Do you care about people who are suffering?

I bet you do. That’s part of the standard “I’m a person” package that you get just by being born a person. I’ll go a step further: I bet you care more about people in a lot of trouble than you do about people who aren’t in any trouble.

Oh, you probably have more complicated feelings than that because of your life experiences. Maybe the word “more” should be replaced with “urgent” for instance, but those two things are probably true in general.

They’re not true because you resent people who are doing well, either. They’re true because people who are doing well are by definition okay, and because one of the things you very quickly after you realized that some people are suffering… is that you’re limited in your ability to help.

You can’t help everyone. You don’t have the time, energy, or resources. You can’t get to them all. It’s not possible.

So, you have to make some choices.

1. How are you going to use your time?

2. How are you going to use your energy?

3. How are you going to use your resources?

Being a compassionate person, at first you’ll probably say something like: “I’ll give everything I have, and everyone else will as well, and then we can get the problems over with as quickly as possible and get back to being happy.”

Except… it’s not that easy.

You can’t get everyone to agree on what to do to help, for one. Sometimes, after you fix a problem, it doesn’t stay fixed for another. Some problems need constant effort just to not get worse. Even trickier, sometimes you can help so much that you move yourself out of the helper category into the suffering category and make the problem harder for someone else do fix. Even more complicated is the fact that some problems regenerate after you totally fix them, and you find yourself using your proverbial fingers trying to plug proverbial holes in proverbial dams, but you’ve never got enough fingers for all the holes.

So What Do You Do?

Over time, and generations, you realize there are some odd properties of team work that can help with solving problems. In essence, if you work with other people, you can magnify one another’s labor in a way that is greater than the individual sum of your labor. Although this is only roughly true, you find that two people working together can do the work of three people working separately. Ten people working together can do the work of a hundred working separately. It goes on and on and on.

1. If you use your time, energy, and resources with other people in the right way you can build a team. A team is some kind of potentially immortal pattern that might last long after all the people who started the original pattern are dead. You can center this pattern around solving one of the problems.

2. The pattern you made with all the other people working together is waaaaay better at solving problems than all of you working separately because you get some kind of network effect as more people get brought into the pattern. It’s like everyone becomes one super-intelligent person based on what everyone in the pattern is best at.

3. If everyone in the pattern, or who touches the pattern, has to make some kind of bet about what the future will be like, your pattern gets more and more efficient and people start being able to re-shape the pattern to solve the problem better and better. In short, people INVENT new ways to solve the problem when they have skin in the game.

Other people see this happening, make their own patterns, and solve other problems. The world overall gets better, bit by bit. In fact, you’ve practically made it so people forgot there was ever a problem in the first place.

That might actually cause new problems.

As new groups of people get brought into the pattern they may forget that the pattern is there for a reason. In fact, human beings themselves are a kind of pattern and maybe they need some level of problems in their environment to calibrate themselves. Without the presence of major problems, people may try to get the problem-solving pattern to do other things and then the problem the pattern was solving comes back, or worse, the pattern starts MAKING bigger problems.

Your pattern is smart, but it’s not sentient. It can think high-level, abstract ideas, or answer if the problem it’s solving is something you care about. Or even if it’s good to have that problem solved. Your pattern will optimize anything, but it’s blind to if the goal is good or bad.

What the hell about we going to do about that?

What Does It Mean to Think About Complex Systems?

With all of these patterns, you’ve run into the problem of Complex Systems.

Let’s talk about that, because it’s probably the single most important organizing principle in the modern world.

Think of two balls and then connect them together with a sting. Congratulations! You’ve made a system, or a pattern, if you will. Pull on one ball, it pulls on the other. Then switch and pull again. It’s nice and simple, isn’t it?

Okay, now replace the string with elastic. So, now you’ve got two balls and then this tension that exists between them that says something about their relationship. If it’s real tense, then those balls are going go react very quickly to one another. If there’s a lot of slack there, that signal isn’t going to move forward as quickly.

Then add in more balls.

