Fairly early in my career, I did a brief stint selling insurance. It was part of my training program at the insurer I worked for, meaning I didn't have to earn my keep from it. That was fortunate in several ways: not having the zero- or low-earning risk of a normal sales job[efn_note]Although on the flip side, it meant I couldn't earn big bucks at it either[/efn_note] allowed me to take a more honest look at my customers and whether or not they really needed the insurance I was going to try to sell. Being a consultative seller at heart, that suited me fairly well - there were just some products I couldn't get around to endorsing, which makes it really hard to sell them.

In our consumerist world, the question "do you really need this or that product" has become somewhat anachronistic, even regarding products that aren't traditionally regarded as consumer items.[efn_note]At some point we'll have to take a look at NFTs - see the last link - on Risklantern. If we can find a risk angle.[/efn_note] Insurance isn't there (yet?), so let's take a look at some criteria for decision making.

The amount of insurance an individual takes out is highly based on their personal risk profile, which again is influenced by personality, culture and a host of other factors including cognitive biases. Which is a more complicated way of saying "it depends". But on what? A few ground rules can help.[efn_note]The list was written with personal risk/insurance in mind. Commercial insurance has some extra factors, although most of this will also work. Also, I am just looking at the question of "if", not of "with whom" and how much to pay. Here, questions of added value for service etc become a strong factor.[/efn_note]

• IMO, there are a few absolute essentials, where the benefit is overwhelming compared to the cost. Prime example is personal liability. Even though the risk of becoming personally liable for huge damages is pretty small,[efn_note]At least in most jurisdictions [/efn_note] the cost is also small. The upside is also not just for the large losses - your kids will tend to break something of the neighbors at some point, and the peace of mind to have your insurer take care of it is not so bad.

• While personal liability is a good bet for everybody, other essential coverages will depend on personal circumstances. Good rule: if you will lose your livelihood if you damage xx, get it insured. For me as a seller that meant that, unlike many of my colleagues, I never gave a blanket recommendation for disability insurance, as some occupations are just more at risk than others. A researcher like myself? Can work in a wheelchair in the same job. A carpenter? Not so much. A pianist? Should always get his hands insured.

• Being aware of your own cognitive biases becomes very relevant especially when deciding on insurance. We are all prone to zero-risk bias - if the risk is already low, why not eliminate it altogether? - and on top of that we are bad at rationally working with very small risks.[efn_note]As Daniel Kahneman noted also with very large ones, but you won't be able to buy insurance for those.[/efn_note] The theoretical calculation is pretty simple: if your risk of something happening is 0.01, i.e one in a hundred, with a potential damage of US$ 1m, and the hazard is something you come across every day, then statistically on average you'll have to pay the 1m once every three years and insurance is probably a good bet. If the risk is 0.001, on the other hand, you might think you'd already be better off putting some money on the side every month. Problem is, it would have to be almost 4.5k a month.[efn_note]Assuming a 3% interest rate.[/efn_note] But do you even know the risk? Insurers themselves know the average risk of groups of insured, but not individual risks, and they don't need to - the law large numbers makes sure it works out.

• At the danger of sounding quaint, the potential for moral hazard can also be a consideration. Will taking out this insurance coverage cause me to take more risks than I would on my own? The extra risk is not a bad thing if we are talking about something like the risk of starting a business to support a family. When we are talking about extra risks for hedonistic reasons (extreme sports come to mind), socializing the cost is a different matter.

One thing you should never do when considering insurance is the ex-post reflection of "I never needed the insurance I bought so far, so I don't need it." Turning out lucky doesn't make the purchase wrong per se; like in poker, you bet on the hand you have and its probabilities, but unlike poker your life doesn't allow you to play many hands so that you can see the superior strategy winning. I come from a nation and culture with something we call "Vollkaskomentalität", comprehensive cover mentality.[efn_note]Don't you just love German compound words?[/efn_note] That means that paying for peace of mind is more highly valued, and zero-risk bias is also culturally stronger than for example in the US. As a nation, we probably have too much insurance, but individually? I am pretty sure I personally don't.

Notes:

A few weeks back, this March, the Internet turned 100000 [efn_note]100000 in binary equals 32, for those who don’t want to count zeros.[/efn_note]. At least what we generally call the Internet, i.e. the World Wide Web, which was proposed in 1989 by Tim Berners-Lee at CERN as a radical new way of linking and sharing information.

The Web changed our life in lots of ways. For example, it’s currently hard to imagine how the still ongoing pandemic with its lockdowns and other restrictions would have turned out without the option for many to work, shop and even meet online via the Web. But the Web also opened the the door for whole new groups of risk.

Mark already highlighted some aspects inherent in the social media layer of the Internet. But cyberrisk – the term we coined for the risks of being connected – go much deeper, and it affects all of us with very few exceptions for truly off-the-grid people.

On the surface, cyberrisk issues seem to mainly affect large companies, as highlighted by major data leaks that were reported in the recent past and that hit millions of users. But while it is true that large hacks make better news (and probably generate more fame for the hacker), smaller organizations where data are less well protected, or private citizens, have the same or even higher cyberrisk.[efn_note]The fact that cyberrisk seems to be larger for larger companies is a form of cognitive bias called “availability bias”. It is similar to the perception that flying is more dangerous than driving, even though it is much safer – a plane crash is highly publicized and more people die at once.[/efn_note]

I just had quick look at a study I wrote about this a while back, where we surveyed organizations on their cyberrisk preparedness, with some concerning results. Around half of organizations surveyed had been affected by a cyber incident, most of which had caused real financial damage, but only 42% did regular risk assessments, and a measly 14% had an institutionalized risk management program for cyber. Those numbers were hardly higher for those organization that had actually had a breach. And these are large organizations (with more than US$ 50m revenue) we are taking about. Even though the data are a few years old now, I don’t expect the situation to have progressed all that much.

While actual black hat hacks are great for movies, they are just the tip of the iceberg in terms of cyber-incidents. I haven’t found exact numbers (especially since the dark figure is high) but my guess is that the vast majority of damage caused is less due to a technological break-in and more to modern forms of social engineering. Probably most well known is the advance-fee or 419 scam, which most people will know by the name of “Nigerian prince scam”.[efn_note]The scam neither originated in Nigeria nor are the majority of scammers located there, but the name still stuck.[/efn_note] In this con, people will receive an email with a promise of large sums of money from dubious sources, but to get the transfer started the victim will need to advance a modest amount of money to pay for bribes, fees or whatever. Surprisingly, this works often enough to be a worthwhile endeavor for would-be scammers; I guess there is still a sucker born every minute.[efn_note]As an exercise in reversing gullibility, “419 baiting”, i.e. scamming the scammers, has become something of an Internet sport, with whole websites devoted to it.[/efn_note]

Other more sophisticated scams prey on users’ lack of tech savvy. For example, when opening an innocuous seeming website, the user gets a popup warning of a computer virus infection with the advice to call a “tech support  hotline”. The user is then coaxed into giving the “tech specialist” remote access to their computer to “fix” the issue – bingo!

Whatever the scamming or hacking method, all exploit the connected user’s lack of protection, attention or knowledge, i.e. the lack of cyberrisk awareness. That is surprising, because a few fairly simple principles can keep the vast majority of people reasonably “cyber-safe”:

• Whether computer, tablet or smartphone, never click on any unsolicited link, even when it looks like it’s coming from a reputable source such as your bank or PayPal. Some phishing emails can look quite real, but your bank would never ask you to click on any link they send you and enter your ID/password there.

• The same goes for unsolicited calls. Offers that sound too good to be true ARE too good to be true 99.99% of the time.

• Maintain reasonable password security. That means not having the same password for everything on the Internet, the passwords being hard to guess (the more random the better) and ideally changing them regularly. Considering the number of website where we tend to have accounts nowadays, that sounds like a bit of a chore, but password safes like KeyPass (which I personally have been using for years) make the task relatively straightforward.[efn_note]For me, the passwords I can choose on my own are always random-looking combinations of numbers and upper-/lower-case letters. They aren’t random, though – I have an algorithm that allows me to create an (almost) unique high strength password for each site while still being able to remember, or rather derive, all of them. I still need KeyPass for the accounts that don’t allow my own password.[/efn_note]

• Don’t do anything sensitive on public networks, whether on a computer in an Internet cafe, or via the public Wi-Fi at Starbucks. At a conference a few years back I heard a really interesting talk by Tony Sales, “Britain’s greatest fraudster”, who laid out how easy it is for intruders to listen in to your device on these open channels.

• On social networks, don’t share information that you don’t want a criminal to have about you. Example: posting your adress with an additional “by the way, I am going on vacation tomorrow”. A few years ago people could be excused for believing their data was secure on social networks, but that time is long past. In my personal netiquette I go a lot further: never post anything you wouldn’t want the recruiter at your next job interview to know about you.[efn_note]Stuff you put up on the Internet almost never disappears.[/efn_note]

• Invest some time and/or money in checking out anti-virus software, again for all devices, and then installing and maintaining it. And of course having it run regularly.

