Hacker News Reader

Comment by dekhn | Hacker News Reader

i have a mini-theory in which any group that allocates resources to existential risks will operate at a disadvantage (in the short term) to groups that do not, thus it's likely that anybody who does this will lose to competitors (businesses, countries) that do not.

FOr example, cloud providers who do not plan for meteor-based regional outages (which occur rarely) believe they can home all their data in a single region. This is much cheaper and will allow the provider a better profit margin, and realistically, not a lot of customers care about their data surviving a meteor taking out US-EAST1.

View on HN

Matt Levine has a great take on this in regards to the airline industry in this pandemic. He talks about all the people upset about all the stock buybacks they did during the boom years and how they should have saved cash for a pandemic.

He runs the basic math, and even if the airlines completely cease to exist, it was still the financially sound decision to return the money to investors. Even if they knew the pandemic was going to happen this year, they made the right decision... and given that this was such a rare event, it would have been extremely foolish to try to maintain a cash cushion for this.

Things this catastrophic are better left to governments for bailouts (which really is just insurance at a societal level) than for companies to save for. Having to save for events this unlikely is incredibly inefficient.

I'm not convinced the premiums paid (taxes paid by corporations) match the risk and payouts given, it's only been 12 years since the last hundreds of billion dollar bailout. But yes, capitalism alone does not work.

Or just let the airlines go bust. New ones will crop up once the situation improves.

Pretty much. When an airline goes out of business, most of the physical aircraft will still be sold to someone unless they fall victim to lack of proper maintenance.

Do you have a link?

https://www.bloomberg.com/news/newsletters/2020-03-17/money-...

Toby Ord talks about this very problem in his book "The Precipice: Existential Risk and the Future of Humanity" [1] -- I totally recommend it if you are into this subject, although I have to admit I am pretty pessimistic after reading it, as it seems that the only way to avoid existential risk is by preparing through international collaboration at a global level. Precisely what I do not think will happen any time soon in the current climate.

[1] - https://www.amazon.com/Precipice-Existential-Risk-Future-Hum...

They cite that book in the article. It seems to be their primary source.

But they do care about power outages, hurricanes, tornadoes (maybe not in US-EAST1), cut fiber lines

The probability any one very rare event is low. Preparing for it helps mitigate risks for many other issues however.

I believe GPS premise is largely correct, although the metaphor used could use a bit of work.

You don't need to worry about a meteor taking out us-east-1 because us-east-1 will take itself out several times over before a meteor ever shows up.

Besides, if the meteor’s big enough, it’ll also take out your customers.

This is a significant issue, any company that set aside resources to protect against a pandemic after the Spanish Flu would have been operating with increased costs for a full 100 years. Yet when the virus hit and companies struggled, I saw plenty of comments saying these companies deserved to fail. I’m all for capitalism and the benefits of market forces, but that’s ludicrous. I’m not saying they should all be bailed out either, companies come and go to be honest, but what we do need to do is ensure our vital interests and the infrastructure we depend on is secure and that’s going to take a pragmatic approach.

The global cost of the virus is astronomical, but protecting ourselves effectively against a new virus doesn’t have to be crippling. The same is true of many other credible risks the article covers. Such preparations could even offer additional useful benefits. It’s just going to take vision and pragmatism I’m not sure our political leaders have in them these days.

>...vital interests and the infrastructure we depend on is secure and that’s going to take a pragmatic approach.

I argue for letting these businesses fail, even if vital, because, it seems, the powers that be have forgotten to take care of them, regulate them, and not get captured by private interests, for decades now. Infrastructure across the the country is literally crumbling. Major dams have failed this year. I fully expect to suffer much more before those with the power to delegate resources where needed will find the will too.

Moral hazard. Why would anyone waste money planning for something that could make the business fail, when they know there will be trillions in bailouts?

We'll solve the problem eventually, I guess, when it becomes clear that there's no money left for bailouts.