Connect every ball to every other ball with another piece of elastic.

Now, every time you do anything to one ball, the effect propagates through the entire system.

You’ve built what I call a maximally recursive web object. For any N balls you have in that web, you’ve got N(N-1)/2 connections between them. What that means in English is that that as you start adding more balls, the number of connections gets comparatively huge.

Let’s start making this even harder.

Double-connect some of the balls. Cut some of the other elastics at random. Add in some different kinds of balls at different weights. Blindfold yourself so you can’t even keep track of it all.

Now, pick two balls at random.

Pull on one of them as hard as you can, and then predict what’s going to happen to the other one.

You’re not sure, are you?

It’s too hard to track the connections, or how what you do will dissipate into the web. There’s too many connections for your brain to follow.

What… well, what the hell are you supposed to do now?

Webs gotta Web if they’re going to Learn

What you’ve built is the fundamental structure underlying economics, evolution, weather, language, the scientific method, and about a dozen other things I’ve thought of so far. What you’ve built is at least in part, a Recursive Web Object. Every time you touch one component, your touch propagates through the entire system at different rates, where it in turn is changed by the signal from other components, or given different weight due to the tension of an elastic, and so on and so forth.

What happens in a web is too complicated for us to untangle.

We can’t directly control output. It’s an impossible task. No one is that smart. Frankly, if you were that smart you wouldn’t be building giant webs to take care of problems. You’d be God.

What you can do, if you’re careful, is control inputs and rewards. In that way, you can TEACH the web to perform a certain task even if you don’t understand how exactly it’s learning or functioning.

So, you get some fancy new balls. We’ll call them Nodes. And you get some fancy new elastic bands. We’ll call them Edges. And the Nodes have the ability to make connections on their own, and the Edges can adjust their own tension. How they do that is going to have to remain mysterious and magical right now, but they can.

Here’s the really rough part. The only chance you have of making that web do anything productive is to… let it try, let it succeed… but also let it fail and let it get hurt. Success and failure are signals, and the Web needs both to make sure it’s calibrated to the real world and is functioning correctly.

Here’s what I mean:

Some of those Nodes are going to be facing the problem you want to fix. We call those “the Input Layer.” Some of those Nodes are going to be facing you and asking if they solved the problem right or wrong, and we call those “the Output Layer.” When the Nodes are wrong, you have to tell them by taking away resources, and then what happens is that information cascades back through the web and rearranges the tension on your Edges and the connections between Nodes. When the Nodes give the right value, you reward them with additional resources. You do that again and again, and as you do that somehow like magic, the system evolves over time to get better and better at solving the problem.

What all this means is that when you build a system, or a pattern, the output will depend more on what you incentivize the web to make than what you intend. It depends on the web’s inputs and it depends on you having good judgment on the output. That means good intentions aren’t sufficient to design a smart system unless those intentions are balanced by good sense and prudence.

Because people have a lot of stuff to do, we tend to maybe not worry so much about what our web is taking in as input, as long as it’s close enough. And if the answer is close enough in the output, we don’t mind. Except as you do that, the web gets worse and worse overall until it starts to become catastrophically wrong. The only truly consistent and fair judge is reality.

So, now you’re back in your Pattern of People, trying to do good in the world. You’re a Complex Pattern, a Recursive Web Object, and you want to do good. Let’s try to do some good.

Why is it Hard to Think About Health Care as a System?

What you want to do right now is fix Health Care in the United States. It’s a mess. Worse, people are dying and they need help NOW so you’ve got to get your Web in place and optimized ASAP. People are screaming, as a matter of fact. All those people you need to help are in a state of emergency. If you even take a second to think about the problem, they’re going to die. They’re going broke and scream because they can’t get the help they need…

Still, you have to understand what’s broken so you can fix it.

So, you descend into a cold logical place where you have to weigh the needs of many and the future against the needs of a few right now.

What is the Pattern already there that’s trying to fix Health Care? Why isn’t it working? What in the hell happened?

What is Insurance?