None of these measures on their own will reduce cyberrisk to zero – a truly determined attacker will always find a way to get at your data – but in combination it makes you be more of a hassle as a target than most fraudsters are willing to shoulder. Worked for me so far, and I have been online since 1994...[efn_note]Remember those old modems with the screechy noise?

[/efn_note]

Notes:

Christian’s excellent discussion on COVID-19, our recent IBM study "An Injection of Hope" on this subject, and zero-risk bias (elimination of a risk category over reduction in overall risk) highlights for me some unique knock-on effects for COVID-19 in the US.  We have had quite an experience on COVID-19.  Due to a combination of American individualism, leadership vacillation, concentration of risk in nursing homes, and sheer size, America has led the world in COVID-19 related deaths, though at current rates Brazil and India may yet catch the US.

America has had some vaccine hesitancy for some time.  Historical incidents resonate deeply with an America founded on the idea that government has limitations.  Our government has not always been as diligent as it could be in protecting the public from legitimate health risks, as show in the US investigation of thalidomide in the 1960s.  By and large, though, in the past the US federal government has erred on the side of caution.  Thalidomide was removed from vaccines voluntarily as a precaution, despite its efficacy as a vaccine preservative, because it could potentially break down into mercury as a sub-component.  The amount of mercury present for one person was equivalent to eating a can of tuna fish.

That non-zero risk, even though vanishingly small, can go viral in America when paired with bad science like the infamous Wakefield study, purporting a link between MMR vaccines and autism on the basis of a study group of n=12.  Add to that the usual American predilection for celebrity and monetization and you get a lot of people (across the political spectrum, by the way) making poor risk decisions.

That culture does work both ways, however.  While America's hands-off attitudes towards governmental risk management can create huge problems, it also can enable huge solutions.  A number of vaccines were researched and fast-tracked in the US system, making their manufacturers $US billions in stock market capitalization and putting the US in a strong procurement position early.  And the US currently is among the world's leaders in vaccine administration, the only large country to break 50 doses administered per 100 residents as of this writing.  (The United Kingdom, culturally similar though smaller, has had a similar set of issues and results).

Our challenges in the US will be twofold from here.

First, we'll need to get to herd immunity levels of vaccination, which is not a given.  As Christian discussed, estimates of herd immunity rates range from 60% to 85%, and it's questionable whether America will get to that higher number.  There is an effect on risk perception as vaccines roll out, though, and polls indicate that familiarity with the vaccine may well increase US population adoption rates.

The second issue, though, will be to replicate that success throughout the world.  For COVID-19 to be truly managed, island solutions will not be enough.  Roughly 10% of the world has had a single vaccine dose, and rates are climbing.  Still, it will be months if not years for the COVID-19 vaccines to even reach all countries, much less all arms.  We have many, many jabs left to go.

The US has a good track record of subsidizing worldwide adoption in these situations.  Competition with China and "vaccine diplomacy" will likely accelerate interest in reducing worldwide exposure, and the number of vaccines in the pipeline worldwide is encouraging.  Now if we can just get folks to continue to manage their risk until then...

My wife received her first anti-COVID-19 jab two days ago, and by the time this post goes online, I will have just received mine, too. A good time to discuss the risk perceptions around the COVID-19 vaccine; especially since just yesterday, three research colleagues of mine, Jane Cheung, Steve Peterson and David Zaharchuk, released a timely new study on the topic, aptly titled “An injection of hope”. Full disclosure: I’ll be shamelessly stealing a lot of today’s post from that report. (But not everything.)

For the study, they surveyed more than 15k consumers in 9 countries. Questions centered around confidence in the vaccines and in the expected consumer behaviors afterwards. Some interesting, and some disturbing, trends emerged.

Roughly two thirds of respondents believed the vaccines are safe, and about the same percentage that they are effective, with wide variations between the countries. That means that about 1 in 3 consumers don’t trust the available vaccines, their country’s rollout efforts or both. Even in countries where trust in the vaccine per se is high, such as China, still 40% of respondents say they won’t get vaccinated – never mind the U.S. were 55% are at least uncertain whether they will.[1]

That is bad news for returning to “normal” life post-COVID-19. The proportion of people who need to get the jab for herd immunity [2] is unknown.  Estimates cited by the US CDC have ranged from 60% to 85%.  So in all likelihood the level of immunization needed for herd immunity is higher than the level of willingness to receive the vaccine. An additional complication is that not all vaccines are trusted equally. The AstraZeneca vaccine in particular has been suspected in some countries to cause blood clots in rare cases, which has put a halt to use of that vaccine in some jurisdictions.

This may misallocate risk, as the risk of blood clot formation has in no way been near the risk of foregoing the vaccine and potentially contracting COVID-19 and its potential complications.  It's not only that people are at more risk from the disease than any vaccine complications.  Even getting in the car and driving to the vaccination site today was more risky than the vaccination.[3] Still, while 70% of my adopted compatriots in France said they will accept being vaccinated, they also say “but not with AstraZeneca” – making the already fraught rollout hereabouts even more fraught.

The vaccine hesitancy inherent in the AstraZeneca scare is a texbook example of zero-risk bias in action. Unfortunately, any sizeable vaccine hesitancy can derail the whole effort. Not only don’t we know where the herd immunity limit is, we also don’t know whether and how well the vaccines will work against mutated strains of the virus, and with every non-vaccinated person, the risk of mutations increases.

The other reason with the high distrust is so scary is in this chart from the report:

Only after vaccination levels have exceeded 70% will even half of consumer feel confident enough to return to their lifestyle from 2019. We’ll need more than that for the economy to get back to humming again, which means vaccination levels that consumers today say we will never reach.

The report has a lot more interesting things to say about when and how those post-pandemic behaviors will look like... details I won’t go into as I invite you to read the paper.[4]  The one thing I would like to highlight is the generational disparity that shows up in the data. Gen Z respondents in particular [5] expect their behavior to change in different ways than their elders: more social interaction, but not in large-scale venues like sporting events, and also more online shopping. My own interpretation of this is: more direct interaction with friends and family (and local stores) where possible, online speed and convenience where necessary. This mirrors my own experience talking to people in my own locality, with many re-valuing social closeness and community, including frequenting local producers and vendors.

I might be wrong in this interpretation, but it does give hope that the excesses of consumerism and globalization that made a pandemic like the one we are living through possible in the first place can be sufficiently mitigated that the risk of it happening again are lowered.

Notes:

[1] 30% say they won’t get the vaccine, 25% are uncertain

[2] Whatever that means in this case – all available vaccines prevent severe cases, but not necessarily catching the bug or passing it on

[3] I got the AZ jab

[4] Another disclosure: even though this was written by my colleagues, I won’t get any benefits for the shameless plug in this post beside maybe a slap on the back. I just think it’s worth the read.

[5] As outlined in the post linked above, the definition of generations is fluid. The report defines them 18-24 year olds – the lower age boundary being a result of not surveying minors.

A while back, my wife and I were watching Ghost in the Shell – the Americanized live-action version, not the original anime. In this cyberpunk story, the heroine is a crime fighter who has a cybernetic brain (among other artificial body parts), allowing her to interface with various machines in the real  world. Predictably, stuff goes wrong, story ensues, etc.[1] What caught my ear and led to this post was a phrase uttered by the heroine’s boss when talking about the above cybertechnology, its uses and abuses: “There’s always a residual risk.”

Indeed there is, as I’ve written before, and it often has hidden cost that causes market distortion. In fiction, markets are less of an issue; the risk has manifested and provides us with a dystopian future (or alternate present reality) that is the backdrop of the story. I happen to like SF, and find the darker dystopian kind to usually be the most thought-provoking, when the creators tries to get reader/viewer to reflect on the world. Some thoughts, as usual in no particular order.[2]

From a risk perspective, the most interesting tends to be the “a previously underappreciated risk happens and makes things go haywire on a more or less grand scale”, with the protagonists fighting directly to mitigate or prevent the consequences. Some authors have made their career out of these types of stories – Michael Crichton comes to mind as probably the most successful. Space probe returning to earth with a deadly microorganism on board? (Andromeda Strain) Crazy rich guy resurrects dinosaurs in a theme park who turn out to be less than cuddly? (Jurassic park.) Many of Crichton’s novels hinge on human error and hubris as their central plot theme, with fail-safes not deserving their name and the residual risk of technology showing up at the worst possible time. It doesn’t have to be Earth-destroying, either, to be scary. IMO the master of these kinds of plots is Daniel Suarez – Kill Decision (about drones suddenly killing people they weren’t told to) is one of the scariest novels I’ve read.

