What is Cyber Insurance? (w/ Daniel Kasper)
4 Video Views·Oct 3, 2024
Every business, even if they're unknown,
and even if they only have a website,
faces cyber exposure.
So a lot of it's basically a misconception
that cyber, there's nothing to get here from me,
says a small business owner with five people
that only has a computer
or maybe a point-of-sale cashier system.
But it's not true.
It's gonna hit you as well.
In today's conversation,
we're gonna be talking all about cyber insurance
with Daniel Kasper,
who specializes in the field of cyber risk
and cyber insurance.
While it's not a pure play
in terms of computing technology,
cyber insurance is almost something newer
and no less technical
with important financial modeling and math considerations
in addition to traditional cybersecurity.
He got involved in cybersecurity
working as an intern for KPMG Germany
and has worked at a managing general agent,
which is a type of wholesale insurance broker
focused solely on cybersecurity.
He has both a bachelor's and master's in economics
and is currently pursuing a PhD in cyber risk quantification
and its macroeconomic implications.
He's the principal behind the website, cybereconomics.com,
which aims to bring some of the research and knowledge
about quantitative cyber risk management
out of academia and into the open.
Without further ado, Daniel, welcome to the show.
Thank you for having me.
A lot of people have heard of health insurance,
car insurance, and words like actuary, broker, agent,
just break it down.
What goes on behind an insurance plan or a structure?
Like, explain to me like I'm five.
So usually the split you do in insurance
is between life and health insurance on the one hand
and property and casualty on the other hand.
Now, there's certainly a difference in modeling
and requirements for capital,
and cyber is a property and casualty insurance type.
So within property and casualty,
for example, your car insurance,
there are also more or less standardized insurance types.
Again, car or building insurance,
you have like hundreds of years of lost histories.
People are very good because they have such rich
and structured data to be able to price
and to come up with sensible coverage options for everybody.
In the insurance business,
you basically have primary insurers and reinsurers,
and the primary insurer is basically the one you go to,
so they're also working together with brokers.
And depending if you're a private person
or a small business, for example,
you usually have one broker you go to,
and they help you to cover your losses
against unforeseen incidents.
Cyber is now a new type of insurance,
and it covers you against several elements,
but mainly against malicious third parties.
So cyber insurance, you would say,
is a form of property insurance then.
Yes, it's property insurance and also liability insurance.
What is cyber insurance not?
So we kind of, you talked about what it is,
but maybe shed some light on some misconceptions
of people might think of, you know,
what isn't cyber insurance?
So the term cyber insurance context,
it usually refers to a man-made risk.
Now, this is the big distinction
that a lot of layman's don't make.
They think it's a matrix type of thing
where you have some hackers or mistakes and errors
at the root of cyber insurance.
And sometimes errors and omissions, for example,
are included in the cyber policy.
Usually this is a different insurance type,
but cyber usually refers to hackers and ransomware.
Now, it's a small level.
For example, take a retail store
or something that's not usually thought of
as a tech fiend business.
So, you know, your mom and pop store down the road,
10 employees, they're selling maybe groceries
or like antiques or something like that.
Now, you wouldn't think that these kinds of businesses
actually need a cyber insurance
and that they don't have this exposure
towards cyber.
But this is, well, already misconception.
And if not now, it will become in the next years.
What are the main selling points or value propositions
for having cyber insurance?
Because on the first thought, people might say,
oh, well, obviously I want to be insured
against hackers and ransomware.
But then they might think,
I don't have that much money for my business.
Or why would a larger company like want something like that?
Starting with data privacy laws,
everybody is exposed to cyber liability to some extent.
So if this antique shop, for example,
because it gets hacked or maybe also of an error
or something like that,
leaks its customer database
or other personal sensitive information,
then they are looking at fines.
On the other hand, we have a ransomware wave,
which especially now during COVID-19
accelerated incredibly.
So we are looking at three or 400% more attacks
with ransomware in the last eight months.
And basically even these small non-tech businesses
also have exposure towards cyber
and they can get coverage
to what's usually covered under cyber insurance.
So if this applies to mom and pop stores
and non-tech businesses,
then of course it also applies to medium sized companies.
And then it also basically goes to large companies
who at the very least,
when they come to be indexed on a stock exchange
and start to face harsh regulatory requirements
as well as punishment as they are not fulfilled.
So the main proposition is for a small store,
look guys, you don't even have a dedicated IT department.
You don't even know how to handle your own IT
in many cases.
Are you really aware and know what's gonna happen
when a breach or something like a hack attacks you?
And the more you go from medium to large,
it is okay guys, everything you have is sensitive.