Money will never run out. Government will just print more money resulting in inflation and other problems. Most of the currencies are free floating and are not backed up by something tangible like gold.

Oh, sure, money itself may conjured such that it never runs out, but the ability of the government to tap the economy's output is definitely finite.

If you start just conjuring money to bridge the gap, you're basically robbing whoever holds money (or other financial instruments denominated in your currency). These parties then work very hard to stop holding money. When that process is done, you're back where you started, only with less trust, higher transaction costs, and elevated risk burdens depressing output across the economy.

Doesn’t appear to be a problem. Print money to accelerate out of the quagmire seems sensible.

Setting aside money as "pandemic" is a problem. Setting aside money for "emergency" helps reduce impact.

1970 gas shortages 2000 dotcom bubble burst 2008 housing crisis

> 1970 gas shortages 2000 dotcom bubble burst 2008 housing crisis

All missed opportunities to change broken institutional structures and to do things differently.

The ability of money to timeshift resources is magical when it works: I can forego goods and services now, save more money, and then purchase more goods and services later.

However, this has limits. It doesn't work for large scale shortages like the oil shocks of the seventies. If everybody had saved money beforehand, there still wouldn't be more oil available in total. When the oil price doubles, you'll only be able to buy half as much with your savings.

What are the preparations that a company would have needed to do in the last 100 years to protect themselves in the event of a new pandemic? Wouldn't they need a plan to restructure to cut costs or find new revenue, and be able to cover cash flow in the n months it takes to do that restructuring? Is it that crazy to expect that from a large profitable company with long-term goals? Consider that it's common "financial advice" for individuals to be able to cover n months of living expenses in the event of an emergency, and that is often much more infeasible for an individual than it ought to be for a large profitable company.

>any company that set aside resources to protect against a pandemic after the Spanish Flu would have been operating with increased costs for a full 100 years

It's not so much setting aside money for a certain very rare event, it's realizing that there are a LOT of low-probability events that are close to catastrophic and the chance of encountering one of these events is quite high in aggregate even if the chance of a specific event in the pool is low.

> and the chance of encountering one of these events is quite high

Indeed, and there are a lot of overlapping disaster preparedness steps that make sense no matter what the triggering event happened to be, so it makes sense to establish, e.g.: robust command and control infrastructure that doesn't rely on primary infrastructure, including communications and chains of command; stockpiles of strategic durable goods like medical equipment, rations, batteries, etc.

Unfortunately, long term, people tend to abandon them. US had many stockpiles before that it no longer does, because they were all sold off as "unneeded".

the unstated prerequisite is for regulation to stipulate a minimum reserve so that the competitive field remains level, otherwise the dynamics would lead to risk ignorers eventually crowding out the disadvantaged risk mitigators. for business risks, i'd support a reserve requirement of 12 months, which phases in as a fledgling company reaches sustainability.

but, a pandemic is not an idiosyncratic risk of individuals or firms, but rather a broad systemic risk, and as such, should be bourne by society as a whole, not each individual and business separately.

Let me understand this: you want a business to have enough capital to pay its expenses for 12 months on zero revenue?

Holy Jesus; that's massive.

Capital is not free. Typical costs of capital today are about 8% (1). You're proposing, essentially, an 8% tax — not on a firm's profits, but on a firm's expenses. I don't want you to think, "oh, that will push a few firms into failure, rendering their entire business unviable." I want you to think, "it will spare a few firms." I also shudder to think of what happens to firm formation in the face of the steep, steep barriers to entry this raises.

You propose a very, very expensive way to build resilience. You're right, though, that the costs would be borne by all of society. Maybe it's worth it? But it's a hell of an ask, and I'm skeptical.