When the risk actually becomes civilization-destroying, the fight gets moved to a different level of mitigation or prevention and it can be successful or not. Those of us who grew up in the Cold War era remember the very real risk of mutual destruction, and the fiction it spawned. Stanley Kubrik’s Doctor Strangelove with a magnificently crazy Peter Sellers? Unsuccessful prevention. Matthew Broderick in War Games? Successful. [3] And then there are the real dystopias, where the catastrophe has already happened and some kind of time travel is invoked to try to reverse it. Terminator (intelligent machines taking over) or Twelve Monkeys (an engineered virus kills off almost everybody) are probably my most-rewatched examples of this, but there are countless others. The Bunker (another virus that kills the world) is a great graphic novel around the subject. And when time travel technology is involved, that itself can be the risk.

Why do these types of stories work to capture the imagination of consumers? It works because it’s so believable – hubris, greed and simple human error are everywhere. After all, the Titanic was supposed to be unsinkable, and technology has a way of escaping its creators in ways that it wasn’t designed or planned for. Simple and relatively harmless example: tech labs still tend to be dominated by men, with the consequence that gadgets that should work well for everybody tend to work worse for women. Anybody who owns a Smart Home device probably knows what I am talking about; I can’t count the number of times my wife shouted at Alexa in frustration because it failed to understand her command, with me repeating it a moment later and it working.

Any human endeavor at a large scale tends to have unintended consequences, with the fixes to the problems we caused having a cascade of consequences of their own. Example: today, we are able to cure many previously deadly diseases using the fabulous tool of antibiotics. But the overprescription of these drugs is starting to cause serious problems by encouraging bacteria to develop penicillin resistance – something we don’t have an answer to yet, but when we do, I am sure it will have unintended consequences of its own.

As a consequence of living so much longer, overpopulation is an issue. We then have to feed so many people, and the industrialization of agriculture has a host of unintended long-term consequences – top soil depletion, desertification, water shortages – and so on.

The stories basically write themselves.

Notes:

[1] Despite the usual cultural appropriation an okay movie.

[2] I’ll be lumping the written word together with film and TV, along with graphic novels.

[3] Remember those modems?

Mark recently wrote about the US First Amendment (the right to free speech) and I am sure the Second (the right to bear arms) will also be examined from a risk perspective at some point. For today, just something from the annals of deadly accidents, another winner of the infamous Darwin Award who couldn’t quite cope with said risks of said right:

To protect his family jewelry, XXX decided on a DIY approach. Recalling his childhood, when a loose tooth was tied to a doorknob and he was told by Papa, "When I slam this door that tooth will shoot out of your mouth like a bullet!" our man rigged a handgun trigger to pull when the door was opened by unauthorized personnel.

It then seems he forgot about his device, because he triggered the booby trap himself, thus becoming Darwin Award Winner #932!

Which brings me to the question in the title. It turns out that in Spanish, el bobo/la boba is an idiot, and the word started appearing in a similar meaning in the English language a few hundred years ago.[1] In that sense, the term booby trap was originally used for pranks. That usage gradually changed, and nowadays it is a fully technical military term. Way to go for Mr. XXX to revive the original meaning!

Notes:

[1] English is really fascinating that way, absorbing elements from many other languages and thus cultures.

Today’s increasingly toxic online environment has itself become a topic of debate. In the US, both political ends are considering changes to social media. They are looking at protections like Section 230 of the Communications Decency Act, which specifies that liability for social communication rests with individual contributors, not platform providers. They are considering antitrust legislation. And they both decry the effects of “cancel culture”.

Funny thing is, a lot of this does not really appear contributive to societal good. Studies have linked social media to depression, anger, and misinformation. For individual participants, there must be some utility — after all, many participate in it for hours a day. The benefit case is not obvious though in real economic terms.

This author does not use social media much for “social media” — I’ve found closed online communities to be a better choice for me, and had elected to #deletefacebook before it became a thing. I have found that many technically-minded colleagues have made similar choices, as awareness of the marketing-industrial complex and its power through technology tends to be higher in that group.  They've opted out.

Social media does matter economically quite a bit though for a subset of people. They have been generally put in the “influencer” bucket, although some have built professional reputation outside of social media.  Brands directly pay such influencers increasing amounts of marketing spend; estimates range from $5B to $15B in 2021. The media channels themselves pay influencers, with top earners making tens of millions of dollars a year. And direct payment for placement is a lucrative business for some, with Kim Kardashian famously charging $1M a post.

For some participants, social media is a major business. And it’s one that is easy to lose quickly. One wrong tweet can derail your audience, with countless examples in the media. And the platform itself can decide to remove your account access, vaporizing your business overnight.

New models are arising to service such risks.  They have migrated quickly from publication to monetization. For those with large social media bases, these can be viable.  Cancellation for such luminaries is overrated.

For the rest of us, though, we face different risks. For those of us running a business, a hobby, or just a group of friends, the risks come from the influencers themselves. If we attract the wrong sort of attention in the social media machine, it can upend our lives. Moving to an independent column is not an option for us. Our business can be threatened.

Is the market providing remedies for that risk? Not really. We could hunt down the influencer who started throwing bricks at our social media presence and sue. And occasionally, that works. But it’s scattershot, and does nothing to really fix the problem.

A “random sparse” enforcement model for social media influence abuse doesn’t really work. Driving down any major highway will tell you this: speed limits are set with such a model, and they are routinely ignored. The Tri-State Speedway, er, freeway, near my house has a nominal 65mph speed limit and these days if you’re not doing 80, you’re getting passed.

The RIAA tried this model to enforce music copyright and reduce piracy, to much negative press over $200,000 judgments against individuals for minor infringement and little practical effect. Such models don’t apply frequently enough to affect risk perception in their desired way — they just appear capricious and mean-spirited. Just like social media these days.

What does work is reducing the incentive imbalance. Music piracy eventually became a minor issue. Enforcement didn’t drive that, and technical lockdown didn’t drive that. What drove it was a change in business models that changed the underlying benefits and risks. It was far easier for most folks to pay Spotify or Apple Music $10 a month than it was to deal with The Pirate Bay and Bittorrent. That’s what worked.

How will this apply to social media? We shall see.

Much of the furor of today’s news and political world seems like it’s “the other side’s” fault. American Democrats decry the social media misinformation propagated by far-right groups, while American Republicans see an increasingly assertive progressive leftist arm overreaching while laying waste to previous social norms. Both sides wildly disagree on politics and culture. Both sides agree, with some differences, that social media is both a powerful agent for change and a problem when wielded incorrectly. All are fed up.

What is increasingly clear is that such suspicion across the political aisle is problematic. Political disagreement is becoming increasingly personal. It has become commonplace for Americans to openly discuss not hiring individuals who disagree with them politically. As we work on reducing discrimination in our economics, we are re-creating it elsewhere.

The risk dynamics of social media may be a contributor to its effects on political life. Consider the risks involved in posting on social media. There’s no question that there is significant risk, as professional careers are routinely impacted by social media missteps, whether recent or many years back, and people are jailed for intemperate tweets. But what are the risk-related upsides and downsides to posting on social media, particularly with political content, and do they contribute to the dynamic?

Posting a controversial thought on social media clearly has major downside risks. One can tweet on the runway and be fired by the time you land. One can gain a following quickly, but one can as easily lose that following, or be banned by the social media provider with little recourse to regain your lost investment in time and audience building.

By contrast, the risks of “piling on” to a criticism of a post, with sentiment that already has some support, are low and potentially negative. There’s little downside risk to being the tenth poster to condemn a tweet others don’t like, and even less risk in being the 100th.

In fact, the incentives for such posters are to outdo peers in levels of condemnation and vitriol. One rarely earns plaudits for “I agree.” But one can earn more following for comments that increase the snark level or the perceived shaming of the original post.

In social media today, almost all the incentives — both for posters and for social media networks themselves — are aligned with polarization and with vilification of the other side. Add in confirmation bias for the general readership, and you get armed camps of followers who have no incentive to hear contrary views, and every incentive to tear into them.

This creates issues at the societal and at the personal level. Polarization increases politically, making it difficult to run a productive government. And that lack of government effectiveness leads to more social media unrest, and more polarization, even at the local political level. Social media can also cause personal issues. Research indicates that many users of social media become more isolated and depressed, and some users have even been driven to suicide by the effects of social media posts directed at them.

Changing the risk dynamic for social media would seem to be worth looking into. But how to start? We’ll consider that in future posts.

As an American traveling throughout the world for many years, I’m asked frequently about our freewheeling lack of speech restrictions in my home country. How can we allow political vitriol, misleading facts, and hatred to be given a hearing?

These rights are protected by the First Amendment in our Constitution’s Bill of Rights. It reads: “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.”

That “freedom of speech” is not absolute. The general principle the courts have settled on is to limit speech only when there is direct and imminent of public harm caused directly by the speech in question.  One cannot yell “Fire!” in a crowded theatre, for example, without repercussions. There are laws against maliciously publishing lies (libel) or false public speeches (slander) to injure another’s reputation. You cannot make threats to kill someone or bomb government buildings, as these pose a clear and present danger to the public.

But that leaves a lot of room, as is obvious when reading our newspapers, watching our television coverage, or perusing Twitter or Facebook.