If one of these databases is set up wrongly
or you get hacked
or maybe there's some very sophisticated stuff.
By the way, I should mention,
it's not just only external parties,
but also internal parties.
There's Gruntland employees, for example,
that are the perpetrators in cyber crime.
And if you don't have that,
then cyber insurance is a very elegant way
to not only cover yourself against the financial hardships
that might face you,
but also to get competent people
that basically do nothing else via the insurance company
that helps you in an incident case.
That's really interesting where you said
kind of everybody is going to start to need cyber insurance
because you look at businesses,
if you are in a brick and mortar store,
you have real estate property,
you're gonna have some kind of fire insurance,
flood insurance,
something to insure against natural disasters.
And when you look at the occurrences of natural disasters,
they're actually much fewer than something
like getting hacked or these man-made cyber occurrences.
And especially even if you don't get hacked,
what if your shop is used as a launch point
to pivot as a proxy to attack another company?
And now you might be on the hook
because you didn't secure your own systems
and you were involved in that.
Exactly.
This is the primary split
that you also do in cyber insurance coverage.
You do it on first party damages.
So basically the damages you suffer as the insured party
and liability and third party claims.
This is basically when your systems get used
to attack somebody else
or a worm or a virus gets spread through your email list
to your customers or employees or whatsoever.
What do you think is hindering
the more mainstream adoption of cyber insurance?
Because people kind of are talking about it
but you don't see it as standardized as other forms.
On the demand side,
so basically the risk owners, the businesses themselves,
they're looking at cyber,
yes, this could happen to me,
but they usually don't believe it
until it has happened to them
or somebody in their immediate circle.
There's this meme of SpongeBob, for example,
where they go,
this is the cyber security budget before something happens.
And this is the one where after it happens
and it's like a wallet full of money getting opened.
Yeah, let's spend everything.
But this sadly is exactly how it is currently.
So the lack of awareness
and the lack of the abundance of invincibility,
thinking this can't happen to me,
nothing can happen to us.
And especially a misconception of what a hacker wants.
So you always have this kind of hoodie guy,
the stereotypical Mr. Robot
or somebody plugged in in the matrix.
Most cyber crime doesn't happen as targeted actions.
Now, these are things that are usually
those advanced persistent threats
are reserved for high value targets
or special kind of thing.
But every business, even if they're unknown
and even if they only have a website,
faces cyber exposure,
be it through ransomware,
which in most cases almost exclusively spreads untargeted,
but also being hit by misconfigurations of your website
and stuff like this.
So a lot of it's basically a misconception that cyber,
you know, there's nothing to get here from me,
says a small business owner with five people
that only has a computer
or maybe a point of sale cashier system,
but it's not true.
It's gonna hit you as well.
Says slowly and gradually change in the last three years.
And now in the COVID year 2020,
we see a huge spike,
not only in ransomware attacks,
which by some estimates increased by 400 to 500% now,
but also in the awareness that's basically associated
with this ever increasing hacks.
And also the frequency that you see in the media
that are covering those high profile cases
where the whole company doesn't work anymore
for months sometimes
because the ransomware has locked their computers.
That's a great point you make with, you know,
the hacking risks and the cyber risks.
It's not like these hit teams of specialized ninjas
that are come and steal things or do things.
It's more like a cat burglary, you know,
just vandalism and pickpockets
and just happening generically.
They're very opportunistic.
And automatic, like an automatic opportunism
basically through most systems.
Okay, so what are some of the factors
that go into a cyber insurance plan?
On the one hand, you have the coverage elements.
So basically what you want to cover under it
or what you don't cover under it.
Again, you split it up between first party
and third party coverage.
Now for first party coverage,
the usually elements are the cost for forensic
and cyber security services.
You have legal firms that are often involved
when you have a data breach.
Sometimes they come in for the client
to basically act as a fiduciary
also against the insurance company.
You have business interruption,
which is one of the most treacherous of all the cyber risks.
That is basically through a cyber attack,
I'm not able to run my business because my IT is down
or maybe even the feds come in
and shut down the whole operation
because I don't know, everything was leaked.
On the third party side,
you usually have defense costs
against not justified claims from a third party.
So when somebody sues you and they say,
okay, we got our personal information hacked here
through your system or through your mistake,
you have media liability, which often is also included,
which basically means somebody hacks your Twitter account
or as in your case, your YouTube account.
These reputational damages are usually also included
in the first party coverage
for the insured party themselves.
And the other big part
in the third party liability coverage are fines.
Now, the US has a long history of quite strong regulations
concerning data privacy.
