Chapter 20: Uncertainty, Risk, and Private Information

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome to the Deep Dive, your shortcut to being truly well informed.

Today we're jumping into something that, well, it really affects all of us.

Uncertainty, risk,

and this really tricky idea of private information.

I mean, you only have to look around, right?

Think about those huge California wildfires a few years back, 2017 to 2019.

Millions of acres burned, lives lost, and the cost just staggering over $50 billion.

Yeah, and that's not even isolated.

2019 also had what?

Tropical Storm, Imelda, Hurricane Dorian, tornadoes, another $18 billion there.

Plus those Midwest floods adding like $20 billion more in damages, homes, crops, infrastructure.

It really hammers home that uncertainty isn't just theoretical.

It's a very real, very powerful thing.

Absolutely.

And we often make decisions sort of assuming we know what's coming next.

Yeah, exactly.

But these events show there's always a risk of, well, significant loss lurking in that uncertainty.

And it's getting harder to ignore.

Climatologists, the insurance industry, they pretty much agree these extreme weather events are becoming more common, likely linked to climate change.

So how do we deal with all this?

Well, that's what we're diving into today, how individuals and really the whole economy try to manage this.

We'll look at how markets, especially insurance, help people reduce risk.

Okay.

But we also need to talk about the limits.

Sometimes risk is hard to spread out.

And this big issue, when some people know things, others don't, private information.

Right, that asymmetric information thing.

Exactly.

And it can actually stop deals that would benefit everyone from even happening.

Okay, so the plan for today, three main things.

Yeah, basically.

First, why do most of us dislike risk so much?

Second, how does a market economy actually let us, you know, reduce that risk usually by paying for it?

And third, what are the specific headaches caused when private information gets involved?

Sounds good.

Let's get started.

First big question then, why do we pay so much to avoid risk?

I mean, that figure you mentioned, the US insurance industry, over a trillion dollars a year in premiums.

It's huge, isn't it?

It really is.

What's the economic thinking behind spending all that money just for like peace of mind?

Okay, so the core idea here is risk aversion.

It sounds simple, but basically most people prefer a sure thing over a gamble, even if the gamble has the same average payoff.

Okay, a sure thing is better than a maybe.

Makes intuitive sense.

It does.

And the why comes down to how we value money or more accurately, the satisfaction we get from it.

Economists call this utility.

Utility pitch.

Yeah.

So let's imagine a family, the Lees, they have an income of $30 ,000.

Next year, maybe they have zero medical costs, maybe they have a $10 ,000 bill.

Let's say it's a 50 -50 chance.

Okay, 50 % chance of $30 ,000 left, 50 % chance of $20 ,000 left.

Exactly.

Now their expected income, kind of the average outcome, if this happened many times, is halfway between, so $25 ,000.

Right, 0 .5 30k plus 0 .5 20k 25k.

But here's the thing, they never actually get $25 ,000.

They get either 30k or 20k.

That uncertainty about the money outcome, that's financial risk.

Got it.

So how does utility fit in?

Why would they want insurance?

This is where diminishing marginal utility is key.

It's maybe one of the most important concepts here.

It just means that each extra dollar you get brings you less additional happiness or utility than the dollar before it.

Okay, so going from $20 ,000 to $21 ,000 feels like a bigger deal than going from $30 ,000 to $31 ,000.

Precisely.

Think about it.

Losing $10 ,000 when you only have $30 ,000 hurts a lot more than gaining $10 ,000 when you already have $20 ,000 feels good.

The pain of the potential loss outweighs the joy of the potential gain, even if the dollar amounts are the same.

Oh, okay.

So the prospect of ending up with only $20 ,000 is really, really unappealing.

Extremely.

Now imagine an insurance company offers them a policy.

Let's say it's a fair insurance policy, meaning the premium what they pay is exactly equal to the expected loss.

So the expected medical cost was $5 ,000.

Half chance of 10k half a zero dollars.

So a $5 ,000 premium.

Exactly.

If they pay that $5 ,000 premium, they're guaranteed to have $25 ,000 left, 30k income, 5k premium, no matter what their medical bills are.

So they swap uncertainty for certainty.

Yes.

And here's the crucial bit.

Because of diminishing marginal utility, having that certain $25 ,000 actually gives them more overall satisfaction, more expected utility than facing the 50 -50 gamble between $20 ,000 and $30 ,000.