Mind you, sometimes freedom of speech is more of a legal right than a practical one.  American courts have consistently held that private publications do not have to provide the same freedoms that the public does.  And even in public, the government does not condone all speech equally.  

We take a very different approach than most other countries.  It is taken for granted elsewhere that some forms of speech may be prohibited. Criticism of the monarchy in Thailand. Holocaust denial in Germany. “Subversion” in Hong Kong, which appears to mean whatever the government in Beijing wants it to mean. “Suspicion of causing harassment, alarm or distress” in the United Kingdom. And those are some of the more open regimes in the world, according to the World Press Freedom Index.

The story in countries like Russia is quite different, as the treatment of Alexei Navalny on his reentry to the country has made clear. He is currently being held on defamation charges, in addition to embezzlement charges already declared invalid by the European Court of Human Rights. With Navalny being a well-known political opponent of Russia’s current leadership, it is obvious that Russia is using speech restrictions to shut down dissent against the government. Political advocacy, of a legitimate candidate, has become hate speech. Protest has become sedition.  Recent events in Myanmar only highlight the problems with this approach.

So let’s look at risk principles and the concept of freedom of speech. Where are the downsides to both sides, and how likely are they?

On the side of restrictive speech laws are an immediate benefit: those individuals injured by the objectionable speech have a less objectionable environment. Much of current "hate speech" law attempts to provide this protection.  Note that hate speech law does not necessary change the underlying prejudices or accusations, but the appearance of support for hateful positions is removed.  In this environment, the risk for making vitriolic speech has increased - a good for individuals.  But the risk of speech laws being misused to stifle dissenting opinion likewise becomes higher - a risk for society.

Many speech restriction laws are well-meaning.  But the risks involved are quite apparent to many Americans who recall the excesses of the Red Scare and McCarthyism. If one type of speech is suppressed today, what kind of political speech will be added to the list tomorrow?  And will the government ultimately be more or less responsible and accountable to its citizens in such an environment?

The WPI Freedom Index cited above, by the way, does not rate America at the top of the list, despite its First Amendment protections.  The reasons are complex, but they boil down to America not always protecting our reporters as well from political interference as we should, accelerated by social media pressure on speech and prosecution of leakers of government information.  In short: the government is leaning on speech it doesn't like.  The cure for that, in this American's view, is more speech, not less of it.  We'll stick with our open speech system, hopefully, with all the chaos it entails, because in the American view, the only safe way to deal with speech you don't like is more and better speech.

Last week, Mark discussed the risks and pitfalls of health care as relating to the US system. A few things went through my mind as I read it; today I’d like to go through them in no particular narrative order.

- It’s very true that health care doesn’t lend itself easily to free market solutions, something that has been described by the late economist Kenneth Arrow in 1963. Not only can’t you decline to participate (well, you can, but it’s not really a good idea), the buyer/patient isn’t even in the position to make the purchase decisions herself except in the most simple cases of “standard” maladies like a flu. The doctor, who can and does make those decisions, has an ethical obligation to put the patient need before market, i.e. price considerations. Add to that health insurance, which adds it’s own layer of costs, inefficiencies, and moral hazard, and it’s easy to see why most countries don’t opt for the US model of a “free” health market.

- The cost vs. outcome discussion needs bit of unpacking into several bullet points. True, the American lifestyle is unhealthy, but  that’s not particularly special among developed nations, and it arguably “works as designed”. Consider this chart: The “obesity epidemic” in the US started right after the aptly named and largely scientific-evidence-free SAD (Standard American Diet) with high sugar and starches, and low fat, was adopted as public health policy guideline. Americans have been following the guidelines, including moving more. Unfortunately, you can’t outrun a bad diet.[1]

- While the obesity rate with all its problems [2] is indicative of the state of health of a nation, despite the popular misconception it’s an outcome, not an input variable. I.e. it’s a consequence of the health care system and of public health policy, not a cause. Sure, it’s individual choices that lead to obesity, but these choices are the sum of the information that is available – see above.

- Life expectancy is not the only outcome that is sub-par in the US despite much higher spending – so are all other relevant ones:

Granted, this chart is a few years old, but I don’t believe the relative order either for costs or outcomes has significantly changed – the US still spends vastly more, with worse outcomes.

- 25% of spending on healthcare is wasteful. Assuming you don’t have a subscription to JAMA (I don’t), this video summarizes the findings.[3]

- Getting back to the topic of life expectancy, infant mortality in the US is also fairly high. You can’t blame that on the infants moving too little or eating too much, and only very indirectly on the parents’ lifestyle.

- I’d also argue with the contention that the US subsidizes other health care systems, at least in the developed world. Even in countries where drug prices are much lower (the linked Scientific American article names the UK as comparison), drug makers still make a profit, albeit not as much as in the US, so the lower prices are not a subsidy in any usual sense of the word. Additionally, many drugs that contribute to high cost (and high profits) are arguably largely unnecessary. My favorite example of this is the current pharma cash cow, statins. The science behind cholesterol lowering is highly dubious anyway. Statins do reduce mortality from heart disease, but they consistently fail to reduce all-cause mortality in all but people with an established history of heart disease. (And all-cause mortality is all anybody should really care about – why lower the risk of dying by falling from a ladder when at the same time you will just die more of everything else?). Add to that the fact that trial data about their efficacy and actual side effects are held under wraps und can’t be verified by independent researchers, and we get a classic case of “trust us, we only want your best - a.k.a. your money”. (Honi soit qui mal y pense.) To complete the link to Risklantern – the risk/benefit relationship of the drug is opaque to the patient, even if she were equipped to and wanted to understand it. They are even opaque to the doctor being pressured to prescribe it. And the “subsidy” – the point where I started – is not from the US to elsewhere, but from the US taxpayer (for Medicare) and the health insurance policyholder (employer/employee) to pharma shareholders.[4]

- The innovation argument is also largely a straw man, see statins. Where it’s really important, drug/treatment development is and will continue to be publicly subsidized, see the BioNTech COVID-19 vaccine. Pretty sure Pfizer will not hurt for partnering with them, either (not sure how much they actually contributed to the research, maybe somebody can comment.) Much more can be said about innovation, patents and regulation, but I’ll leave that for another post.

- Oh, and yes, Mark did largely describe the German health care system. Which arguably is not the best either, although much more affordable and nobody goes bankrupt because she has the misfortune of contracting a severe or rare disease, and not the same job lock risk either. Personally, I think single payer systems work best for something like healthcare where choice is much less important than outcomes and where externalities and other market failures negate the benefits of competition – see first bullet point. Some rationing is unfortunately inevitable, whether you do it based on wealth/income or other criteria (e.g. age, severity, necessity) is a choice society has to make. If I were to design the perfect health care system from a risk perspective I would probably ration based on individual risk choices. Food for thought for another post.

Notes:

[1] One of the authors of the linked article is Tim Noakes, sports physician and one of the world’s leading authorities on long-distance running, who literally tried to outrun a bad diet.

[2] Obesity is simply defined as having a BMI above 30, regardless of body composition.  Dwayne “The Rock” Johnson is obese!

[3] The Indicental Economist, the blog I got the video from, it a great resource on anything health care and health insurance related.

[4] One day I’ll write a post about the invisible hand, free markets, and what Adam Smith actually meant when he wrote all that. If I can find a risk angle. Suffice to say that if we were to attach a generator to his grave, we could create a lot of energy given the speed he must be spinning…

In one of our early posts, Mark talked about the risk of dying, or rather death statistics. For example, in 2010 about 400 people died in the US by falling from a ladder. Want to know why? Because Minnesota has too many lawyers, that’s why. Don’t believe me? Have  a look:

The more lawyers, the more deaths.[1]

Interestingly, even more people (718) died by falling out of their bed. And who’s fault is that? The lawyers' in Puerto Rico, of course. The correlation is almost perfect! [2]

Of course, none of these charts prove anything – I have them from a cool (to a nerd like me) website call Spurious Correlations. The host, Tyler Vigen, takes a bunch of data points, and plays around with the correlations between them, and then allows you to make charts like the ones above. Fits very well into our risk theme – lots of data on various hazards, with truly silly pairings, such as “the more nuclear energy the US produces, the more people drown in a swimming pool.”[3] Good for a bit of fun, check it out!

Note:

[1] The correlation here is 0.81, so pretty high

[2] 0.96

[3] But I did say in the linked post that nuclear energy is uneconomical. Who know what other nefarious things it does…

SpaceX has been the leader in the field in terms of launch frequency and payload deployment, with its Falcon-9 system posting launch reliability rates of 98%. It recently had a minor bobble as booster stage B-1059 failed to land after its sixth flight. Not long ago, space watchers fretted about the ability of a rocket to be reused at all. Now we fret that they are not being reused enough!

