after covid-19 closed the borders, I had to ship my furniture from Sweden to a Danish island in the Baltic Sea. At first, I tried markets.

It wasn't easy.

The only shipping company that quoted reasonable-sounding prices turned out to have a business model based on making people's belongings disappear. Alas, I had already signed the contract when I figured this out – a contract that on closer inspection turned out to have a clause that stipulated that if I wanted to break of the deal, I would have to pay the equivalent of a month on a median Scandinavian salary.

This is a very gripping story, but to cut to the chase: after having convinced the criminal truckers that it was in their best interest to drop the case, I called my friend Torbjörn –

"Could you get a vacation next week?"

"Of course." (This is Sweden.)

– and that saved me about 20 days' salary.

By which I mean, the amount of money I would have paid a shipping company would have cost me 20 days of labor more than what I had to pay for renting a truck. And that is assuming I would have employed a shipping company with a criminal record, mind you. If I would have employed a credible company, it would have cost me ~50 days of work.

Now instead I vaguely owe Torbjörn something on the order of 2 days.

= My network of trust was 10 to 25 times cheaper than markets.

I didn't think much about it at a time. Asking friends to help you move is common enough.

But then my wife noticed something: the roof of our new house was leaking.

Contacting a few carpenters, selecting those that had good reviews online, we got some price figures. The carpenters could solve the problem at a cost of ~2 years of median salaried work. Was that reasonable? The material would cost something like 3 months' salary. Retiling a roof is a fairly fast operation. Two weeks? Three? No, my back-of-the-envelope calculation tells me that 2 years salary is not a reasonable price.

But what to do?

My wife emailed a hippie. Rumor had it that the hippie and his crew sometimes did construction work – but only if they liked your vibe. I put on my knitted sweater.

And this might not align with your notion of hippies – but after meticulous studies, he unveiled a construction design with improvements of the 10X variety that start-ups would kill for.

Organic materials. Reduced heating costs. Four times longer life expectancy! Half the price.

The foreman of the hippie construction company explained that his anti-market business model (giving ridiculously good prices, but only to people he knew or predicted would reciprocate the kindness) allowed him and his employees to support their families on 15 hour work weeks.

That immediately raised two questions.

1. How can I find more hippies?

2. Why are markets so expensive?

Let's look at the second one. (I look at the first one here [LW · GW].)

Safeguards and asymmetries

I won't pretend to know all of the reasons why market transactions are more expensive than transactions between friends. But two obvious ones are costs for safeguards and information asymmetry.

By reducing the amount of trust needed to enter into a transaction, markets have connected people that otherwise probably would have considered piercing each other on spears. Like me and criminal truckers.

This is a feat. But it is not a free lunch. Making sure that we respect people we don't trust is expensive.

In Debt, the late anthropologist David Graeber notes that the creation of markets, as a realm separate from our network of trust, is made possible by the existence of several resource-intensive institutions:

Economics assumes a division between different spheres of human behavior that, among people like the Gunwinngu and the Nambikwara, simply does not exist. These divisions in turn are made possible by very specific institutional arrangements: the existence of lawyers, prisons, and police, to ensure that even people who don't like each other very much, who have no interest in developing any kind of ongoing relationship, but are simply interested in getting their hands on as much of the others' possessions as possible, will nonetheless refrain from the most obvious expedient (theft).

In a non-market society, services and gifts were exchanged in complex networks of relationships. Extracting the market out of those relationships takes work; work is not free.

When laboring as a software consultant, it felt like 50 percent of my time was wasted in meetings with clients. They drew up legal documents. We had endless follow-ups where they tried to make sure I was doing what I had promised. And on top of that, we had insurance to cover for mistakes and potential disagreements. And both parties paid heavy taxes to support a justice system that could keep the other party in line.

Those safeguards were expensive. And I don't need them when I code with friends.

The other thing that makes market transactions expensive is information asymmetry.