Insurance is an old pattern, and it makes a lot of sense on the surface. When something doesn’t happen very often, but it would be really bad if it did happen, a large group of people can give a little bit of their resources to the pattern. In return, the Insurance pattern recognizes when that rare occurrence has happened, and delivers an abundance of resources to the person who was hurt so they can recover.

It works great for cars, because people don’t get in car accidents that often.

It works even better for houses, because people’s houses don’t get destroyed that often.

If you’re young, it even works for when people die suddenly because most young people become old people.

Here’s the thing about Health Care, though… everybody gets sick. It happens to people all the time. Most people will break two bones in their lifetime. We get cold, flus, food poisoning, at least in a mild form probably every year.

The Health Care Pattern of the United States is based on Insurance, and Insurance doesn’t make any sense to fix Health Care because Health Problems are Common. In fact, as you dig deeper into the problem it’s worse because most people can only get Health Care through their employer, and that artificially ties people down to jobs and takes away some of their resources without them ever getting the chance to see what they lost.

As you look you see the Input Layer of this Web is the Hospital Billing Department and the Output Layer is handled by the Insurance Company.

The Health Care Web isn’t optimized to fix health problems, or make them cheap. What you’ve found is monstrous. The Health Care Web is optimized to make everything part of insurance, right up until insurance costs as much as it possibly can, as a segment of the bigger Web of the Economy.

The Insurance Company and the Hospital Billing Department are locked in a predator prey relationship, and what they’re both trying to get better at is taking your money and giving you little in return.

What is Buying Power?

Say you have a lot of resources, and you declare that you wish to buy a lot of something. You need a billion widgets.

Widget makers from all over the land approach you and show you their widgets, and you get a chance to judge them and test them out. Because you’re a big widget buyer, widget makers want to come to you to sell their product because if you buy from them it’s a big reward.

Buying Power is the basic pattern of State Health Care, like in the U.K. or France.

Because the state is the only buyer of Health Care services, anyone who wants to supply health care has to approach the state and outbid their competitors on the state’s terms. That state gets all the choices because it controls all the power.

This gets a low price on widgets, or in this case health care.

What are the Trade-Offs?

I know what you’re thinking.

You’re thinking: Let’s do the European model, duh! Of course that’s the best way! Let’s do that right now!

And honestly, if we had to choose what we have now versus that model, I’d have to agree. There’d probably be degradation over time without the U.S. for reason I’ll explain below, but between those two choices, Buying Power seems to do better than Insurance. Insurance is a broken model for trying to provide all of your health care needs. It fundamentally doesn’t make sense, because you’re paying both for the Health Care and the administration of insurance which adds unproductive cost.

Here’s the problem though… that’s the way the whole world used to work when we had Kings and Queens. Only one person was able to buy things. That worked great when you had smart people in charge, but not great when you didn’t. Also, nobody invented anything new.

If you’re taking something that already exists, and have smart buyers, Buying Power can deliver you a great price.

But what if you’re talking about something that doesn’t exist yet? It can kill innovation. People who invent are going to have to put a lot of time and energy into making something new, and because there’s only one buyer they’re going to have to trust that they’ll be able to sell it. It’s like asking a super smart person to play the lottery, especially if their invention is difficult to explain. Why would they do that? That’s not likely, so they’ll probably spend their time inventing something else unrelated to medicine.

Another weakness is that your output layer isn’t necessarily being judged accurately. Reality is the best judge, and if you have multiple webs hitting the market, usually one method will win out over the others and it’s better if the people managing the web are the people the web is being built to help.

Insurance, though, you know just can’t work the way it’s designed. Insurance is only good for things that don’t happen that often, or to that many people. We know that isn’t a good idea to fix health care. However, on things that don’t happen very often, insurance is really good at driving prices down and making those kinds of bills affordable because the group can carry the cost and the administration doesn’t have to be that big.

So, you decide, is there a way we can use Buying Power and Insurance together?

What Do We Want?

Let’s make a list of what we want.