The long-term play for SpaceX, however, is their Starship program, its launch system for both large-payload flights to Low Earth Orbit, and for longer-term missions to the Moon, Mars, and beyond. SpaceX has turned reusability into a major asset. Instead of launching one rocket and expecting it to make it all the way to a distant target, launch one to orbit, refuel, and then have much more capacity to perform tasks in space. Refuel can be accomplished by ferrying fuel up to an orbital tanker facility (maybe as simple as “just another Starship”) and running many flights. On current estimate, if SpaceX can work out the orbital fuel transfer capability, a total of 7 Starship flights can position a fully-fueled Starship in orbit, with massive delta-V capability for work in space.

That model will require one thing: extremely low-cost reusable rockets at a scale similar to the Saturn V that launched the Apollo missions. This sounds challenging when you consider the Boeing/NASA Space Launch System, a vehicle of similar capacity to the Saturn V, has spent $18B over several years, including a billion dollars just for the test stand. But SpaceX has prototype SN10 on the stand at their Boca Chica facility now, with several more in production nearby, funded mostly privately.

Estimates for Starship production costs vary, but $100M is probably the upper bound and the cost may be considerably lower. They plan to build 100 of them so they’ll need to be cheap. And at that cost, they can afford an iterative, test-to-destruction program.

Hence where the risk comes in. When your economic costs are so much lower than competitors, and the entire system is designed around volume production and reuse, the risk of a single flight failure becomes much more manageable. SN8 crashed and exploded on landing. SN9 did the same, for different reasons. SN10 may well do the same. The data gained is worth the production loss. The cumulative risk, however, has caught the attention of US regulators. The US Federal Aviation Administration, which holds responsibility for public safety regarding aircraft, are now investigating — and in some cases, delaying — SpaceX’s aggressive testing program.

This all boiled over a bit with delays imposed on the SN9 test by the FAA, in a rather pointed Twitter discussion by SpaceX CEO Elon Musk. American space fans immediately jumped all over the FAA, accusing it of being behind the times, too slow, and in the way of progress.

It’s not that simple. The FAA has guidelines from Congress for regulation of space programs, and Congress is not the most nimble of bodies at the best of times. The guidelines call for risk thresholds to the public in the 1/1,000,000 range, and for clear processes to be in place to manage to that level of risk. Boca Chica is six miles away from tourist destination South Padre Island. Starship stands over 150 feet tall and is fueled with liquid methane, which makes impressive fireballs on impact. Small wonder the FAA wants to review mishaps.

SpaceX has not been spotless in their record. Local government agencies keep the public away when a launch is scheduled, but the responsibility for ensuring a clear range for launch lies with the launch provider. There have been multiple pre-launch range clearance issues, with one passerby driving right past the SN9 prototype fueling up for a static rocket test. And the FAA SN9 delay, it turns out, was due to SpaceX launching SN8 with an incomplete signoff on their failure blast radius analysis, an important detail to get right when calculating risk to the public from giant exploding rocket ships.

Note, however, in Elon’s tweet that he took pains to criticize the regulations, not the regulator. There’s a couple of reasons for that.

Regulators are there to help manage risk, not to completely eliminate it. The FAA knows this and has been quite responsive for a government agency in working with SpaceX. The language used by the FAA tells you something, though. They refer to the SN8 “mishap” because that’s what the regulations call it, even though all parties knew SN8 was unlikely to pull off a landing on the first try. The regulatory guidelines, however, presume that you really don’t want to crash your expensive rocket. They were written in a time where that was not expected at all. So the risk regime needs an update. And, as it happens, one was ordered by the President in 2018 and will be rolled out in March.

But Musk, I suspect, does not really want the FAA to get lost. The FAA helps SpaceX manage risk as well. Rockets that were approved by regulators tend to be presumed to be operating to a reasonable standard and not one of negligence. So if SpaceX makes a mistake and damages property or people, they have some independent review and “best practice” assertions to fall back on. Likewise, insurers are increasingly playing a role in managing risk for the space industry as it matures. Insurers like predictability, and regulatory review and signoff provides that predictability.

Regulators may be slowing down SpaceX a bit. But they are providing important risk management services to all parties as payment for that delay.

One of the questions I get asked by foreign colleagues is how we deal with our "crazy" US health care system.  Our primary-private care mix certainly has unique challenges.

On the plus side, you can get some of the best care and cutting-edge treatments in the world.  I've had relatives survive serious diseases and live many additional years because of that feature.  But you can also end up with  substandard care at extremely high cost, as other relatives who did not have good medical insurance coverage have experienced.  And it is possible to get out-and-out bad care, though our litigation and malpractice insurance systems tend to weed out the worst.

Sometimes the system gives you freebies -- COVID-19 testing and vaccine has been declared free-to-consume by the US Government and compensation is being paid behind the scenes. Sometimes, though, it is far from free.  Medical expense is the #1 cause of personal bankruptcy in the US, with estimates of as high as 60% of all bankruptcies.  Our tax system does not subsidize the system much, as you need expenses above 7% of income just to qualify for any tax relief.

The real challenge though is the random nature of costs.  Costs can range from $1K to $10K and up for the exact same procedures, provided by the exact same doctors and facilities.  The very largest private health plans can negotiate lower rates, and do.  Our US-sponsored Medicare (elderly) and Medicaid (poor) government-provided coverage plans can negotiate (demand) even lower rates.  With this kind of risk pooling, though, hospitals and doctors can, and do, make up their shortfalls by charging extremely high rates to customers outside of those systems.

Add in the on-demand nature of health issues, and a lack of certainty on what procedures or treatments you might need when visiting a doctor or hospital, and the results are somewhat chaotic.  The system feels random.  Those least able to evaluate get the highest bills.  Those least able to plan ahead get the worst rates.

Another problem: job lock.  The US system imposes very high risks on those who switch jobs amid healthcare concerns.  If you lose your job, the US legal framework allows continuation of health insurance coverage for a limited time, then no coverage.  The costs to reestablish health coverage after losing it are quite high - $1000s a year.  Smaller businesses trying to provide health coverage to employees can be impacted by taking on a high-risk individual in a small business health policy.  Buying solo coverage tends to be even more expensive.  This all creates more issues and externalities, and the practical impact is, you're stuck.

America's somewhat "raw capitalism" culture tends to presume that a free market with parties competing for the best business model will optimize results.  The invisible hand will cure all ills.  American politicians on the right tend to talk about healthcare as a free market, but it is not truly free -- you can't decline to participate, or shop around for the best service or value, when you are ill.

Provider costs do show some mixed correlation to quality.  Skeptics tend to note America being #1 in spending but #26 in life expectancy, but those numbers distort the picture.  The American lifestyle does not lend itself to long healthy life.  Our obesity rate is 42%, many of us drive everywhere, and our restaurants and diet tend to kill everyone they touch.  Plus, while our pharmaceuticals industry contributes to our bills, they also indirectly subsidize much of the rest of the world's health treatment.

We probably need to consider a more risk-based approach to regulatory guardrails.  We could specify a base set of coverage and let the market compete on a standard package.  We could provide that package to all fairly, using our tax system to ensure that the poor were covered and the rich pay full freight.   We could build in the ability for private insurers to provide more coverage as an upside to incentivize insurer competition -- our Medicare Advantage program does some of that with Medicare and reports high satisfaction rates.

Funny, I think Christian would tell me I just described the German healthcare system?

One problem for the US adopting such a system is that we can't share these risks and costs across borders.  Should drug rates be uniform across nations?  The US view might be heading that way.  May spur innovation, may set off firestorm.  Health risk is fundamental, should be shared by all, but nations that can't agree on the Kyoto protocol are unlikely to agree on sharing health care costs.  But sharing the costs equitably across nations will be needed, as the US system will eventually top out otherwise.  Either other nations will need to fill the health treatment innovation role, or the US will need to recoup its costs.

Fair warning: even though the hook to risk on this post is the Cold War and its arms race, it will be a bit of a ramble across several subjects. When I took some notes for other posts, more and more things ended up in the foot notes, so I decided that might warrant an own post. So here goes.

Following on my last post on population statistics, let’s start with generations. These are a cohorts of people born in a span of years which share certain characteristics.[1] The first time I read about them I was pretty surprised because the timespans seemed all wrong to me. In the US way of reckoning, I am Gen X, but in Germany I am part of the baby boom cohort. The declining birth rates that are one of the defining characteristics of GenX simply happened about 10 years later in Germany than in the US. This is not really surprising when you consider how differently World War II affected these two countries – baby boom in Germany started when reconstruction had ended and the German “Wirtschaftswunder” (economic miracle) was well under way.

The generation after that is GenY in the US, and Generation Golf in Germany. That wasn’t from the sport, but rather from the Volkwagen Golf, also known in the US as Rabbit, [2], and the generation got its name is because it was that first car that many members of that generation drove. (We used to jokingly call it “the car of the German individualist”.)

My point? Attempting to shoehorn generations into one internationally fitting pattern is understandable if you are a marketer trying to sell internationally (or a scientist looking to make comparisons), but misses one crucial point: generations are shaped by shared cultural experiences due to common events – similar to the symbols, heroes and rituals I talked about earlier in the cultural underpinnings of risk perception.