Asymmetries of information are fundamentally a good thing. If you know something that I don't, we can pool our knowledge and achieve things that are too complex for a single mind. When dealing with parties we trust, asymmetries of knowledge create complementarity: you know how to do something that I can't, that's why we transact.

With friends, this is usually the end of the story. But when dealing with strangers, asymmetries of information is also a cost center. By strategically withholding information others can mislead and exploit you. That pushes prices up.

When we were redoing our roofs, the hippies were friendly – they revealed what they knew. We had a better understanding of the situation after talking to them.

The professional carpenters, on the other hand – though they looked as friendly! – were not an inch above proposing designs that would break, and quoting prices uncorrelated with their costs. Because why wouldn't they? Really.

Someone who has done an intro class to economics could probably argue that the markets would correct for this. Isn't crooked carpenters a great opportunity for honest ones to increase their market share?

No. Not if the consumer never realizes what happens, which they usually don't – because of asymmetries of information. Usually, the customer only sees the wholesome, friendly-looking attitude – and gives a rave review after being exploited.

So market transactions are salted with costs for safeguards and salted because we get manipulated. Therefore it is often better to stay away from the market.

A Coasian theory of networks

For some reason, companies seem to be much more sensitive to this fact than individuals, at least if one is to believe Coase's theory of the firm.

According to Coase, firms expand until the cost of producing in-house matches the cost of arranging the legal work, etc necessary to outsource it.

(A firm can in this context be seen as a network of trust. It is a set of relationships that are defined by non-market transactions – colleagues making requests to each other, counting on goods and services being provided without payment, for the greater good of their firm or something.)

But for some reason customers don't do this. They don't try to figure out what is better served by non-market transactions within their network, and what transactions it makes more sense to do in the market. They just mostly outsource everything. (And the few things they do produce in-house – say food – might not even be the things where they have a comparative advantage.)

So what is the optimal balance between markets and mutual aid? It depends on who your friends are. Warren Buffet famously never trades with people that require lawyers to draw up contracts.

There are a lot of areas where you can save time and money by helping a hippie commune harvest in exchange for a new roof, or giving a friend investment advice in return for shipping your furniture. I don't know how far you can scale that until you hit diminishing returns. But much further than most people do.

Cross-posted from my blog.

22 comments

Comments sorted by top scores.

comment by Selueen (selueen) · 2021-06-01T07:23:12.395Z · LW(p) · GW(p)

I see a few problems with trust networks that are not generally present in the markets.
I'm glad that your experience was mostly positive, but I'm aware of many examples where things are more tricky.
Part of it comes from two very different but common attitudes towards transactions between friends/family. Some people think that every work should be paid, always. Others expect and provide free help. 
These positions are clearly non-compatible and predictably lead to conflicts, especially when people don't communicate their position clearly. They often think that their position is obviously right and don't even consider the alternative until the conflict arises. 
Another problem is that in transactions with friends/relatives there's often a pressure to work informally. Which is a risk - if they fuck it up you can't even sue them. And it's not that unlikely that they do - you probably did not select them based on their responsibility and their expertise in whatever field they work in. So you might lose all the resources AND hurt your relationship on top of it. 

It's not that these problems are completely unavoidable, but people do get burned.
Some personal examples, so that I don't just parrot stories I've read on the internet.


Extremely bad example.
One of my relatives was technical director and de-facto co-owner of one local ISP. De jure he was nobody, he never got around to do all the paperwork - partly because he trusted his "friends", partly because there were some complicated issues, partly because he is rather lazy. Years and years of no consequences, until they decided to sell the company. Guess who's opinions was no considered and who got nothing out of the deal.  I know, it's extreme and it's not only trust related, but these things do happen. 


Somewhat good example.
Back when I was working in e-shop, our courier got sick. We could not realistically hire replacement in time, and outsourcing would be extremely expensive. My boss asked our sysadmin to help the company out. I remember him discussing that decision with someone - "I know he would not decline, he is a nice guy, but he is too shy to name the fair price, and he would be disappointed if he does not get paid fairly". He ended up paying him slightly more per hour than he paid out actual courier. 