For example, you have the HIPAA
that concerns health-related data.
So somebody like a hospital, for example, leaks data,
then there's a huge fine usually associated with that.
And on the side of normal consumers,
you have, for example, in 2003 already,
there was a Data Privacy Act in California.
So if through a hack,
you have costs that arise from such an incident,
then these are usually also covered
in the third party coverage of a cyber insurance policy.
Then it comes down to the hard financials.
The one hand is the coverage,
so basically how much is the given policy
pay out as a maximum.
Then you have to deductible,
which is the amount of money that you have to pay
for any given damage
before the insurance policy starts paying out.
And then usually you have associated
with all of these sublimits.
So for example, media liability might be 1 million,
the whole policy might be 10 million,
and the costs that are covered for cyber forensics,
for example, might be 5 million.
So these are basically the main factors
on a cyber insurance policy.
That's interesting,
because it looks like the structure
behind a cyber insurance policy
isn't very much different
than something like health or car.
It's very, it's kind of inherited those qualities,
such as if you're driving a car and you wreck your car,
those first party damage claims,
you might be covered like the repair costs
at the auto store.
So now the people servicing you
or the lawyers involved
or the doctors and medical professionals involved
will get paid out or you'll get covered
because the insurance company
is paying those service providers.
And now if you wreck somebody else's car
or injure someone, I don't know,
politician or someone who's well-known
and you're drunk driving,
there would be liability, media, reputational stuff there.
So we talked about kind of the pricing structure
where you pay your premiums to the insurer
and for a payout in case you get hacked,
and there's a deductible that you have to co-pay into
before you get your payout.
So how much does a coverage or a plan actually cost
like for maybe a small, medium, large,
a hundred, a thousand, 10,000 employee type of firm?
I mean, usually employees
is one of important factor in pricing,
but the most determining one is the coverage amount.
So if you're looking for a million in coverage,
if you have 100, 1,000 or 10,000 employees
might be a difference of, let's say 20%.
So 10,000 employee firm pays 20% more
than what a 100 employee firm pays.
But if you increase the coverage
from 1 million to 2 million, for example,
then the price might double or go by 50%.
So rather than looking at employees,
it makes more sense to look at coverage
to give you a broad sense of what it costs.
And for a coverage of 1 million,
and let's say deductible of 50,000,
they're probably looking at annual premiums
of about one to $1,500.
And if they want to increase that again,
then 2 million, for example, in coverage,
then they might be looking at 2,000 to 2,500.
The important metric here is basically
the relationship between coverage and the price.
And again, this is way more sensitive
to what's an increase in coverage than an employee size.
Now for large firms, and this is effective for all firms,
basically which industry you're in.
Now hyper, for example, for hospitals,
there you're looking at way higher premiums
usually than in a brick and mortar store.
So for hyper, for example, if you have a 10 million coverage
with a deductible of maybe 500,000,
you're looking at annual premiums
of maybe a quarter of a million dollars,
again, depending on several factors as well.
But this is basically what you're paying.
A small business pays like 0.1, 0.2,
maybe 0.5% of the coverage
someone wants in cyber insurance.
And something like a hospital
that's looking for a larger coverage,
like 10 million, is paying 2.5% of that as annual premiums.
How do you quantify cyber risk
in financial and economic terms,
such as getting hit by ransomware,
or getting hacked?
For a coffee shop, it's very different
than maybe for a hospital.
The first thing you mentioned is also very fitting already,
and this is the split between economic damages
and insurable losses from that.
There are some types of losses
basically associated with every cyber incident,
which are not coverable,
or not covered at least currently
by cyber insurance policies.
So you have the split between what's insurable
and what is the total damages.
Then the split between a coffee shop, again,
as a maybe brick-and-mortar kind of business,
versus something like a hospital,
which has a lot of private data,
and again, for example,
covered under the Healthcare Data Privacy Act.
Say under HIPAA, for example,
where you face definitely harsher penalties
than the coffee shop.
You have to say that the modeling between cyber
or for cyber damage is still very nascent.
So while insurers have taken on
a tremendous amount of risk already,
in the billions of dollars in premiums written,
and in a bad year,
so usually express that in the actuarial sciences
as a one out of 100 event.
So in one out of 100 cases,
what's the loss that we're facing,
for example, in 2021,
given my cyber portfolio of whatever,
a hundred million in premiums?
We're looking at like 50 billion or something.
So the whole cyber insurance industry
could be looking at 50 billions of damages next year,
if it's a one out of 100.
The question really is, okay, how do we model this?
And let's say a straightforward approach
is like looking, okay,
what are the cyber damages,
cyber incidents of the past?