While the student movements of the US and Germany in the 1960s (just to pick the two countries I know best) had obvious similarities in their roots as peace protest, and in their appearance – similar hippie look, similar music etc. – the binding force in the US was the anti-Vietnam war protest, whereas the German student movement was much more fundamentalist (and leftist) in nature, mainly due to the perception that Nazi sentiment hadn’t been sufficiently eradicated.

A peace movement which binds the Generation Golf was the protest against stationing nuclear weapons in Western Germany in the 80s. Many of our readers will remember the 1984 song “Russians” by Sting, or the 1983 film “The Day After”, both dealing with Cold War risks and its consequences. They touched a nerve internationally. But as with Flower Power, the criticism and protest of the NATO Double-Track Decision to station Pershings in Germany was more fundamental in nature: short of an actual nuclear war, a possible scenario was Germany becoming a conventional battlefield in an armed conflict between NATO and the Warsaw Pact – a situation the US would not have had to contend with. Again, a shared experience that helps shape a generation.

So even though Mark and I are born around the same time and so both theoretically GenX, that doesn’t tell you anything about  our risk perceptions (or any other perceptions). Context is important.

Note:

[1] You can argue whether these characteristics are real or cliché…

[2] Sorry, Americans, but what kind of a stupid car name is that? At least the Beetle looks like one…

Seven years ago, my wife and I adopted a few changes in our lifestyle, one of which was switching to a diet low in carbohydrates. At the time, low-carb was considered a fad by many nutritionists [1] and not very well known among the public. What people did tend to have heard of, though, was the paleo diet, so I just said I was doing that. To which many exclaimed “but why would you want to eat like a caveman? They died at 27!”

Which brings me to one of my pet peeves in the field of risk – population statistics. Even serious newspapers often either get it wrong, or muddy the issue. Consider this statement from an article in the Guardian from 2019 which criticizes the plans to increase the retirement age to 70 in the UK:

In Glasgow, boys born between 2015 and 2017 have a life expectancy of just 73.3 years – meaning under this plan, many would never reach pensionable age.

On the positive side, the author didn’t say that Glaswegians could only look forward to three years of retirement. The statement is still a weak argument for a number of reasons. For one, we are talking about the year 2085 – in the science fiction genre that wouldn’t even be considered near-future SF anymore. Additionally, 73.3 is an average, which on its own doesn’t tell you much.

Let’s have a look at some life expectancy data [2]:

A quick glance at this chart would seem to tell us that Americans die first among the developed nations I selected (I more or less randomly picked the larger countries where our readers seemed likely to be), and the “primitive” peoples (the three colums on the right) live about half as long. Right? Well, yes and no.

The chart shows e0, which is “average life expectancy at birth”. That’s in principle the same number the Guardian article above mentions. But if Glaswegian boys have an e0 of 73.3, and the UK as a whole has 79.5, some people in the UK have to live to over 85 to even out the average. E0 differs by sex, socioeconomic status, region and a host of other factors. It (obviously) goes up with better medical care. It falls with greater inequality (i.e. worse access to medical care). So e0 on it’s own is an interesting indicator, but won’t tell you how long you have to live.

But the cavemen's life was still nasty, brutish and short, right? Their e0 is only half of ours, after all?

Again, not quite. Besides being an average calculated from birth, the chart above also doesn’t tell us several other things: when and why people die, and how long they can expect to live if they reach a certain age.

Before the discovery of pathogens and the effects of hygiene, infant and child mortality was extremely high. In 16th century England, 12% of all children born would die in their first year. (Link) That number is around 1% or less today in developed nations. This drop was the leading cause for the steep increase in e0. Looking at “untouched” hunter-gatherer peoples today, i.e. those without much contact with us or other peoples (the !Kung and Hadza on the chart above), on average 57% of the children survive to age 15.

The next big cause of death for caveman was accidents. Hunting a mammoth is dangerous, never mind the competition from saber-tooth tigers and other fun carnivores. So in a caveman mortality table, we would see another dent around the prime hunter ages of 18-25 (for men). For a cavewoman, the major cause of death was childbirth – similar age, similar dent.[3]

The fact that I talked about mortality tables shows that my background is insurance – a life insurer generally wants to know the probability of somebody dying. In population statistics, you mostly use life expectancy tables. Here’s an excerpt of what that looks like for the US:

The first row at Age 0 is the e0 from above, and the longer you survive, the higher your average life expectancy gets. A (US) woman reaching retirement age today can expect to live another 20 years, i.e. to an age of 85.6. And that’s even though when she was born in 1955, her e0 was somewhat lower than it is for a newborn today.

Back to our hypothetical caveman. Again we can use the modern hunter-gatherers as a template for what their life expectancy table would have looked like:

Even “real” prehistoric caveman, once he got past the mortality bumps, could expect to live to an old age comparable to modern humans.[4] The modal age at death e.g. for hunter-gatherers like the Hadza is 76. That number shows the age at which the most individuals in a population die, i.e. the peak of the adult mortality distribution curve. IMO modal age at death is a better indicator of longevity than an average age, which can be skewed by outlier groups.

These hunter-gatherers generally just die of old age – no diseases of civilization like diabetes, Alzheimers or CVD. That’s why I think living like a caveman – combined with sensible hygiene, perks of medicine and some other modern amenities, will get me the most bang out of my life buck.

Notes:

[1] There are still a few who have missed the bus and think it’s a fad. After 7 years I can confidently say a. it works, b. it is easily sustainable and c. I have never been healthier

[2] Sources: For the countries: data.oec.org -  newest available data (2018 resp. 2017) for the hunter/gatherer/etc: Gurven, Michael and Hillard Kaplan. “Longevity Among Hunter-Gatherers: A Cross-Cultural Examination”. Gurven Lab. 2007. https://gurven.anth.ucsb.edu/

[3] I am aware that I am perpetuating the sexist stereotype man=hunter, women=babies, despite growing archeological evidence that the whole tribe including women and children were involved in the hunt, including the actual killing of game. But that's a different topic outside the current scope of the blog, so right now I'll stick with what we see in modern hunter-gatherers.

[4] The Gurven/Kaplan paper discusses why the prehistoric data, which are based on archeological findings, should be taken with a grain of salt

One area of risk I think is going to get a lot more play in the near future is the impending reiteration of the Space Race.  With the rapid expansion of launch capabilities by several countries and by private enterprise, we are moving into an era where new things are possible.  The US has the current lead on launch count with 44 in 2020, but China, Russia, and the EU all can launch in volume, and the 114 launch attempts worldwide are the most since the end of the Cold War in 1990.

There are two big changes driving this investment.

The first has been the explosion in private satellite deployment, with 5,700 satellites in orbit and Starlink launching another 100+ a month with approval for 12,000. And with the advent of much smaller, more  standardized cubesats, we can launch hundreds of satellites a year with new capabilities in communication and observation.  This has created opportunities for private imaging services, many of which will measure risk much more precisely.  We can image homes, farms, and businesses in near-real-time.  We can see through cloud cover.  We can identify pollution sources on site.  Our visibility into commercial risk factors will continue to increase.

The second has been sheer launch capacity.  Starship, Long March 9, SLS, New Glenn, and Yenisei are all on track to launch 100,000 kg+ to Low Earth Orbit, and today's Falcon, Delta IV, Long March 5B, and Soyuz can all launch at least 20,000 kg at volume.  By contrast, the US Space Shuttle could launch 24,000 kg.  This capacity will make new business models possible as cost-per-kg drops by orders of magnitude.  Space tourism, moviemaking, and reality TV are all in planning stages.  Private space stations can be built.

The space market has a number of sources of risk for customers: delayed launches, failure to launch, destruction of the satellite or human cargo, and simple budget overrun and schedule overrun.  There are many risks to manage.

Up until this point it has all been bespoke engineering.  Space travel is the kind of edge case that is hard for insurers to manage.  Traditional launch has been one-time-use.

The US Space Shuttle pioneered the concept of reuse and cost savings, but the move to a partially reusable craft imposed unacceptable complexity/cost ratio tradeoffs given the technology available in the 1980s.  NASA development processes were highly disciplined and worked hard to mitigate risk inherent in their edge cases.  But these systems are complex.  The potential for human error, multiplied by many of parts and interfaces, simply makes risk inherent. Errors creep in, sometimes embarrassing ones, such as the 1999 destruction of the Mars Climate Orbiter over a failure to convert one figure between metric vs. Imperial units, or the loss of surface probe data in Russia's Venera probe mission to Venus because an ejected camera lens cap got in the way.

Enter SpaceX, funded on a relative shoestring by Elon Musk in 2001 with some of his eBay proceeds.  Bringing agile development processes from software development into space hardware development was their approach.  The many risks of space launch failure would be mitigated by reducing costs by factors of 10 or 100.