1. We want people to be able to get medical care.
2. We want people to be able to get medical care without going broke.
3. We want people to invent things to fix illnesses better than what we have now.

How do We Get What We Want?

I’m going to lay out what I would do if I had a magic wand to fix the Health Care system, and then I’ll talk about what I think is actually possible as a bridge solution. What I think we need is a new legislation that combines the strengths of both systems, as opposed to what we have now, which I honestly think in some ways combines the worst aspects of both.

1. We need to drastically reduce what is covered by insurance, globally, for everyone.

Insurance works for infrequent catastrophic events that require a large amount of resources to correct. Let’s let insurance solve those problems and ONLY those problems.

If you’ve got toenails for skin, we’ll make sure you’re covered. If you’ve got the flu? Eh… I think you can take care of that one on your own. If you can’t, I’ve got an idea for that too, but you still need to pay out of pocket.

We’ll establish an average, based on the average cost of insurance right now. Anything with a cash cost below about $350/mos, which is about the monthly premium cost to an individual per month, you just pay out of pocket. As things get above that, we start adding them to the insurance list.

2. Insurance should be money you just say goodbye to every month, but you shouldn’t plan for your Health Care that way.

As we’ve said, insurance isn’t a workable model for Health Care. It makes more sense for you to buy low priority items, like getting your arm broken, out of pocket. So, we’re going to take a portion of what you were paying for insurance before and we’re going to create a Health Savings account.

Your Health Savings account will be managed by a bank that invests your money in medical technologies and returns interest to you. In fact, we’ll create a bunch of tax breaks and incentives to make it very lucrative for the bank and those technology companies to do that. What’s more: if you get to zero in your account, you are lent money by that bank regardless of credit.

The trick is that you have to, by law, contribute a certain percentage of your check into that account every month, and you’re going to be charged interest on that loan so that nobody else is losing their money. No one can come after your assets, and you can’t be sued for that money, but so long as you have an income you have to pay a portion of it into that account.

Every year, the performance of these accounts will be determined against inflation. If the return of these accounts falls below inflation, we take the least common medical concerns that caused the deficit and we add it to the insurance pool.

So, we’re giving some powers to government here that it doesn’t have right now:

A. The government decides what’s insurable and what has to be paid out of pocket.
B. The government forces you to put a portion of your income up to a certain amount in a health savings account
C. The government mandates that a bank not be able to sue your for that money, but also gives tax breaks and incentives for them to invest this in medical technology

3. The Government puts a Bounty on Everything on the Insurance List

Right now, if a medical technology company makes a breakthrough they have a Monopoly on it due to patent law. That’s actually pretty good as it incentives people to create and gives them some assurance they can recoup their investment. However, I think we can do better.

We want multiple companies to try to make products. Also, research and development and production are two very different skill-sets. They don’t necessarily belong together.

So, everything on the Insurance List gets a bounty. What that means is that if you develop a technology that lowers the out of pocket cost of treatment for that illness to less than $350/mos you don’t have to do anything beyond the development and showing that it works. You just get a check for inventing it and the copyright goes to the public domain.

The cost of each bounty will be set at some percentage of the total cost of care for each illness in a given period, and every set period of time where a cure is not found the percentage will increase up to a certain threshold. The cost of those bounties will be taken out of your insurance money, and the development of those cures will be driven by firms funded through your Health Savings accounts so some of that money comes back to you and everyone wins.

Over time, there will be fewer and fewer items on the Insurance list, until the only thing left is Death.

This way we can make sure that people are still incentivized to cure things, and that nobody gets a Monopoly on a life-saving cure that will prevent prices from coming down.

What you’d hope to see here is dedicated research teams at universities racing one another to collect the Bounty, and then medical supply firms around the company taking those breakthroughs and producing them at inexpensive prices.

This lets us use Centralized Buying Power to get what we want: Cheaper Stuff, without also having to give up innovation.

What was that again?