What of that can we attribute in terms of damages?
So can we quantify what was lost?
And then also in the next step,
what of that would have been insured
or would have been insurable?
Now this fails to some extent.
There are definitely data sets out there
and there are vendors that offer data sets for that.
And this is basically the baseline
that we had in the market for maybe five years now.
One of the key characteristics of cyber
is that the threat landscape is ever evolving.
So even if we had the perfect estimation for 2019,
for example, for any given company
or industry sector or like the whole world,
we not have like the same predictive value
as for car insurance,
where if not autonomous driving comes into play,
we basically can look at 50 years of loss history.
And we also know that the 51st year
would have very similar losses to the period before that.
So these are basically the challenges.
And there's certainly somewhere
it's very similar actually to pandemics.
When it comes to ransomware attack,
it has more characteristics to a pandemic
than for example, to car accidents,
because the spread and also the recovers and so on
is more similar to how COVID for example,
spreads than how a car accident spreads.
We try to model cyber as every other line of insurance.
This is not as potent here
because the factors that I mentioned,
and this is basically ongoing research,
which is of great interest for academics
as well as to the industry sector.
And this basically includes the primary insurers,
the reinsurers, capital market,
everybody who is a stakeholder into this,
including for example, cyber security firms
who come in to remedy the claims.
Everybody is looking at how to model these damages
and even more importantly,
how to model the correlation between damages.
So just to recap, let's bring it back to cars.
You have 50 years, maybe even 80 years worth of history
of car accidents and the amount of damages
that each of those have incurred everything
from the little dings or fender benders
to like total chaotic multi-chain highway disasters.
You have a amplitude in terms of the price differences
and the cost for insurance.
We could call that a one sigma or two sigma events
in terms of the standard deviation.
And we know that in the entire existence
of the car industry,
there's only been maybe three sigma
or four sigmas in terms of damage.
Kind of like if you go to a hiking park
or next to a river,
they mark the water elevations throughout history.
And so you have your 10 year flood, 50 year flood,
100 year flood.
And because we don't have a very much history
in terms of cyber and the data,
you could have a 12 sigma event.
Maybe that's wanna cry
that just completely destroys your entire model.
And now your insurers are out of money
because they didn't account for that.
Not only out of money,
they're basically broke in some like very bad scenarios.
And the requirements for insurance companies
basically don't allow you to take on that much risk
when there's so much uncertainty involved there.
The point you made about the floodings
and the peaks of the floodings
is a very famous problem in actuarial sciences.
Let's say we have 10 years of loss data
or in this case, flooding data.
How can we predict a 100 year event basically?
How can we get from a sample of 10 years
to what's the worst case out of 100 years?
Basically having to take into account 90 years
for which we don't have data.
And yeah, this is basically the problem
that we have right now.
So what are some broad categories of cyber risks out there?
Maybe ones that are actually insurable
and ones that are not.
One of the vectors that is insurable
and probably most dangerous in 2020 is ransomware.
And for ransomware,
what is covered is not only the cleaning of the system.
So basically getting rid of the ransomware
and unlocking the systems,
but also possible payment of ransomware,
usually via Bitcoin or another cryptocurrency.
And even the negotiation process,
which as we have seen,
is to be honest, quite ridiculous sometimes.
Where basically the insured party
or the insurer or the third party
that's tasked with doing the negotiations
is basically like in a chat interface.
They're like wiggling for thousands of dollars or something.
The hackers say, okay, we need 30K or something
to unlock your systems.
And then they go, yeah, look, we can only do for 10.
And then they go like, okay, do 25 and so on.
So it's like a silent auction sometimes.
And all these threats are usually covered
and all these costs that are incurred by that
are covered by a cyber insurance policy.
When I say usual, it means like 95% plus of policies.
But again, in 2020, this is the most dangerous vector.
Also the reason why to get cyber insurance
in the first place.
And besides ransomware on the first party side,
we also have the costs that are associated
when a hack occurs.
So looking basically for the leak that was possible,
an unpatched server or something like this,
maybe a Citrix client that wasn't properly configured.
And basically the reinstatement of systems
back to whatever the normal level is.
And usually also a small forensic analysis
to see if this was the only leak.
So to make sure, okay, we have patched it
and nobody can enter.
And then what might not be as obvious
is the massive legal costs
that are usually associated with cyber risk
on first party side, as well as on third party side.
You can see, for example, if you have stakeholders
so if it's a stock listed exchange,
exchange listed company that you're looking at,
of course they have to do a lot of compliance
in the aftermath of a cyber attack
to not only inform the stakeholders and shareholders,
but also to make sure that it's properly fixed
unless you want to face huge liability
from shareholders going forward.