To conclude, my main points are: 

  • Don't rely purely on trust if stakes are big 
  • Critically evaluate your reasons to trust the person
  • Clearly communicate your expectations and listen to their expectations
  • Consider all the risks, not just the material ones 
Replies from: henrik-karlsson
comment by Henrik Karlsson (henrik-karlsson) · 2021-06-01T11:06:11.550Z · LW(p) · GW(p)

This is a great point that I had overlooked. I have similar bad experiences now that I think about it, mainly from starting a company with a friend, whom I should have questioned more on technical skills, but didn't because of our high trust. That was a nightmare.

I need to think a bit more deeply about why relying more on trust has worked so well this time, because as you say, it is not a given. Your bullets are a good starting point for a more nuanced understanding of the comparative costs of markets and trust networks. Thanks!

comment by Dagon · 2021-05-31T23:42:53.674Z · LW(p) · GW(p)

Fully agreed with the underlying realization: trust (enforced by norms, shared culture, and friendship) is FAR MORE efficient than markets.  If you have social capital and trust relationships in areas that match your needs, definitely use them.  Also, continue to invest in such things - they're more valuable than money (at many common margins, at least; exceptions abound).

I disagree that markets reduce trust.  Markets operate with less personal trust, and can achieve many things that a network of trusted contacts just can't.  And it's not evenly distributed, either.  Some people wouldn't even have needed to rent a truck - they have friends who'll lend it.  Some people don't have friends able/willing to drop everything and perform labor for them.  In your case, you used the market to get a truck, and a trust network to get the labor.  Great, but having the market as fallback for those parts you couldn't otherwise do made it possible.

In truth, the efficient point will vary by task and individual (and their network) - almost nothing interesting is purely market or purely non-market, but some mix of the two.  Which is the main point of theory of the firm - the blending of market transactions for many purposes and many timeframes with non-market (mix of contractual and social norms) activities for others.

Replies from: henrik-karlsson
comment by Henrik Karlsson (henrik-karlsson) · 2021-06-01T06:00:44.309Z · LW(p) · GW(p)

Good points! I agree with everything, and the way you puts it is clearer.

You're definietly right that markets do not reduce trust. That was a bad turn of phrase: what I tried to say, and what I think you are saying, is that markets enables transactions with lower levels of personal trust.

And it is very true that markets are great enablers when they allow you to do things not possiblie within your network. Expanding my network to include someone with a truck might (or might not) be an efficient investment for me, for example. I don't think it would be rational to exit the market. But it is important to remember that there is another direction one can look in, and that sometimes investing in growing your network is more efficient than raising your salary.

comment by Pattern · 2021-06-02T02:08:05.952Z · LW(p) · GW(p)
That immidiatly raised two questions.
1. How can I find more hippies?
2. Why are markets so expensive?
Lets look at the second one.

Based on the name of this piece, I'm not surprised you went there, but the first question sounds like it might change your life.

Replies from: henrik-karlsson, henrik-karlsson
comment by Henrik Karlsson (henrik-karlsson) · 2021-06-22T06:26:20.172Z · LW(p) · GW(p)

So here's my attempt at the find-more-hippies-question: https://www.lesswrong.com/posts/dktESGGtYJwwdTjeo/scaling-networks-of-trust-1

comment by Henrik Karlsson (henrik-karlsson) · 2021-06-02T07:36:34.761Z · LW(p) · GW(p)

To be clear, by hippie I mean any highly trustworthy person that is willing to do non-market transactions with an extended network, or something like that. But yeah, actively looking for more people like that is having some interesting effects on my life.

This post could be read as an introduction to a (hypothetical) sequence about using and scaling networks of trust. If there is interest, I might write another post detailing my observations so far. Any thoughts?