Even though the average income is the same, 25k in both cases.

Even though the expected income is identical, they're essentially paying to avoid the possibility of that low -income $20 outcome.

Because a dollar is just worth more to them when they have less money.

That willingness to pay for certainty is risk aversion.

Yeah, it's really clear.

It's about the utility curve being concave, right?

Yeah.

Getting flatter as income rises.

That's exactly how economists model it.

If you picture that graph, utility climbing but flattening out.

The utility they lose dropping from 25k to 20k is more than the utility they gain rising from 25k to 30k.

Taking the insurance avoids that bigger potential drop in well -being.

Okay, that makes sense.

But surely not everyone feels this way or feels it as strongly.

You're right.

The degree of risk aversion varies a lot.

Two main reasons.

One is just, well, personal preference.

Some people's utility drops off really steeply.

With lower income, they hate risk.

Others have a flatter curve.

They're less bothered.

Like thrill seekers versus, you know, someone who always buys the warranty.

Kind of, yeah.

The other big factor is your starting point, your initial income or wealth.

Losing a thousand dollars is a catastrophe if you're near the poverty line.

Maybe barely noticeable for a billionaire.

Exactly.

The marginal utility of that thousand dollars is vastly different.

And then theoretically you could have someone who is risk neutral completely indifferent to risk.

Their utility curve would just be a straight line.

But that's pretty rare for most meaningful risks.

This also explains the common mistake people make, right?

Judging a decision only after they know the outcome.

Like, I bought insurance but didn't crash, so I wasted money.

Right.

That's hindsight bias.

The decision to buy insurance was rational before the fact because it bought you peace of mind and protected you against a potentially large loss.

Even if that loss didn't end up happening, you paid for the reduction in uncertainty itself.

So people are willing to pay the fair premium, the expected loss.

But you're saying they'll even pay more.

Often, yes.

Because risk feels bad.

Risk averse people are usually willing to pay a premium that's higher than the strict expected loss.

An unfair insurance policy in technical terms.

So the liees might pay, say, six thousand dollars for that insurance, even though the expected claim is only five thousand dollars.

They might, yeah.

If having a certain twenty four thousand dollars, which is thirty K, six K premium, still gives them more utility than the fifty fifty gamble between twenty K and thirty K, they'll do it.

There's a limit, of course.

If the premium was seven thousand dollars, leaving them with twenty three thousand dollars, maybe that's less utility than the gamble so they wouldn't buy it.

But that willingness to pay a bit extra, that's a profit margin for the insurance industry.

That's a big part of it, yes.

It allows the industry to cover its administrative costs and, you know, make a profit while still providing value to risk averse customers.

And we see this in other places too, like warranties on laptops.

Definitely.

Those are basically mini insurance policies.

You pay a bit extra, maybe more than the expected repair cost, to avoid the risk of a big unexpected bill if it breaks.

Increases your expected utility.

OK, so we get why we want to reduce risk.

How do markets actually make that happen?

You mentioned Lloyd's of London.

Right.

A classic example.

Back in the 18th century, shipping was incredibly risky.

Pirates, storms.

Lloyd's started as a coffee house where merchants ship captains and people with capital could meet to basically buy and sell insurance on voyages.

So people were trading risk itself.

Exactly.

And that's the first big principles of how insurance markets work.

Trade and risk.

If you dislike risk more than someone else, you can pay them to take it off your hands.

It's mutually beneficial.

Like selling a house next to a noisy club.

You hate the noise.

Pay someone less sensitive to noise by accepting a lower price to take it.

Perfect analogy.

The second principle is that some risks can actually be made to, well, almost disappear through diversification.

OK, let's unpack trading risk first.

Back to the 18th century merchant.

Ship and cargo worth a thousand pounds.

Maybe a 10 % chance of total loss.

OK, so the expected loss is 10 % of a thousand pounds, which is a hundred pounds.

Right.

But losing the whole one thousand pounds might ruin him.

He's highly risk averse for this particular amount.

So you might happily pay, say, one hundred and fifty pounds for OK, willing to pay more than the expected loss for certainty.

Meanwhile, there's a wealthy investor, maybe someone underwriting at Lloyd's for them.

A thousand pounds is not insignificant, but not ruinous.

They might be willing to take on that 10 % risk of losing a thousand pounds if they get paid, say, one hundred and ten pounds for it.