SpaceX has taken unconventional approaches.  Their Starship development program was started as a garage-built prototype, with welders hired from water tower builders.  But they iterate at speed and massively.  The program as of this writing is on its 9th Starship prototype.  SN8 launched a 12km test flight that validated much of the design but lost pressure on landing and ended badly.  No matter, when SN9 was ready to launch a month later (even after falling off its assembly pad) and SN10 through SN15 are in the assembly line... and can all be informed by the data from the previous prototypes.

Meanwhile, the NASA/Boeing-led Space Launch System consortium has taken a more traditional approach, but has yet to launch after nine years and $18B.

The iterative approach has other benefits.  SpaceX's first commercial launch platform Falcon 9 proved out low-cost rapid reuse, and other startup launch firms such as RocketLab and other countries including China and Russia are pursuing reuse as well.  Reuse of launch equipment was originally seen as highly risky, with the US military declining to use early versions.  The safety record of SpaceX has been quite good, however, with F9 launch success rates near 100%.

Reuse is now being touted as a "flight-proven" benefit.  Proven reliability can trump precise design and engineering.  The US Air Force has now approved reuse where there is no conflict with individual missions (reuse does cost some fuel and therefore limits range).  And NASA certified the SpaceX human transport Crew Dragon system in part because the system had demonstrated mission success through multiple cargo runs to the International Space Station, thus testing and validating the design through iteration.

And with volume comes the possibility for financial risk offset.  Insurance will be a key driver as this market matures, both to manage the risk of launch and as a customer of these services.  SpaceX President Gwynne Shotwell cites insurance as a driver of market maturation, giving launch customers the means to offset what has become manageable launch risk.  As space becomes less a one-off chancy adventure to a more know commodity, the market will only accelerate.

The big takeaway: iteration and cost reduction can dramatically reduce overall risk.  Keeping your bets "small and fast" tends to help manage risk over the long haul... even as far as Mars.

While we like to think of ourselves as rational beings – and economic theory assumes in its models that we are – the reality of human decision making is quite different. That is not to say that in general humans are stupid or irrational. Rather, there are different “modes” of decision making, which nobel-prize winning behavioral economist Daniel Kahneman [1] calls “fast” and “slow" thinking. Fast or System 1 thinking is intuitive and heuristics-based, whereas slow, System 2 thinking is the rational one. We do most of our decision making in System 1, and it’s good that we do: slow thinking is too, well, slow, for normal, everyday use. Not only that, it takes up much more resources, in time, energy and information necessary. So not deciding everything rationally is a good thing.

The success of the heuristic approach to decision-making depends on correct application of lived experience, and can appear quite “magical” at times. Think of a fireman having a “sixth sense” of the precise moment and location of where the floor of a building will be collapsing – he has heard the warning signs hundreds of times before and without conscious thought puts together the pattern of destruction. Or an experienced GP [2] who can tell with a few glances (and sometimes sniffs) at her patients what's wrong with them.

Issues with the intuitive approach can arise in several ways. For one, it is inherently practical, i.e. tied to things that can be experienced, not abstract concepts and complex issues that often arise in modern world decision making. Secondly, it is prone to systematic errors in judgement, biases that affect our thinking and lead our built-in heuristics astray. These cognitive biases can also affect our slow thinking, but only if we let them.[3] Being aware that we have these biases and what they are will allow us to improve our (slow) decision making.

Psychologists have catalogued more than 100 more or less distinctive biases that affect our thinking and decision-making. Today, I’ll just list the top thirteen that affect our risk taking and risk decisioning. Please be aware that the list itself, and its order, is subjective. Anybody who disagrees may do so in the comments.

1. Confirmation bias. Arguably the most common (and dangerous) bias, confirmation bias causes us to seek out facts that confirm our views and ignore those that contradict it. It’s incredibly hard to shake confirmation bias, and even trained scientists fall prey to it when the data they collect contradicts their favorite theory. As it causes us to – purposefully or subconsciously – leave out a subset of data, decision-making is automatically suboptimal. Echo chambers couldn’t exist without it. Confirmation bias can be seen as a protective mechanism against cognitive dissonance.

2. Survivorship bias. This is a type of sampling error that only looks at data from successful ventures (the survivors) instead of taking both successes and failures into account and analyzing the factors for both. History is full of examples of this bias in action; one of the more famous is when in WW2, the allied command examined where returning planes had been hit by enemy fire and proposed strengthening those areas.

3. Agent detection. In an earlier post about COVID-19 I wrote about Type I errors and how they might have evolved as a persistent trait. Type I errors are not a cognitive bias on their own, rather they happen as a combination of biases; specifically the previous two and agent detection and its two cousins anthropocentric thinking and anthropomorphisms. These describe our tendency to think about everything in human terms – like ascribing rational thought to animals and seeing patterns and intent behind even the most random occurrences. Understanding when noise is random, and when there is a pattern, is important for correctly assessing risks and rewards in decision-making.

4. Anchoring. This is the tendency to rely on one piece of information (usually the first one acquired) above all others when making decisions under uncertainty. Marketing relies on this cognitive bias pretty heavily. Anchoring can go so far that test subjects walk slower after a test where they were “fed” words associated with age or illness.

5. Default effect. When faced with a set of decision options, we will often chose to do nothing, i.e. the default. For instance, countries that make organ donation consent the default from which the individual has to opt out have much more donors that countries where individuals expressly have to opt in. Similar approaches apply to subscription models with automatic vs. explicit renewals.

6. Loss aversion. In utility terms, most people perceive the loss of an object as worse than the previous gain of receiving it. This is related to other effects like the endowment effect (people will demand more to give up something than they are will to pay to acquire it) and the sunk cost effect (e.g. persisting in a project that has no prospective future value). In rational decision making, only future risks and rewards are relevant, everything else is “water under the bridge”.

7. Dunning-Kruger-Effect. Ever read of the survey where 80% of drivers thought they were better than the average? That’s a result of the Dunning-Kruger-Effect: less skilled people tend to overestimate their abilities relative to their peers, whereas highly skilled are the opposite. It’s cousin is the Hard-Easy-Effect: we tend to overestimate our ability to master hard problems, and underestimate the easy ones. In high risk situations, the two can be a potentially deadly combination.

8. Gambler’s fallacy. The erroneous belief that past events influence future events in random settings. “The roulette wheel came up red five times in a row, so black is now due.” The difficulty in recognizing this one in oneself is that sometimes past events, i.e. priors, do matter for future risk calculation and decisions.  Related is the “hot-hand fallacy”: the believe that if somebody had success in a random event they will be more likely to have success again. Both will mess up your risk radar.

9. Egocentric bias. There are several instances of this group of biases, such as the illusion of control (overestimating one’s influence on events), the illlusion of validity (overestimating the accuracity of one’s judgements), or the restraint bias (overestimating one’s ability for restraint when faced with future temptation). We all know people who behave like that, but we ourselves would never fall prey to it, of course we wouldn’t. If you just thought that last sentence, that is called “illusory superiority” due to a “bias blind spot” – all egocentric biases.

10. Neglect of probability. Making decisions under uncertainty is tough, but it’s important to make best estimates of the probabilities of risks and outcomes. This cognitive bias describes the tendency to forego estimating probabilities, i.e. ignore probability altogether.

11. Normalcy bias. This bias is, or can be, a result of cognitive dissonance. A disaster or other bad outcome that has never happened (to oneself) before is ignored, or refused to plan for. COVID-19 denial comes to mind here, although, as discussed, that can have other causes, too.

12. Projection bias. We don’t really understand ourselves over time and how we change. Projection bias is our tendency to overestimate how similar to today we will be in the future regarding our preferences and values, thus leading to decisions that aren’t optimal for our future selves.

13. Zero-risk bias. This bias runs through various of our discussion on this blog, especially regarding risk mitigation and residual risk. It is the tendency to prefer complete elimination of a (sub-) risk, instead of a much greater reduction in overall risk.

As I said, this is just a subjective selection, and there are many more biases, some of which influence risk taking negatively. Depending on future topics, we might revisit or do a deeper dive into one of the above. Stay tuned.

Notes:

[1] If there is one book aspiring and experienced risk takers and decision makers should read, Mark and I agree it’s Kahneman’s “Thinking Fast and Slow”. Link up in the bibliography

[2] For the US readers: primary care physician

[3] It can be argued that being aware of our biases will also improve our fast thinking. Probably true, but that will be much harder.

Christian and I were discussing paying for our childrens' educations in our various systems - Christian is through the gauntlet and has one child through the European system, while Mark has two pre-teens and is planning for the expense in the US system.  The contrasts between those situations are instructive (though depressing for me).

US system is pay-as-you-go, with some substantial financial aid for those less fortunate.  Costs range from free (with scholarship) to $10K a year (for a quality in-state university option) to $70K a year for full-freight education at a highly rated private university.

US college costs have been escalating at 6-7% a year for many years now, well above inflation.  The US has been subsidizing higher education via student loans and grants, while doing nothing to control costs, so college unsurprisingly has climbed in expense due to basic supply and demand economics.  The worst case scenario for my planning works out to roughly $1M total expense.