Thinking in public like this has been great - several of the comments have helped me clearify my thinking and pinpoint new opportunities and potential pitfalls.

Replies from: Pattern
comment by Pattern · 2021-06-02T16:50:29.238Z · LW(p) · GW(p)
This post could be read as an introduction to a (hypothetical) sequence about using and scaling networks of trust. If there is interest, I might write another post detailing my observations so far. Any thoughts?

I'd be interested in that.

comment by Arvid Kästel (arvid-kaestel) · 2021-06-01T18:56:29.851Z · LW(p) · GW(p)

I think an economy of formalized trust or reputation is a very interesting concept. Instead of money we would interact with reputation scores or something similar. So your capital is basically your reputation or more precise your historical input to output ratio. The key aspect is that a reputation transaction is not zero sum while an economic transaction is. So i.e. releasing a patent would raise the inventors reputation without lowering reputation of the users of the idea. The economy has at its best increasing returns dynamics, but the monetary system is based on a stone age analogy of giving a X for a Y.  Which is kind of weird!

Replies from: ChristianKl, henrik-karlsson
comment by ChristianKl · 2021-06-03T09:50:37.235Z · LW(p) · GW(p)

There are countries like China were reputation is more central then in the West. A powerful person who sufficiently ruins their repuation gets thrown into jail in China. A huge problem with this is that it sets up incentives to kill free speech and is one of the key reasons why you have less free speech in China then in the West. 

Replies from: jmh
comment by jmh · 2021-06-04T01:47:17.677Z · LW(p) · GW(p)

I was also under the impression that China, and many other Asian countries where status and "face" is highly important were generally considered low trust cultures -- meaning you expect your neighbor to behave opportunistically if you're not watching closely.

comment by Henrik Karlsson (henrik-karlsson) · 2021-06-01T19:58:55.389Z · LW(p) · GW(p)

Could you expand on this thought?

It is an interesting point that there is something non-zero-sum about a reputation transaction. "The hippie" mentioned in the essay worked very much on this assumption. By being extremely trustworthy he creates a reputation that expands his network of trust, so he can get more help with things he needs done etc; but being helped by him also raised my reputation, as I got his stamp of approval. (This can happen in a market transaction as well, since markets and relationships of trust bleed into each other.)

But I wonder how that can be formalized without being game:able? I'm a bit fascinated by experiments in the blockchain community, like bitclout, were peoples reputation (or whatever to call it) is a token that can be traded, so there is an incentive to invest in people who one precieves will grow in reputation. But I can't wrap my head around how a system like that would behave if you pour in people and have them figure out ways to game it.

What's your intuition as to how one would implement what you suggested?

Replies from: arvid-kaestel
comment by Arvid Kästel (arvid-kaestel) · 2021-06-02T19:52:38.084Z · LW(p) · GW(p)

Well I'm working on a longer text on the subject, but the short version is analogous to the situation between nations with sovereign currencies. How is this trade possible if you don't only send goods but money? If there is a solution for this situation there must be something that should, at least in theory, work for individuals. 

One important point is that reputation is public and that it is not isolated to one situation. But a sketch is that my own reputation is affected depending on whom i interact with. So optimally i trade with people of good intent. Good intent would be to maximize not ones own value but aim for the social optimum. This is not however possible with ordinary money since transactions are zero sum. We need a system that can handle something similar to the clark pivot.

MMT is actually a step in this direction, since allows a trusted party (the state) produce money in certain areas where we have increasing return dynamics, thus not creating inflation. It has other drawbacks though.

I'm also playing with the thought of using reported utility as trust. Since you cannot experience more than a limited amount of "happiness" per time it acts as a trust guard. That would lead to the possibility of a system that creates a utilitarian optimum. It would strike the "optimal" balance between redistribution and incentives. Long story short...

comment by ChristianKl · 2021-06-03T10:46:55.742Z · LW(p) · GW(p)

Trust is a very valuable commodity. However it's a mistake to assume that just because you know someone you should be able to trust them.