So more than one hundred and pounds expected loss, giving them an expected profit of ten pounds.

Exactly.

So the merchant wants to pay up to one hundred and fifty pounds.

The investor wants at least one hundred and ten pounds.

There's room for a deal.

Maybe they agree on one hundred and thirty pounds premium.

Both are better off.

The merchant gets peace of mind.

The investor gets a return for using their capital at risk.

And if you imagine lots of merchants and lots of investors.

You get a market precisely.

You get a supply curve for insurance as the premium offered goes up.

More investors are willing to supply insurance coverage and a demand curve as the premium falls.

More merchants want to buy coverage and where they need it.

That's the market price, like one hundred and thirty pounds in our example.

And that leads to an efficient allocation of risk.

The risk ends up being held by those most willing or perhaps best able to bear it in this case, the wealthier investors rather than the potentially bankrupt merchants.

But you mentioned a catch earlier, private information.

Yes, we'll definitely get back to that.

It does complicate this neat picture.

But first, let's talk about the other

diversification.

Making risk disappear.

How does that work?

It works when risks are independent events.

This just means one event happening doesn't make another event any more or less likely.

Think about flipping a coin twice.

The result of the first flip tells you nothing about the second.

OK, or maybe historically the risk of a ship being hit by pirates in the Caribbean versus a typhoon in the Indian Ocean.

Yeah.

Probably independent.

Mostly.

Yeah.

So imagine our ship owner, Joseph Moneypenny, has two ships.

If he sends both to the same risky place, say Barbados, and there's a 10 % chance of disaster there, he has a 10 % chance of losing both ships.

Right.

A 10 % chance of total ruin.

But what if he diversifies, sends one ship to Barbados, 10 % risk, and the other to Calcutta.

Let's say also 10 % risk, but independent risk.

OK.

The chance of losing both ships now is the chance of disaster in Barbados times the chance of disaster in Calcutta.

So 0 .1 times 0 .1.

Only 0 .01 or 1%.

Exactly.

Diversification dramatically reduced the chance of losing everything.

His expected loss is the same overall, but the risk profile is much, much better.

Less chance of the really bad outcome.

So spreading your bets reduces the volatility.

That's a good way to put it.

And you don't need to own multiple ships.

Even small investors can do this by buying shares, small pieces of ownership in many different companies through the stock market.

Right.

Don't put all your eggs in one basket.

It's the same principle.

And the most powerful form is pooling.

Think of a huge health insurance company.

They have millions of policyholders.

Each person's risk of getting sick is largely independent of others.

OK.

By pooling all these independent risks together, the insurance company can predict its total payout very accurately, even though predicting any one person's health is impossible.

The individual uncertainties average out across the large pool.

So diversification and pooling are really powerful tools for managing risk, but there must be limits, right?

You can't diversify away everything.

Absolutely not.

Diversification only works well when risks are independent or at least not strongly related.

It breaks down when events are positively correlated.

Meaning if one bad thing happens, another bad thing becomes more likely.

Exactly.

Think back to wartime.

If French privateers were active, British ships were at higher risk everywhere, not just in one specific sea lane.

The losses were correlated.

And modern examples.

Oh, plenty.

A major hurricane hitting Florida impacts many properties in the region simultaneously.

That's correlated risk for insurers there.

Global weather patterns like El Nino can cause floods or droughts across multiple continents.

Literal events, too.

Like a war affecting oil supplies impacts businesses globally.

Definitely.

And economic downturns, recessions, they tend to hit many companies across different sectors all at once.

That's systemic risk, highly correlated.

So you can diversify away the specific risk of one company doing badly, but not the risk of the whole economy tanking.

Precisely.

There's always an irreducible core of risk that diversification can't eliminate because the underlying events are linked.

Lloyd's of London learned this the hard way again in the corporate liability claims related to asbestos exposure.

Lloyd's had insured against these risks, assuming they were sort of independent company specific issues.

But they weren't.

No, asbestos was used everywhere.

The claims turned out to be highly correlated across many companies and industries.

The scale of the losses was massive and systemic, almost brought Lloyd's down because diversification didn't protect them as much as they thought.

Wow.

OK, so that brings us to the information you call it.

Yes, same idea.

It just means one side in a potential transaction knows something relevant that the other side doesn't.

And this hidden knowledge can really mess things up.

Mess things up how?