Because this Web takes your ability to pay for Health Care as its input, and presents treatment Options as Outputs, this system is overall designed to bring down the cost of Health Care while also making sure new options are continuously presented to make everyone healthier.

Can this Happen?

Oh, no. Not at all. This would be so disruptive the present system would just collapse and never shift over into the new system and leave no one around with the wherewithal to implement this system.

What Could Happen?

1. We could make insurance companies and our medical providers accountable, for a start.

Right now you walk into a hospital and nobody can tell you what your treatment is going to cost, even if you have insurance. We should pass legislation demanding they tell us the price and be bound by it, before we agree to care. In fact, for things like the flu and a broken arm, they should be obligated to advertise those prices so you can choose which medical provider you want to visit.

That would incentivize hospitals to bring down their prices and agree payment plans in advance so you know what you’re getting into. Hospitals would have to compete for your business.

2. We could start to create incentives for alternative health care systems.

What if instead of paying an insurance company, you paid a subscription fee to your local hospital? And that would count as insurance in the eyes of the government and the hospital would agree to cover you for catastrophic care. Even better, what if you had that arrangement with a doctor or a group of doctors instead of a company that owns the hospital?

That way the doctor is incentivized to try to correct what’s wrong with you so you stop coming back, and is incentivized to innovate, because the doctor makes money by fixing you as cheaply as possible. Namely, this gives your doctor skin in the game and makes sure that the money we spend is actually going to the health care providers and not administrative costs.

Maybe the doctor has some kind of Re-Insurance (Insurance on Insurance) so that if you’re ever not local and something happens you’ll still be covered, but if the doctor has skin in the game that means you’re probably going to get better care overall. Also, it’s probably cheaper for the doctor to work this out as it would give them direct control over pricing and make sure that they buy responsibly from medical providers.

3. Allow people on government health care to experiment with some of these systems

I know this is probably an unpopular viewpoint, but it’s unfortunately born out in the evidence: Government involvement makes things expensive. When the government gives out what is perceived as free money, prices go up. It’s supply and demand.

Right now, there are a lot of people not getting good healthcare.

Maybe if we slowly grow that subscription model and give doctors a direct line of sight over their care and patients some people would want to move over to that model. It might be cheaper and a good strategy until we can update things completely. It could foster a powerful economic sector of informed doctors who are the first to have their hands on the money when it comes through the hospital, so they’d probably lobby government to make sure they kept that power which would in turn make the system better and better.

4. Get Rid of Cost Plus Contracts

The government pays absurdly inflated prices for basic care, because it allows contractors to fulfill orders at a cost (how much it cost the company to produce it) plus a certain profit. So the contractor has no incentive to drive down costs. Rather, the government should offer a fixed amount of money, and demand the company deliver. This would drive down prices overall, including for the average consumer, as the Supply and Demand curve would begin to normalize.

5. After this has gone on for a while, incentive Health Savings accounts

Give people giant incentives to put their money into a Health Savings account. Let them pass it onto their children tax free when they die, for instance, to avoid the Estate Tax, because that would incentivize people to save for their old age, increase the Buying Power of the average consumer looking for Health Care, and make sure we’re pumping money into the medical technology sector.

In Summary

It’s sick that people work their whole lives, and get sick for a few months at the end and lose everything they’ve ever worked for. Costs are inflated far beyond reason due to bad economic models, and innovation is incentivized to create Monopolies. Every part of the current pattern sucks.

One day, I’d like elderly people to have a nice robot (at the cost of a car) that takes care of them, and provides them with the cheap medical care we’ve created, until they fall asleep and don’t wake up, and then the robot gets sold to another person getting to the end of their life. Plus, leave something for their kids so their family gets wealthy over time instead of getting reset to zero because someone went to the hospital for three days, and it somehow cost half the wealth they had ever created in their entire life.

Categories: Uncategorized


Evan · October 13, 2018 at 9:40 am

I am pleased to announce that I have discovered the habitat of the reclusive Recursive Web Object. I’m wondering: how would you contrast the RWO with the law of unintended consequences?