(1) if you have better figures than http://people.stern.nyu.edu/adamodar/New_Home_Page/datafile/... then I'm all for 'em

it's not that incredible. most going concerns already have the working capital, current assets, and receivables to hold out 3-6 months.

and the cost of capital isn't the right rate to be applying here. a better estimate is the spread between the cost of debt and their short-term securities rate (like the interest rate on current assets), which could even be negative (that is, making money). certain businesses can even be financed via their receivables alone.

damodaran is a good reference. his book was the basis of my finance and valuation classes. but it's dense and not easy to implement without going all in. it really taught me that finance and valuation are arguments, not analytical solutions.

Capitalism doesn't seem to really incentivize "vision and pragmatism" at any level of government these days. Social media and entertainment-journalism have compressed attention cycles so much that planning for next year or next administration might as well be lumped in with planning for the next global catastrophe 100 years from now in terms of payoff and optics.

The "It's the companies fault for not being prepared" line originated as a satirical jab at the "personal responsibility" rhetoric the right has used for decades to systematically dismantle any semblance of a social safety net in the United States.

That's just evolutionary theory applied to groups. Bacteria that allocate resources to existential risks (e.g. antibiotics) operate at a disadvantage to bacteria that do not, except in environments with frequent encounters with antibiotics.

Definitely. Another place the mathematics translates pretty cleanly to is financial leverage.

If you take a loan to expand your business and pull forward profits, or even rent property instead of buying it, you take tail risk but in normal times tend to outcompete.

I think it's an ironclad argument for regulation (and maybe government intervention) in insurance markets, and a good argument in many others -- Even though I'm generally in favour of some leverage.

The right answer may not always be the risk compensated answer. The social gains to be had from ignoring the risk could be worthwhile in the broader picture, even if some groups fail.

Yes, precisely (my background is biophysics). Selection pressure leads to rapid population shifts when one has a favorable mutation. However, you can look at eukaryotes and see that absolute efficiency isn't the metric by which they are being evaluated.

That's a pretty good theory. It even shows up in the start-up world: those companies that invest in good security and proper IT infra have a short term disadvantage compared to the ones that don't, which actually may have the effect that those that don't eventually win out at which point they will either end up with a massive security issue or they get wise and fix things in time. It's not a level playing field and tricks like this can easily make a big difference in outcome.

There's a great SlateStarCodex essay on this called "Studies on Slack", where this tradeoff of investing resources in preparation for low-likelihood catastrophic events is examined.

Here is a link to the full article on LessWrong - https://www.lesswrong.com/posts/GZSzMqr8hAB2dR8pk/studies-on...

Shame you can't link to the original Slate Star Codex copy.

This is basically the case for many government regulations, e.g. bank reserve requirements, building codes, food and safety laws, etc.

So it's a coordination problem. If only there were some sort of Organisation (in this specific example, for Health), whose remit was the whole World instead of particular groups, maybe we'd be able to coordinate to avoid a tragedy of the commons?

Probably one of the reasons why security is at the bottom of everybody's priority list.

This is a good example where good marketing can make up for the disadvantage.

If you explain to your customers what you did to prepare for eventualities, those among them who are risk averse may decide to become customers and pay you a little extra.

The US government could find the money to be fully prepared for pandemic diseases in its couch cushions. The problem is not a lack of resources but rather a lack of operational capability to observe, orient, decide and act.

And even if you do prepare, the preparation might be for nil (or very little) advantage if a government will just bail out your competitors anyway.

that works for a profot seeking institution, but preparing for existential risks doesn't have a disadvantage in national security. it's the profit motive that is short sighted, and you've accidentally mistaken "any group" for "any group organized for the purpose of making profit."

It's compounded because of the interdependance of companies and countires.

Elon musk is a counter example to your theory.

Not quite. Elon's strategies involve finding product/market fit in the short term for technologies that will be critical in the long term. But the companies themselves are on a knifes edge of survival. For example, if a meteor struck tomorrow, neither Tesla or SpaceX would have any advantage over their competitors. And if anything, because he runs his companies so close to the edge, his companies are uniquely susceptible to external catastrophic events.

Reply on news.ycombinator.com