I don't find the statistics right now but if my memory is right people generally underrate the chance of various criminal behavior from acquaintances compared to strangers. Competency of friends is often overrated compared to strangers. 

On average your friends are unlikely to be the best in the world or even in your city at what they are doing in their job. While the hippies in your example might be more skilled then the average carpenter they are an expection and a lot of people do now average carpenters. 

Replies from: henrik-karlsson
comment by Henrik Karlsson (henrik-karlsson) · 2021-06-03T10:59:59.896Z · LW(p) · GW(p)

Good points. These are important considerations to keep in mind when deciding when to rely on markets and when on friends, and also when deciding whom to include into your network of trust. I try to be highly selective about whom I transact with outside of markets, as well as in markets. Generally, the people I trust are not perfectly correlated with the people I consider my close friends. But the bias that makes us overrate people we trust is important, and I will try to correct for that, which might make markets more attractive compared to networks of trust.

comment by Tao Lin (tao-lin) · 2021-06-03T03:49:49.803Z · LW(p) · GW(p)

These problems sound significantly worse in Sweden than in the US. For example, roof replacements cost 8-30K out of 30K median income here. The problems you talked about are real and fundamental, but I think low hanging problems with the market in Sweden are responsible for most of the cost.

comment by Srdjan Miletic (srdjan-miletic) · 2021-06-01T03:07:54.444Z · LW(p) · GW(p)

So I think it's interesting that the market seem so expensive for these tasks. It makes sense for the carpenter case due to the information asymmetry but I don't see what there aren't more affordable moving companies in your nation.

As for markets vs trust a few thoughts:

  • You seem to be of the view that most consumers (irrationally) go to markets for most of their transactions when they could be relying on trust instead. Is that really true? Many of the goods I get, relationships, conversations, essay feedback, sex, childrearing, etc... I get through non-market trust based mechanisms. Isn't it the case that we just don't think of non-traded things as goods?
  • You seem to be of the opinion that trust is massively more cost effective than markets in many cases? Again I'm somewhat sceptical. I think your selection procedure is to take things you do on the market and find the few of them that are far more effective when done outside of the market. This leads to a biased sample because you're selecting specifically for things the market does badly.
  • Another thing to consider is the cost of trust based networks vs markets. Specifically, you're limited to in-network people and hence to only a very small subset of goods/services.

(By the way, still think the "trust can really help with transactions" gist is super interesting and useful. Even outside of laws and just in terms of cultural norms, living/working in a low-trust culture can be a shocking experience to anyone accustomed to a western european country)

Replies from: henrik-karlsson
comment by Henrik Karlsson (henrik-karlsson) · 2021-06-01T06:03:16.394Z · LW(p) · GW(p)

Thanks for your non-market essay feedback!

I don't know exactly why shipping companies are so expensive in Scandinavia. Part of it is of course high tax and insurance cost, but that can't explain the entire difference in cost compared to doing it your self. Maybe shipping is unattractive as a job for various reasons, so we have an undersupply of shipping companies?

And you're definitly right that my post is cherry picking galore. I should have underlied that it was more of a hunch! I don't have data to answer the question whether people in general could gain by moving more of their consumption of the market (or vice versa: gain by moving things they now do with in their network unto the market). What I do have is my single data point, and I've been finding a lot of ways to reduce how much I need to labor to be able to satisfy my needs by growing my network at the expense of my salaried career. (After having the realisation I share in the post, I've also expanded my network to supply me with food, and I've set up so I can loan a car instead of buying it.)

So the point I'm trying to make is a slightly different one than "trust is massively more effective than markets". Its that one can have a higher marginal return on ones time if one consciously explore if what one wants to consume can best be served by the market or by one's network. For me, I've been able to get higher returns by investing more in growing my network. (Though there are some things that I now do outside of markets that might be better served by the market!) But again, that's one data point, and one has to go case by case. My intuition though is that since that more people could gain, given that for example the shipping companies operate though one can easily beat them by a wide margin by having (cultivating) good friends. And I thought that could be useful to share.