I like the idea of emphasizing the relationship between patient and provider—creates buy-in for both parties, which is crucial and also beneficial for all parties involved. Although I’m not sure how doing so will decrease administrative costs. It seems to me these administrative costs would be similar, if not more, than the costs that an insurance company would normally bear—you’re just deferring that overhead of administrative processing to the provider or hospital instead of the insurance company.

Paramount in any healthcare debate is the x-factor of the human mind. With any single payer system, the government must necessarily force its citizenry to buy in to the system. While this does create buying power and reduce administrative costs, the bigger issue at stake is that of freedom. People (especially Americans) don’t like to be told what to do. Forcing Americans to buy into a system that inevitably redistributes its revenue to subsidize the many different contexts and lifestyles that exist within the American landscape is simply unpalatable for even the most altruistic among us.

Thanks for your thoughts, Andrew.

    AA Peterson · October 13, 2018 at 1:07 pm

    A high level principle I have is: Set goals like a person, but think about how to achieve them like a machine and break the machine into components.

    When I think about freedom, I try to think about what that actually systemically means in the web object described above. If human beings are Nodes, and our relationships are Edges, then freedom is the condition such that Nodes are able to control and generate their own Edges so long as their Edges don’t destroy/harm other Nodes. If you look at it this way, you can look at a system without having to say is this free or not? You can look and say “is this more free or less free?” because you can look at the movement of Edges between Nodes and their degree of shift over time. I feel what I’ve described above, because it presents more choices and transparency and gives more freedom of action to each individual person is more free than the present system.

    The system is also designed to become more free over time. If there are abundant healthcare options when you are sick, and conditions above that shrink administration over time, then when you are injured you are interacting with an environment that gives you increasing options to choose from. With technological improvements over time, you’d arrive at a condition where there is no insurance because all illnesses have cheap treatment.

    As for unintended consequences, that’s sort of what Recursive Web Objects are. It’s an understanding that what we try to do when we change the world is unlikely to work because most solutions involve 1. Changing Nodes 2. Changing Edges when what we should be doing is finding cases where blind systems have incorrect inputs and outputs and repairing this so the system corrects itself over time. There needs to be a much more robust engineering science for this, because I think this is how we need to manage ecosystems, etc, and I dont’ think we have a choice if we want to do that or not if we’re going to claim the stars.

Evan · October 17, 2018 at 8:57 am

I suppose one way of looking at the RWO is as a method of trying to herd the cats of unintended consequences, no?

Before we can accurately answer the question of “is this more or less free?” we’d have to get down the brass tax of concrete numbers and percentages of income people will be relinquishing to bureaucrats in Washington. This is the most crucial part of the debate that I feel like you neglected—correct me if I’m wrong.

I like the idea of the Health savings account and how the power there is spread out among banks who invest that money to incentivize innovation. But it seems like the glaring issue is that of how to pay for all of the healthcare procedures that are above that $350 cut off (or whatever number we decide). Taxation alone simply will not foot that bill.

How do you ensure that the government owned monopoly of buying power in the health care industry does not end up actually causing all the widget makers and widget providers to stop making and providing widgets for us and instead make “fidgets” or instead make and provide widgets for another land?

In order for your model to work, it seems like there would need to be some sort of Node to incentivize health care providers to actually practice here in the US. You said it yourself, government ruins everything it touches. Ok, you didn’t say it quit like that, but this principle is true in basically every metric. The only reason our military is good at what it does (debatable) is because the purpose of the military is to break stuff, which is exactly what government does when it touches anything, so they go hand in hand!

I dunno, I really like the theory of this, but like you said, I cannot ever foresee it being implemented. Especially when you factor in all of the political passions that taint health care nowadays (transgenderism, abortion, assisted suicide) these issues are intractable. Fun to think about, nonetheless.

Where’s the line of the intellectual’s role in planning and managing complex systems and the good old fashioned Invisible Hand of Adam Smith? I think I prefer the hand over the web.

Leave a Reply

Your email address will not be published. Required fields are marked *