And on third party side, again, liability.
So every lawsuit basically gets involved
and also the media liability
when you're basically damaging the reputation
of somebody else through a hack.
And on the stuff that's not insurable,
and here we should say what's not usually insurable.
So there's some policies where you can buy it
as an endorsement basically.
So you pay a little bit more
and then you get these inclusions.
On the exclusion side,
things that are not covered under cyber,
a war exclusion in many cyber policies.
So this came in at NotPetya
because the official line now of the United States
and the United Kingdom that it was Russian hackers
that were responsible.
And NotPetya was mainly used on Ukraine servers.
So the theory is that Russia used it against Ukraine
to weaken their cyber security and whole economy.
And as it was like a warlike kind of weapon,
these are usually stuff that's not covered
under affirmative cyber insurance policies.
And this, by the way, is a point of ongoing litigation.
There's a very famous case
in the cyber insurance law sector between Mondelez.
They are the producers of Oreo and Toblerone.
You know, huge multinational company.
And they're suing the insurer Zurich
to the tones of 100 million in a policy
that's in Mondelez's eyes
should have been triggered by NotPetya.
And Zurich claims this warlike exclusion,
basically saying, no, we're not going to pay
because this was like a warlike attack
and therefore we will not pay out this 100 million.
Besides this exclusion itself,
usually what's also not covered,
what is usually available in other lines of insurance
is intellectual property.
So it's very hard to quantify
how a loss of intellectual property directly hits you.
But again, in other standalone insurance types,
this is then insurable,
but you're out of the box in cyber insurance.
So essentially in a cyber insurance plan,
there's certain factors already built into the model
and things that you can't include or account for,
such as getting a bomb dropped on your car
beat by a foreign actor, that is not insurable.
And something like the intellectual property,
all of the losses would have incurred
or all of the money that you would have made
if you didn't lose your IP,
there's no real way to determine that.
So that's also not insurable as well.
So it's kind of these trickle on knock on effects.
By the way, you find this ripple effect,
this interdependencies everywhere in the cyber value chain,
be it in coverage, be it in potential losses,
be it in actual losses,
be it in correlation between damages,
be it on the modeling side,
be it how the insurance sector
between primary and reinsurance is set up.
So dependencies are basically everywhere
when it comes to cyber and really looking at the risk.
So let's say I'm McDonald's, right?
I'm some big multinational company
and I want me some cyber insurance.
Walk me through the process of what I need to do.
So as McDonald's, you're a huge player,
you have a huge exposure to cyber
and you will likely not get as much coverage as you want.
From current numbers,
basically the biggest cyber insurance coverages
that are out there are between a billion and 2 billion.
Now at these sums, again, where you have these huge exposures
as huge companies,
you're not going to your local agent or broker.
These are usually specialty insurance products.
And the biggest marketplace for that is Lloyd's in London,
which has about 10% of, I think,
the global insurance premiums overall
and is especially focused on these risks
that are not insurable anywhere else,
be it because it's hard to price,
for example, like the voice of a singer
or the legs of a football player,
which by the way, all happened in London,
or be it because the coverage amount is not large enough
so that a normal insurance company
would be comfortable taking it on.
So McDonald's probably directly or indirectly goes to Lloyd's
and says, okay, we have such and such exposure to cyber
and we want to buy a billion in coverage.
Then you go through all the coverage elements
that you want to have covered.
Now, in the case of McDonald's,
you have a lot of employees on your payroll.
Also, you know, I would imagine a high turnover.
So you're very liable against third-party losses there.
And data privacy is probably,
is most likely the most important risk
that you want to have covered.
At Lloyd's, you have the so-called syndicates,
which are like more or less small insurance companies,
but they're usually all sitting
in the Lloyd's building in London.
So probably what's going to happen there
is that you have like 10 syndicates
and they're going to slice it up.
And they go like, okay, I take 5% of this
and syndicate B is taking 10% of this
and syndicate C is taking 20% of this.
And again, this will be not a straightforward process.
So it's not as easy as getting car insurance
as this, besides the audit,
you have a lot of negotiations as well,
not only in the price you pay, the premiums,
but also in what is covered and what is not covered.
I would imagine within like two to three months
for McDonald's.
Okay, so let's scale it down just a little bit.
What if it's just some local, very local credit union
or some very tiny boutique bank
with maybe under a hundred million in assets,
maybe under a hundred employees?
This gets a lot easier
and more straightforward at this point.
So as a small retail, for example, or a bank,
the two possible channels is either through your broker
or agent that you have all your other insurance policies.