Two main ways.

The first is called adverse selection.

This is about private information regarding the state of things like the quality of a product or the riskiness of an individual.

OK, like knowing something about what you're selling that the buyer doesn't.

Exactly.

The classic example is the lemons problem with used cars.

Imagine you're selling your car, you know, if it's secretly a lemon always breaking down or a plum, a really good car.

But the buyer can't easily tell just by looking.

Right.

So what does the buyer do?

They know some cars out there are lemons.

They can't be sure which ones.

So they're only willing to offer a price that reflects the average quality of used cars, maybe leaning towards the lower end because they fear getting stuck with a lemon.

OK, so they offer less than a plum is really worth.

Yes.

And if you know you have a plum, that low average price is insulting.

You might just decide not to sell it at all.

So the good cars get driven out of the market because buyers can't distinguish them from the bad ones and won't pay top dollar.

That's adverse selection.

Potentially beneficial trades.

Someone wants to buy your good car.

You want to sell it.

Don't happen because the buyer lacks information.

And this happens in insurance, too.

Big time, especially health insurance.

Imagine an insurer offers a policy at a standard premium based on the average person's health risk.

OK, who finds that premium most attractive?

People who secretly know they are sicker or higher risk than average.

It's a bargain for them and the healthy people.

They look at the premium and think, wow, that's expensive.

I'm healthy.

I probably won't need much care.

So they might opt out.

So the insurance pool gets skewed towards sicker, costlier people.

Exactly.

Which forces the insurer to raise premiums next year to cover those higher costs.

Which drives away even more healthy people.

Until potentially only the very sickest remain and the premiums become unaffordable and the market collapses.

That's called the adverse selection death spiral.

Yikes.

How do markets even function then?

How do they fight this?

Well, they develop strategies.

One is screening.

Insurers try to gather observable information that might be correlated with the risk.

Like car insurance asking your age, driving record type of car.

Exactly.

They can't know exactly how safe a driver you are.

That's private info.

But statistically, young drivers with speeding tickets, driving sports cars are higher risk.

So they use that observable data to screen and charge higher premiums to higher risk groups.

Seems unfair to the careful young driver, but statistically sound for the insurer.

It is a tricky balance.

Another strategy is signaling.

This is where the party with the private information takes some costly action to credibly reveal it.

Like the used car dealer offering a warranty.

Perfect example.

Offering a warranty is expensive if your cars are lemons, but manageable if they're plums.

So it signals quality to the buyer, allowing the dealer to charge a higher price.

And just building a good reputation over time.

Absolutely.

A long standing business implicitly signals reliability.

You trust they aren't trying to rip you off with hidden problems because they want your repeat business and good word of mouth.

Okay.

So that's adverse selection, hidden characteristics.

What's the other problem?

The other big one is moral hazard.

This isn't about hidden characteristics.

It's about hidden actions.

It's when someone changes their behavior, often taking more risks because someone else is bearing the costs.

Because they're insured, for example.

That's the classic case.

Think about, there was this weird arson epidemic in New York in the late seventies.

Some landlords whose buildings were maybe losing value had them heavily insured.

They had private information about their own actions.

Maybe they just stopped maintaining fire safety systems, or maybe in extreme cases, they even hired someone to, you know, burn the place down because they knew the insurance company would pay out.

Wow.

So the insurance itself created an incentive to be reckless or even destructive.

That's moral hazard.

The fact that your actions aren't fully observable and someone else picks up the tab if things go wrong distorts your incentives.

It doesn't have to be illegal like arson.

It could just be, you know, not bothering to lock your car or driving a bit less carefully because, you know, insurance will cover dings.

Or maybe like a salesperson who gets a flat salary if the boss can't perfectly monitor their effort.

Exactly.

Their effort is private information.

They might slack off because their pay isn't directly tied to their sales effort.

That's why commissions are so common.

They combat moral hazard by aligning the salesperson's incentives with the company's.

So how do insurers fight moral hazard?

They can't watch everyone all the time.

No.

So they use tools like deductibles.

You know, the amount you have to pay out of pocket before the insurance kicks in.

Right.

Like the first $500 of a car repair.

Exactly.

Because you have to pay that first $500 yourself, you still have a financial incentive to be careful to lock your car to avoid minor accidents.

It means the insurance coverage isn't 100 % so you still have some skin in the game.

