I don’t normally write about breaking news, as others do a great job of this. But since we do talk often about markets and the various ways in which complex systems work thought a riff on risk, and how it spreads around our economic engine, and how much of it is irreducible, was apropos. It might even be helpful. Enjoy.

There's risk in everything. The places we see them the least are the places we know the least about. Most of the modern economic structure exists to move risks around, and successfully manage them somehow.

The core theory of this is that risk ought to be held by those who understand it the best. That's a large part of what the financial system does. It moves risk around and gets paid for the privilege.

Transaction failures

Let's look at an example. Say I want to go buy a laptop. Bear with me if some of these sound dumb, there's a reason. Here are a bunch of ways in which that transaction could go wrong.

• The laptop could be fake. I thought I was buying a Mac and turns out it's sold by Apples, a fake company.

• The laptop could be faulty. I thought I was buying a new Mac and turns out it's a refurbished crappy one.

• The store could defraud me. The cost is $2000 and they charged $2500.

• The store could reject me and my money.

• The store could reject my payment method.

• The store could accept it and then change the amounts.

• I could buy it and then not pay my bills

• I could buy it with stolen money.

• We could transact in good faith but the currency falls apart

• We could transact but the intermediary financial system collapses

As you can see, there are a rather large number of ways. However, most of these fall into just a few buckets:

• Fraud - from the buyer or the seller. The good being sold or bought is not as per the contract.

• Credit risk - the risk that someone will go bankrupt before they hold up their end of the transaction. Again, on either side, though mostly with the seller

• Market risk - the risk that while the transaction is happening, the market shifts and prices change dramatically. Less relevant with consumer goods like laptops, but more relevant with any instrument that has longer duration (buying a bond, buying a house, marriage)

• Liquidity risk - this one’s special to money, because it’s a risk that ongoing obligations can’t be fulfilled because you have no money. Like if you owned a house but zero in your checking account.

• Operational risk - the risk that someone will screw something up!

There are also others, like general business or reputational risk - the risk that by doing something you’ll look dumb now and forevermore.

To do almost any of these though, you touch the financial system. In most cases that means you touch several banks. Yours, where you keep your money, and the seller’s which receives the money, amongst many many intermediaries in the middle.

Which can sometimes cause problems. Because-

The problem

A short aside on the banking system. I promise this will be short.

Many people have ideas on things they could do. Some of those things help other people, by giving them things they want, or inventing new ways to do things.

Doing things takes money. But doing things can also make you money. So you need someone to help you do things so you make money by doing things, and then, maybe, split some of that money with them. Congratulations, you’ve invented equity.

But this is risky. What if you can’t afford to lose your money, but still want to help those who want to make things? Well, they can do things with money that are less risky. Like making sushi. Everyone loves sushi, so chances are your money will be safe. So you’re happy to give them money in exchange for a promise that they’ll pay you back a fixed amount of money. Reduce the variance of equity investments, so to speak. That’s debt.

If you want this to happen all across the economy, where people from Paris want to fund people in Bordeaux to build baguette shops and capitalise on the craze, then you probably need an intermediary to move money around from a lot of people in Paris to those in Bordeaux. That’s banks. If you incentivise them by letting them make some money as well, they do this job well.

They normally do it by lending out the money of people from Paris to those from Bordeaux. Except the people from Bordeaux who took the borrowed money can also do the same and lend to people from Lille. And Lille to Marseille. That’s fractional reserve banking.

But of course if the banks screw up, then the entire ball of yarn gets tangled. Especially if they screw up all at once. So you don’t want them to go totally wildcat. Which means you want them to give them guardrails, and monitor that they’re not going stir crazy, like watching children in a playground. That’s regulators.

Sometimes though, you have to have a way of making sure that the entrepreneurs building things are making things that are useful. Like they’re not wasting their time trying to make perpetual motion machines, or weird social experiments about the concepts of money, or getting some underpaid chap to cycle desperately fast with a pint of milk. Which means someone needs to loosen or tighten the amount of money that the banks get to give out to people, giving banks a sort of high (or low) accepted minimum. That’s, kind of, the Fed.

If everything works, you have enough money being made to support enough banks who give it to enough people building enough cool things that we collectively get flying cars.

If.

How we solve them

For some of these problems, we choose regulations to solve them. For some we choose markets.

If it's to do with fraud of any sort, we try to get our legal system to sort it out. This works okay today, though it still is an issue for anything that isn't heavily monitored or easily analysed.

If it's to do with counterparty credit risk, we try and fix this through legal systems and various insurance schemes. We try get a situation such that it's not outright fraud, but incompetence, we get regulators to try and make sure this doesn't kill everyone. The trick is to guess at what the actual level of incompetence is likely to result in ahead of time, and prepare for it. So, you know, its not likely to be perfect.

For something like buying a laptop this isn't too bad. But if you want to get a car loan or a mortgage or buy a money market fund, then you rely on regulators to maybe cap the problems this might cause. Like how the FDIC tries to make sure you're not wiped out if your bank's executives really like bottle service.

If it's market risk, like you want people to be saved a bit from the market turning against them, then mostly it's caveat emptor, with some restrictions on the ‘emptor’. You have some laws protecting people from their own idiocy, like stopping non accredited investors from losing money in startups.

If it's a systemic problem, like currency collapse or your banks collapse, that's backstopped by various bodies who create insurance solutions or implicit “put” options. But there isn’t a hell of a lot you can do about it, which is why we try fix it as soon as it happens.