This usually then is a so-called add-on policy
or an endorsement of cyber.
Now this basically is in your fire insurance,
they ask you, okay, now this new fire insurance type
that you made, it doesn't include cyber.
Do you want to include cyber?
And then you can book it on top of your fire insurance,
of your property insurance,
of every other insurance type basically.
So this is basically having this as an add-on
and the easiest way to do this is go to your broker
or whoever is your channel
to get all your other insurance products.
Now the other way, which is also very similar actually,
is to go into the internet.
This would be more like direct insurance sales.
And then you contact directly through a website
or like a phone number.
One of the carriers, they usually give you more likely
a standalone cyber policy.
There it is then, okay,
we have here the cyber standalone insurance for you,
which covers, for example, dimension first party losses
like breaches and ransomware and so on.
So broad brushstrokes, what are the number of players,
brokers, customers, dollars?
How do you kind of measure the overall scale
of the cyber insurance market?
What does that look like?
So there are some regulatory filings
that have to be made in the U.S. for example.
And as the U.S. is probably somewhere
between 60 and 80% of the overall cyber insurance market
in the world.
Now the best estimate that is currently out,
they've pegged the whole cyber insurance market worldwide
at five to $6 billion.
And the growth rate in the last year was about 30% annually.
And other important metrics are of course,
the number of carriers that cover,
that offer cyber insurance,
which currently is about in the U.S. 160 providers
as far as I know.
And the other important metric
is how the cyber insurance market is performing.
So the general property and casualty market,
so all your car insurance, building insurance,
your home insurance, everything like this.
If you take together all the premiums
that the insurance sector is taking in,
and you basically look at it
against all the claims that are made,
basically everything that's paid out.
The whole PNC industry makes about one to 2%,
call it net margin.
So in 100 billion that the insurance sector
in the PNC niche takes in,
they're making about one to 2 billion in profit.
Now the whole PNC market has about,
I think 3 trillion in outstanding premiums currently, 2019.
For cyber in the U.S. market in the last three years,
the margin was more like 20%.
So in the last five years,
and again, the market is growing by 30%.
So in 2020, the market,
it's like compounded interest, right?
So the market in 2020, if it grows by 30% every time,
was I think like six times bigger
than it was five years ago.
And in all those years, on average,
the cyber insurance carriers,
the insurance sector made 20%, roughly speaking,
as a net margin on every dollar they've taken in in premium.
So a cyber insurance market
has made 10 times as much in profit
on the same number of premiums taken in
as the whole PNC industry.
And this is a very surprising fact
because we know that the underlying risk of cyber
is ever increasing.
Now this will be very interesting to see
because through COVID basically,
the risk itself, most insurance policies,
they are like 80 or 90% of insurance policies
entered in at the 1st of January.
So on 1st of January, 2021,
we will see how the market reacts,
but we already know from basically industry sources
and what the big insurers are putting out there,
that the damages, of course, have increased.
As I said, the ransomware claims
have increased by 500 or 600%,
and therefore also the price
and also the willingness to provide capacity by re-insurers
is decreasing in terms of capacity
and the price is increasing.
So we are also in a very interesting phase right now.
And in the next few months,
we will see how exactly COVID also impacted cyber
in quantitative terms.
Qualitatively, we already know it has been a huge hit
and the cyber damages are increasing.
Wow, that is fascinating.
So as a recap, you know,
the traditional insurance industries,
non-cyber insurance,
are making almost 10 to 20 times less in profit
than cyber insurance,
even though the cyber insurance,
you would expect them to be riskier
and have to pay out more payouts than the traditional ones.
And 30% year over annually means
it doubles every little over two years.
So, you know, two years from now,
you would see a $5 billion market
become a over $10 billion market,
which is quite large.
So what are some of the biggest players
in the cyber insurance field
that actually offer coverage and plans?
Like who are the big name marquee guys
and maybe versus some of the smaller players
that will provide it?
Looking at the market,
the U.S. market is by far the largest one,
probably accounting for 60 to 70% of global premiums,
maybe even a little bit more than that.
And there are now about 160 U.S. carriers
who are offering cyber insurance.
And this has tremendously risen in the last years.
So basically, you know,
every medium to major insurance company in the U.S.
will likely have a cyber insurance policy.
Now, the actually five biggest players,
and we know that because the National Association
of Insurance Commissioners, I think it's called,
they let every insurance company that's licensed in the U.S.
file out a so-called cyber supplement,
but they really have to give concrete numbers
to how much their portfolio is,
how much policies they have in force, and so on.