And deductibles also help with adverse selection maybe.

Like riskier people might choose lower deductibles and pay higher premiums.

That's right.

They can help people sort themselves out based on their own assessment of their risk, which gives the insurer useful information.

There's another example in the business world you mentioned with franchises.

Yeah, it's a great illustration.

Why do companies like McDonald's use a franchise model instead of just hiring managers for all their restaurants?

Good question.

Well, a salaried manager's effort is hard to monitor perfectly.

It's private information.

They might not work quite as hard if they get paid the same regardless, but a franchisee, they've usually invested their own money, their own capital.

Ah, so their own success is directly tied to how hard they work and how they run the place.

Exactly.

The franchise model, by making the operator bear more of the direct financial risk, inherently combats moral hazard and incentivizes hard work.

They try harder.

This is fascinating stuff.

It really shows how information, or the lack of it, shapes everything.

Now, you mentioned a real world case study that ties a lot of this together.

Pure RE Insurance.

Yeah, it's a really compelling story.

Ross Brickmuller founded Pure RE Group of Insurance companies.

They started writing homeowner policies in Florida right after hurricanes Katrina and Rita in 2005.

Wow, talk about jumping into the fire.

Most insurers were running away from Florida then, right?

They absolutely were.

Major national companies were drastically cutting back coverage because the hurricane risk seemed overwhelming and critically highly correlated.

So how did Pure RE succeed where others were failing or fleeing?

Brickmuller had a really smart two -part strategy that directly tackled the problems we've been discussing.

First, he used aggressive screening to deal with potential adverse selections.

Also.

Pure RE decided only to insure high -value homes over $1 million,

and specifically homes that were relatively new, built to stricter codes, had solid construction, good shutters, high -impact windows, etc.

Ah, so they weren't just insuring any expensive home.

They were selecting for homes where the owners had already demonstrably invested in reducing risk,

lower -risk clients within a high -risk area.

Precisely.

They screened out the higher -risk properties within that category.

That was step one.

Step two was tackling the massive correlated risk of hurricanes through diversification and pooling on a huge scale.

But how could a relatively new company diversify away hurricane risk in Florida?

They didn't do it alone.

They bought reinsurance, basically insurance for insurance companies.

They paid huge global reinsurance firms to take on about 75 % of their potential losses from a big chunk of that concentrated Florida risk to companies that operate worldwide.

Exactly.

For those giant global reinsurers, Florida hurricane exposure, while significant, was just one small part of their massive globally diversified portfolio of risks.

They could pool it with earthquake risk in Japan, flood risk in Europe,

liability risk everywhere.

For them, it was a manageable, diversifiable risk.

Pirari used screening to get a better -than -average pool of risks locally,

and then used reinsurance to diversify away the big correlated catastrophe risk globally.

You got it.

It was a brilliant combination of strategies based on sound economic principles.

And did it work?

Amazingly well.

Pirari didn't just survive, it thrived, grew like 30 -40 % per year for a decade, expanded across the US, and was eventually bought for over $3 billion.

Wow.

That's a powerful example of understanding risk, information, and markets.

It really is.

It shows these concepts aren't just academic, they're crucial for navigating the real world.

So, wrapping things up, what have we seen today?

Risk and uncertainty are, well, they're just fundamental parts of economic life.

Our dislike for risk, our risk aversion, comes from diminishing marginal utility losses, hurting more than gains feel good.

Markets have developed amazing ways to cope, trading risk to those more willing to bear it, and making some risk effectively disappear through diversification and pooling.

But it's not perfect.

No, especially because of private information.

Adverse selection, hidden characteristics can kill markets, and moral hazard hidden actions can distort incentives.

Which leads to counter -strategies like screening, signaling, building reputation, and using things like deductibles.

Exactly.

It's this constant interplay between risk, information, and incentives.

So, for you listening, maybe start looking for these dynamics around you.

Why does that warranty exist?

Why does your health insurance have a copay?

Why might a contractor prefer an hourly rate versus a fixed price, or vice versa?

Yeah.

Understanding these ideas gives you a new lens to see how deals are structured, how markets work, and why sometimes they don't work as well as we'd hope.

So, maybe the final thought is this.

The next time you're in a situation with uncertainty, or where one side seems to know more than the other, how might thinking about risk aversion, diversification, adverse selection, and moral hazard change how you approach it?

Something to think about.