And we need those options, because without those we basically have no ways to globally coordinate ourselves, or to stop the larger, more systemic, risks from causing catastrophe. As we’ve seen in our history time and time again.

What actually happens

Now with all of these justifications, what can we actually believe will happen? Well.

• For many of the normal purchases, mostly the laws work. Your meats okay, milk is fine, laptops work. For smaller purchases there's fraud, but that's often baked into the price, since not everything can be cost effectively monitored.

• For the counterparty issues, we've created a Byzantine web of trust to help us all work with each other. We can't trust that the companies who serve us won't die out always, but it's more reasonable. It's totally fine for immediate purchases, mostly fine for buying goods, fine for services which is delivered afterwards, and kinda fine for financing of any of the above. Basically for the areas where the contract extends for months or years, you need ways to trust the other party to the contract will, you know, survive.

• So we ask the financial institutions to adhere to a bunch of regulations of keeping enough money aside such that if something happens they're covered. We tell them to sure, go take risk and try to make money, but not too much money

  1
  .

• With market risk, basically that your belief in the market might turn out to be wrong, we're slightly more caveat emptor

  2
  . You want to yolo into GameStop, you can. Mostly. It's a little paternalistic, like you can't buy cough medicine past expiry date, but it mostly works.

Which in our system means that the minimum amount of fraud possible isn't zero. The optimal number of bank failures isn't zero. The optimal number of bankruptcies isn't zero. We’ve chosen some amount of risk as “acceptable” so that we can continue doing business without spending all our excess capital on preventing something bad from happening.

The level of irreducible risk is the price we pay for not having perfect foresight.

The core thesis of managing the economy is that many of these risks could be reduced, like fraud, but for almost anything else the core idea is that the irreducible risk has to be placed with those who understand it the best.

You should try to make sure people are okay, and only get hurt for the risks they take, but basically it means sometimes people will get hurt and you have to figure out ways to help them

Why risk belongs with those who understand it

Back to the idea of me buying a laptop. Fraud is reasonably well taken care of by legal system. Duping is taken care of by the buyers for the rest of it. Payments risk is taken by the banks who do sophisticated modelling of who's likely to stiff them, including on the buyer and the other bank. But even they don't take any system risk, which is taken by the banks themselves in part by paying insurance, and tamed by various regulatory bodies in response to the regulations the government passes.

So when you hear SVB CEO lobbied to reduce their regulatory burden to not have to keep a lower reserve ratio, it's a little like asking the bank to increase your credit limit. It's under the presumption you're not going to do dumb things and that you're good for it

If you, however, are the types of person who is really good at “reading” the Fed, or probably have a view on the bond market, you might say hey, I'll take the duration risk.

If you're the type of person who really really understands companies, who want to borrow working capital, you might say hey I'll accept credit risk of this chap not paying me back because I know how to read who’s good and who’s bad. You might be a little less trusting than bank of mum and dad and get some collateral, but still. Broadly true.

If you're the type of person who thinks they are fantastic at handling market risk, you might even play options on GME and ride the wave. Or like take the other side of someone who says they don't like some risks and want to hedge it.

If you're the type of person who fully grasps how startups work, and their peccadilloes, and the fact they're equity rich and money poor, you might bank them and do much better.

In other words, you choose what you to do based on what risks you are good at understanding.

Quoting Byrne quoting Lebron:

Agustin Lebron's The Laws of Trading makes the point that since traders are being paid to take risks, they should only take the risks they're being paid to take. If your edge is in predicting oil prices, and you find a way to express a view on oil prices that also involves credit risk (say, buying an oil company's junk bonds) or counterparty risk (doing an over-the-counter bet with a flimsy counterparty), you're basically volunteering your firm's capital for some unpaid charity work on the side. This is frowned on, for good reason.

From the regulatory point of view and from our point of view as consumers it's worth noting this is playing whackamole. If you loosen rules on credit risk (who can borrow) you get problems and though they're different problems to if you loosen rules around how long you can hold an asset before you get returns (interest rate risk), both hit the banks by making them less liquid than you would like them to be, given sufficient depositor need to withdraw cash.

There’s no escaping some risks, only moving it around to those who want to hold it, and who might be the best at understanding it.

The only question for us therefore is, do we understand the risks we’re taking on! The depositors at SVB clearly didn’t. The shareholders did. If the economy goes into a freefall post a financial crisis, a lot of people are going to be hit by risks they didn’t take on, which is what the Fed and others are trying to prevent.

And because this isn’t a clear “opt out” system, but rather an “opt in” system, there’s leakage. You do end up with counterparty risk to counterparties you didn’t know about. That why this is hard!

For instance, since the topic of underexamined black swans is still in the air, I'll leave you with one.

Imagine if tomorrow visa and mastercard mysteriously close up shop, immediately and irrevocably, it will massively increasing demand for people to pay for things in cash. Withdrawals will increase, ATMs will freeze up, and even well capitalised banks will find themselves working overtime to convert bits that determine ones creditworthiness to atoms in the form of a currency note. It will grind the economic system to a halt. If you run a business, this might well sink you. If you wanted to buy bread, you might well end up going without.

Will this hit Walmart? Amazon? Their suppliers? Distributors? Any number of service jobs? Will this sink Citibank? Maybe not. But … the prevalence of this risk too is in the eye of the beholder.

We’re all taking this risk by choosing how we buy things or where we keep our cash. We just think its unlikely.

If you liked this essay, you might also like:

1. People always put their money into the futures they predict

2. In the investing foxhole, there are no atheists