The currently biggest insurer for cyber
after portfolio size is CHOP,
then followed by EXA, U.S., AIG,
Beasley, and Traveler's Insurance.
So you mentioned the audit for the large players.
Would the smaller players going through the add-on
or endorsement have to go through audit and negotiation
as well of their own computer systems and their processes?
That depends on the size
and also on factors like the loss history.
So if you're like businesses that has been hacked
like two years ago,
then this will likely trigger an audit
if you're not super smart.
So if you're not making like less than 100K
in turn over a year,
usually the loss history itself will trigger an audit.
But this barrier is very flexible, basically,
for non-affected businesses so far.
So in Germany, it again depends on industry.
So, you know, which industry the business is in,
also on some other factors,
the bank with 100 million in turnover, for example,
they would likely face an audit.
And if they slim down to like, I don't know,
30 to 40 million, you probably don't have an audit.
But again, this is very jurisdiction and country dependent.
So McDonald's would have like more or less standardized
audit after ISO 27001, for example.
So, you know, in terms of effort,
it would be like a 30-person day, for example,
that are required for this audit.
If it's like a smaller company,
again, maybe on the fringe of this barrier
to get an audit or not,
it might be like a one or two day job
where you have remotely people looking in there,
please show us your policies,
please show us your backup management.
Some of them are also self-assessments
and self-questioneers.
For very small companies without audit,
basically everything is a self-assessment
because you have to check the 10 security boxes, for example.
And it's a very fluid, basically, process.
And some of them are like mini audits, basically,
just for cyber insurance.
Some of them are more standardized audits.
Pen testing usually is not part of it,
at least if it's not a huge company like McDonald's.
And a business continuity plan, for example, at some size.
So if you have like 50 million turnover
in the financial field,
it's like 200 employees, I would imagine.
Then you also would have an audit going for,
if you have a business continuity
or IT system services continuity plan in place.
So this would be part of an audit,
but pen testing and all this technical stuff.
And the gist of it is, yes,
it's still more primitive and non-technical
than a cyber security professional would think currently.
But as mentioned, the market is hardening,
security standards are going up
and the price as well is going up.
But this is likely not to continue
in the next two to three years.
So let's say I get hacked and now I need to file a claim.
What's the process for doing that
once I already do have cyber insurance?
Most carriers offer a 24-7 hotline,
depending on which jurisdiction you're in.
So in Germany, there's some players basically
that offer you to do all the claims management themselves.
So you call them and then they contact you,
see, okay, what's the problem?
And then they either have in-house teams
to do the ramification
or they get third-party vendors and service providers
to help with you and they do like the management and so on.
Now, the U.S. way for compliance reasons
seems to be that the U.S. insured party
usually contacts from a list of suppliers and vendors
and basically does the contact and so on on their own.
So in this case, you basically have this list
and you start your cyber insurance policy
where you have all the players, KPMG for example,
or more technical, smaller boutique
kind of cyber security firms,
and you contact them yourself.
And they go like, hey guys,
I'm insured with, for example, CHOP,
the biggest insurer currently,
and I need to have somebody look at this.
I'm pretty sure we have a ransomware case here.
And then basically the supplier contacts you
or sends maybe the bill to you,
maybe to the insurance company,
and then the claim gets paid this way.
So do your actuaries who have their math models
and actually go in and determine the level of risk
to provide you insurance,
how specialized in cyber expertise do they actually need?
Or are they working maybe in tandem
with a customer's own in-house IT and security staff?
You're seeing now cyber teams
that are consisting of cyber actuaries.
So these are basically people who only do modeling of cyber.
They definitely have a lot of cyber expertise.
I have six people.
Some of them are cyber risk engineers,
cyber risk professionals,
and some of them are more or less traditional actuaries
that also might be modeling other than cyber damages.
And you mentioned the IT personnel of the affected party.
These are usually not involved in the modeling,
but maybe in the underwriting process.
So the underwriter who makes the decision,
do we insure company A or do we not insure it?
And if so, to which conditions?
And so the actuary usually doesn't have the direct link
to a customer's IT personnel.
And on the other hand, of course,
you try to gain from the claims
and like a secondhand kind of information through that.
But it's not like they're usually
in contact between each other.
But your own cyber risk engineers,
and especially on the claims side within the company,
they are definitely in talks.
And generally speaking,
cyber is like a very collaborative topic.
And it's a little bit like mixed martial arts, for example.
That's the way I think about it.
So in the 1990s, for example,
when mixed martial arts was born and quickly,
it was clear that jujitsu and like wrestling
and going for the jugular,
like the weak points of the human body
is better than boxing and taekwondo and so on.