Replies from: ChristianKl
comment by ChristianKl · 2021-06-22T11:38:08.466Z · LW(p) · GW(p)

Maybe shipping is unattractive as a job for various reasons, so we have an undersupply of shipping companies?

Even if it's unattractive, a global company like Fedex could come in and likely still higher unskilled labor for it.

Replies from: henrik-karlsson
comment by Henrik Karlsson (henrik-karlsson) · 2021-06-22T15:03:38.581Z · LW(p) · GW(p)

Let's hope! Probably that throw away guess is not the correct one for why shipping is so expensive here. There might be some regulatory mess I don't see.

I do find it interesting, though, that certain jobs might be undersupplied because of presitige and social factors. I think Scott Alexander mentioned somewhere that plumbers make about the same as doctors, in the US. So if you where a rational economic agent, you would go into plumbing, because its less costly to learn, less stressful etc, but that doesn't happen, I guess, because - who knows a cool plumber? (Super Mario not counting.)

comment by Adam Zerner (adamzerner) · 2023-05-16T06:19:24.420Z · LW(p) · GW(p)

This is a super interesting and thought provoking essay. Thanks for writing it! Although it does feel overstated in this essay, I really like the core point of how trust makes things better, and how there are costs to enable transactions when trust is absence (police, lawyers). It's new, plausible and powerful: a combination I don't come across all that often.

"Markets!" my econ professor would proclaim in a thick Armenian accent. "If only they were a machine! People would love a machine where you can put in wheat and pick out a helicopter."

That is a marvelous machine.

Wow. It never clicked to me like this. Thanks. 

Someone who has done an intro class to economics could probably argue that the markets would correct for this. Isn't crooked carpenters a great opportunity for honest ones to increase their market share?

No. Not if the consumer never realizes what happens, which they usually don't – because of asymmetries of information. Usually, the customer only sees the wholesome, friendly-looking attitude – and gives a rave review after being exploited.

100%. I've realized this more and more as I've grown older and seen many examples of it.

One sticks out. I worked for a company that builds software for other companies. Ie. one of our projects was building the mobile app for CVS. They charge like $150/hr for developer time, pay developers $40k/year, consistently deliver projects that are late and flawed, and have constant chaos internally.

I was a naive 22 year old when I worked there. It boggled my mind that they stayed in business! How do they survive? I get that inefficiency might last for a little while, but surely they'll be discovered as imposters and go out of business eventually, right? Well, it's sort of the opposite. They get to say on their website that they built the mobile app for CVS. Boom: social proof. My understanding is that it's this sort of positive feedback cycle of social proof that keeps the money flowing in. That along with, as you say, asymmetries of information.

Asymmetries of information are fundamentally a good thing. If you know something that I don't, we can pool our knowledge and achieve things that are too complex for a single mind.

Nitpick, but I think what you meant to say is that given the existence of information asymmetry, it's good that people can work together. If you could wave a magic wand I think you'd probably want to just transmit the knowledge to the less knowledgeable person.

And the few things they do produce in-house – say food – might not even be the things where they have a comparative advantage.

It's great to see you note this! It's always seemed weird to me that in our hyper-specialized society, people still cook so much. Cooking in bulk is more efficient. Why are there not places where you could get a huge pot of chili for like $1.50?

In his podcast, Adam Ragusea talks about how in east Asia it's often cheaper to eat out than to cook at home and so lower income people mostly eat out. I just re-watched it and am not really seeing a coherent explanation. "Rice" is the one word explanation. He goes into the history and stuff. But rice is cheap in the west too. He mentions labor, but talks about how that doesn't explain it because countries like Japan don't have cheap labor.