The real challenge for Americans, though, is that the bills for even those of limited means can be $50k-$100k, which as a relative share of wealth and income is a similar financial hill to climb.  Tuition has in effect become an indentured claim against students' potential future earnings.  The US has exacerbated these issues by adding in for-profit universities and the inability to discharge student loans through bankruptcy.

But the hidden issue is that, for less advantaged students, this represents much higher relative risk.  Cost/benefit studies have consistently shown that the high end of the US university market has a good ROI.  Even though the investment is quite high, the earning potential benefits substantially outweigh the investment.  Hence the mercenary college admissions environment that led to the Stanford scandal.

The cost benefit for middle-of-the-road institutions is far less clear.  There are many colleges charging top-tier tuitions but not providing that top-tier earnings impact.  This makes risk hard to manage for prospective inbound students.  It is easy in the American system to put oneself in deep debt and not be able to find employment afterwards to match.

In this environment, those least able to evaluate and manage the risk -- the less-well off who are not earning a scholarship, often first-generation university attendees of limited means and experience -- are bearing the brunt of bad risk outcomes.

Contrast this with the European system.  Tuitions are much more reasonable relative to average income, with heavy subsidy.  And there are alternative paths for those not wishing to go the university route, with strong vocational education options.  Those options are further subsidized by structured apprenticeship programs where the costs of education are offset by meaningful work and experience built alongside the education.  Potential costs and future income do not get too far out of sync in this setup.  The US could learn from this.

Unfortunately, US policy may not go down this route.  Despite vocational education being brought up in the 2020 campaign, we appear to be going further down the uncontrolled costs road with loan forgiveness being discussed.  This will be popular short-term, but won't do anything to address the underlying risk.  In fact education will become even more unmoored from outcomes, as bad risk decisions will be rewarded by this subsidy while good decisions - either high savings to afford an education, or rational individual choices about the cost vs. benefit balance of tuition - will be left to fend for itself.  The rich get richer, despite well-meaning subsidies.

There seem to be several meanings and uses of the term in insurance and economics, but for the purposes of this post and our blog in general, I’ll stick with a fairly simple definition: Moral hazard is when an agent does not (or not fully) bear the negative consequence of his/her own actions, and takes dubious actions with knowledge of this situations. Modern economies are full of examples of potential moral hazard:

• Insurance coverage per se can create a moral hazard when the insured acts riskier than the insurer originally calculated, in part because they have insurance and know resulting losses will be made whole. In that case, the premium is too low, with the insurer bearing the cost of the risky behavior.

• Corporate limited liability creates moral hazard as the fallout from bad investment decisions often goes way beyond the liable capital. (Bankruptcy protections can have a similar effect.).

• Another source of moral hazard in corporate systems is the principal-agent problem: normally, C-suite and directors act as agents of the company and its owners (the principal) without bearing the full cost of their actions. Corporate governance does not generally hold decision makers accountable for downside effects of decisions, as the constant reporting of C-level "golden parachute" compensation packages indicates.

• Packaging individual debt into debt derivatives (e.g. CDOs) removes the direct link between credit default risk and credit decision and thus creates moral hazard.  Investors are insulated from the downside risk of credit default and such risk rebounds in poorly understood ways, creating surprises that led to the financial crisis of 2008.

• Industry bailouts create moral hazard, as long as they are advertised in advance and/or expected. The same is true for too-big-to-fail companies – gains are privatized whereas losses are socialized.

Moral hazard bear some resemblance to externalities, which also come from a disconnect, in that case between the the producer and bearer of a cost. The main difference is that externalities are market failures because the price of a good or service does not reflect its actual cost and so the market cannot work efficiently. Moral hazard, in contrast, is only a potential market failure – the excess risk the agent takes does not necessarily materialize as a loss.

On the other hand, the examples show that for better or for worse, moral hazard is fairly inherent in our Western capitalistic systems. Insurance has become a vital underpinning of our economy which has allowed not only the wealthy to engage in commerce, but also many a small business owner. Limited liability corporations are less clear cut – even Adam Smith, the patron saint of free markets, expressed his doubts.[1]

Mitigation is possible. Insurers try to mitigate moral hazard via exclusions and loading [2], and of course more and better data. Financial regulation, and legislation like Sarbanes-Oxley, attempt to get a handle on other forms. But ultimately, like with many of the other types of public risk we write about, there needs to be the political will to keep moral hazard in check, and that’s not something that’s on the horizon.

The subject of political moral hazard would fill a book, not a blog, but we hope to shine some light on that over time.

Notes:

[1] Adam Smith’s criticism was of a specific form of corporation, the joint-stock company, but as his criticism was of the principle-agent problem it creates, it is still valid. Our modern form of LLC was legislated more than half a century later in the UK.

[2] Loading is when insurers increase the premium for extra risk factors. E.g. being overweight in health insurance, sky-diving as a hobby in personal accident, renting out to strangers in home insurance. Regulation curbs some forms of loading as discriminatory; whether that’s a good or bad thing is open for debate.

Nuclear energy is not the only issue in public policy and residual risk.  One big public risk issue in the US has been the evolving balance between the insurance industry, government programs, and development along the US Golf Coast.

Big dollars are at stake.  Estimated costs for the largest US hurricane losses, all of which have been in the last 20 years: Katrina at $170B,  Harvey at $131B, Maria at $94B, and Sandy at $74B.  Not a coincidence that all four of these made landfall near major metro areas and all were relatively recent.

These record losses are not necessarily because these hurricanes were stronger than large past hurricanes, or that we were unlucky in where they made landfall.  A main factor is the escalating property values in coastal regions.  The Miami region for example has appreciated 102% over the last ten years and almost 200% since 2000.

In the US system, flood risk is covered by the National Flood Insurance Program, but wind and rain damage are specifically left to the private insurance sector to cover.  Since many insurance prospects are not well-versed in the details of coverage, there are often gaps in coverage/understanding and surprised when a peril materializes, leading to high costs, unaffordable claims remediation, and litigation.  Estimates of the percentage of homes in Houston that were uninsurable flood losses range as high as 70% of affected homes.  Damage sources are often unclear or blended, leading to additional disputes about wind vs. flood coverage and loss cause, which inflates systemic costs even further.

Some public policy is aimed at reducing or mitigating these risks.  Building codes are much stricter in many Gulf coastal regions, mandating hurricane mitigation tools like storm shutters.  Public works like levees, windbreaks, and shoreline management measures can help manage flooding.  These systems are often left to local jurisdictions, however, leading to knock-on effects: one town's levee merely moves the flood water to another town.  As a result, flood modeling and prediction is complex and prone to error.  Catastrophe modeling to predict and manage hurricane risks are compute-intensive, proprietary, and prone to misestimating.

Given this complexity and the generally fractious state of US politics, all of this is managed with little regard for risk.

Insurance can have significant personal costs -- the average homeowner's policy in Florida is over $3000 a year and that does not include flood or windstorm coverage, purchased separately at $1000+ each.  This is a major expense for a lower-income household budget, and these households are often forced by housing costs to live in more flood-prone areas.

The political reaction to this has often been to mandate market rates, either through subsidy or by forcing insurance rates down.   But this runs up against insurance regulatory bodies, managed at the state level in the US, who must also manage insurer solvency to ensure claims will be covered -- and often also run a state-funded insurance entity as an insurer-of-last-resort when insurers are forced to leave the market by a combination of artificially low rates and solvency requirements.  At one point Citizens PIC, Florida's insurer-of-last-resort fund, held 40% of all property risk in the state.  At this level the state risk may become unmanageable, where significant or repeat hurricane losses in a year may threaten the state's finances as a whole.  Well-meaning measures such as a 10% cap on rate hikes in Florida, coupled with large claims payouts and reinsurance hikes, imbalance the system even further.

The real problem is the "hidden cost" of hurricane risk, which inevitably creates moral hazard.  If risk is not effectively priced, it incentivizes building that is not sustainable economically in the long run, as evidenced by comparing photos of Miami Beach now and 10 or 20 years back.  The problems are getting worse, with overbuilding now exposed to land settling / sea rising and associated flood risks... even ground flooding on sunny days in the Miami area.  Public demand to "do something" will only result in throwing good money after bad with stopgap measures like dam projects.  Venice tried that and it has not worked well so far... and that was for a unique tourist draw and UNESCO World Heritage Site.  It might be easier to just move the South Beach clubs.

The ultimate need is for politicians to recognize the need to price risk appropriately.  In some cases that even means stopping rebuilding, which the US is seeing more of in recent years. In the aftermath of Katrina, the very worst flood-risk neighborhoods were not rebuilt.  And support has been increasing for mandating tougher building and reimbursement rules when disaster does strike, to avoid repeating the same mistakes.

Our political systems should make sure those who can shoulder the load of risk management are able to do so.  There may be need to help the poor relocate after hurricanes and other natural disasters, vs. rebuilding homes doomed to further problems.  But we need to do so with a clear view on risk.  There's no reason to keep subsidizing penthouse condos and beach houses in glitzy coastal enclaves.