Then it was like pure mixed martial arts in the beginning,
but then it converged.
So then it wasn't like the judo fighter versus karate,
jujitsu against wrestling.
Then it became just mixed martial arts.
And this is basically the process that I'm seeing right now.
We have really dedicated cyber professionals.
Most of them have like industry experience
and other insurance types.
And cyber security lines.
But I'm pretty sure we will come quickly to a point
where people might right off the university, for example,
start to only do cyber risk modeling,
cyber risk engineering,
really looking at this thing as a whole.
So do you ever see cyber insurance becoming available
for maybe individuals?
Because a lot of times people themselves
are subject to maybe ransomware or getting phished.
Yes, there's already some personal cyber products
out in the market.
This market is definitely way smaller
than the corporate cyber insurance market.
And the similarities between personal cyber insurance
in terms of what they cover to corporate cyber insurance
are very, there's not that stark of a difference
between them.
So for example, reputational harm, of course,
makes for person, for private persons, natural persons,
as much sense as for a corporation.
The same goes for data recovery, for example.
If my MacBook, which I, maybe I'm a student, right?
I saved a lot of money to have a MacBook
if that gets bricked or is unusable.
Of course, I want to have this insured as well
if it's like the result of a ransomware attack
or a breach or something like this.
And then liability, maybe,
of course, you can still be held liable in cyber terms.
Again, your Twitter account is perjuring
maybe your current firm you're working for,
and then you lose your job, for example,
as a result of a cyber attack or something like this.
So the coverage is very similar to corporate coverage.
The market is definitely there.
As you mentioned, tech non-defined people
also have the same or almost the same exposure
as more tech-related people.
And the market for personal cyber insurance
is certainly behind the corporate one,
but it's likely to also pick up steam
maybe in tandem with the corporate sector in the next years.
So you're an economist, or in German,
they call it a Wirtschaftswissenschaftler.
What got you into cybersecurity in the first place?
So as you mentioned,
I have a bachelor's and master's degree in economics.
During my master's,
I was basically a working student for a small consultancy.
Now, they were focused more on small businesses,
more on business administration.
And I wanted to definitely do something with computers.
And a school friend of mine,
he was a very good friend of a manager at KPMG
in my local town of Cologne.
And basically, I mentioned that to him,
we set up the interview, that went well.
And I basically became an intern
and then a working student for the cybersecurity service
of KPMG here in Cologne, Germany.
What was your experience like working at KPMG,
you know, doing cyber risk consulting?
It was amazing, certainly.
The intern and the working students here
that lasted like one and a half years.
During the internship, which was full-time,
I was on a project for a big financial corporation
here in Germany.
The cool thing about that was it was a huge project
with I think like 18 distinctive working streams.
And I was working in the business continuity
working team of that.
And since continuity management
basically looks at recovery plans,
what happens in a disaster case when your IT falls out
or other perils impact your business.
And the insights I gained there were very valuable.
So your latest project is a website
called cybereconomics.com.
So can people find you there, right?
And, you know, maybe tell us a little bit
about that project you're working on.
When this interview goes live,
cybereconomics.com will be up and running.
If you're interested in some of the topics
that we've discussed today,
starting from what is cyber insurance
to the academic research into cyber insurance,
and especially how the current hurdles
that we can work together as academia
and the industry to work on this kind of thing,
then cybereconomics.com is certainly the right place for you.
And we have a lot of resources.
So we first and foremost see ourselves
as a top-level summary for all topics cyber-related.
So, for example, we touched on so many topics today,
like what is in a cyber insurance policy,
how does the market look like,
what are common exclusions,
how is silent cyber affecting the market, and so on.
So there are so many topics today
that actually could be their own video.
And cybereconomics.com, if you're interested in that,
regardless of your background,
is the first place to give you a broad overview
and also all the information
and keep you up to date with developments in the cyber market
and also basically make sure
that you're at the pulse of the whole development.
There are so many stakeholders basically in this whole thing.
It's so complicated.
So many questions are still unanswered
and so many are actually unasked
that we really need a way to get every interested party
or people who might get interested in them together.
And this is basically what cybereconomics.com strives to be.
And I hope that in the future
that we can also do further videos
on the topics that I cover at cybereconomics.com.
So guys, if you want to see that,
don't forget to like and subscribe
as is usually on YouTube.
And yeah, so thank you very much, Ricky, for having me today.
It was a really pleasure.
Yeah, so I'd love to do something,
talk about cyber risk more.
I certainly learned a lot talking with you
and hope to see you soon.
Thanks, Ricky.
Have a good day.

