Comment by wheels | Hacker News Reader

Comment by wheels | Hacker News Reader

Parent Full thread

> And I see absolutely no reason why it won't happen again.

It will happen again. And it was clear that it'd happen at some point. And worse will happen. Shit happens, especially at scale.

> I think this problem kills the business model.

Nah. Let's do a little back-of-the-napkining:

• Let's assume that this is the first incidence of such.

• Let's assume that Brian's quoted 2 million nights of rentals is accurate.

• AirBnB's fees are 6-12%. To make things easy, let's assume an average rental of $50/night and 10% fees.

• There. AirBnB has now made about $10 million in revenue.

• Total cost of fixing this is maybe tops of $100k.

So this eats into their margins by 1% assuming the numbers are roughly correct and assuming that insuring against such would cost about the same thing as paying out of pocket (but would also cover the more catastrophic cases). Maybe it's a bit more and it's a couple percent, but it doesn't destroy the model by any means.

Also, if they offered insurance, they could do it as a default-on-with-scary-red-confirmation-text-if-you-uncheck upsell at a premium (say, cost + 20%) and make more money.

> But they have to change their company to address this problem

Going over the top on this one is the cost of waiting too long. They need to turn this story around. The next time, when they have a process in place for this, they just let the process take care of things and don't end up with a shit-storm on their hands.

> Let's assume that Brian's quoted 2 million nights of rentals is accurate.

The reason they haven't done anything is that 1 in 2 million isn't accurate. 1 in 50,000 is probably stretching it. Consider the accidental AirBnB trespassing/squatting reported by another HN reader here: http://news.ycombinator.com/item?id=2820644

It's starting to look like AirBnB's business model relies on tricking people into making a bad financial decision[1] and then ignoring and intimidating them when things go bad.

[1] 99.99% chance of gaining $100 with 0.01% chance of losing $50,000 is positive expected value, but if losing $50,000 would bankrupt you it isn't a game you should play.

Once you do this as a matter of policy, you are open to fraud. Now you have to spend a significant amount of your resources on fraud detection - perhaps most of them. That's what I mean by they become an insurance company; their day-to-day operation would be indistinguishable fromt that of an insurance company.

Both ebay and PayPay went through this, and that's what they spend most of their time doing. But ebay and PayPal don't need to support the open-ended kind of claim - "My entire apartment has been ransacked and my identity stolen" - that AirBNB does.

Insurance (and by extension dealing with insurance fraud) for this sort of thing is a solved problem. They don't need to do it in house. They just pick up a policy from one of the institutional insurers and pass on the cost with a markup.

Edit: To clarify, by "this sort of thing" I don't mean "Airbnb like businesses". There's nothing particularly novel about insuring homes, their contents or vacation rentals. An institutional insurer can crunch some numbers (with estimates on the average values of contents, caps on liability, frequency of incidents, cost of claims inspection, rate of fraud) and figure out what the risk profile on that is and then mark that up and sell it to Airbnb.

What's actually unclear to me, however, after crunching out a revenue estimate for Airbnb (which I'd assumed was higher) is if they're a big enough fish yet for an insurer to bother with them.

Then they're an intermediary for the same kind of insurance that hotels and bed-and-breakfasts have. I don't know if that's even possible to do - how do you approach a company for a policy that covers hundreds of thousands of homes around the world?

I would bet dollars to doughnuts that a traditional holiday apartment will only be able to get insurance that is viable if they meet a number of criteria (removal of valuables, cheap furnishings, inspection etc) that would not be available to someone putting their primary residence up for piecemeal rental on AirBnB

I doubt it is a solved problem or AirBNB would be using it.

Imagine what the negotiations would be like. They need insurance for: -An open-ended number of people worldwide (constantly changing) -Any sort of property from shitty apartments all the way to fancy houses (no estimate of value possible, requires appraisers) -Damage/theft caused by people with unverified identities (all they have is CC, stolen?) -Any and all damage (including accidental death of people staying there?), including stolen identities and fire damage (to other properties as well?) -No security requirements for properties (fire alarms? carbon monoxide detectors? checking up on the place?)

I have a feeling that would cost more than 10%.

Imagine negotiating insurance for a mobile business where people are flung around 30+ feet in the air with no safety harness, others stick their heads in the mouths of lions or are "sat" on by elephants, and still others stick flaming swords into their throats. Meanwhile, children are riding on devices assembled by people who may not have a high school diploma (at least that's the impression I get). These businesses are generally very well insured.

There are insurance companies that deal specifically with unique, high risk insurance situations that are only applicable to very few clients. You don't buy these policies from Nationwide (directly), but they likely have a subsidiary which handles policies like this. In turn, those subsidiaries take out insurance policies on the high risk policy (it's turtles all the way down).

Disclaimer, my FIL used to work for an insurance company who dealt with high risk policies.

That's not exactly the point. An insurance policy for a high risk activity - let's say skydiving - is very well definited. There are statistics on accident rates. There is a pretty defined payout when someone dies. The problem is not the risk level, it's the undefined nature of the problem.

AirBnB doesn't know the value of the property they are renting. They don't have a long enough history to figure out what these kind of things could cost. What if I rent out a mansion, are they as liable there as in this case? Does AirBnB limit their pay out? These are questions they will have to come through at some point.

> Does AirBnB limit their pay out?

Yes. And make it clear that the payout is limited in the TOS. That way AirBnB/their insurer assume some well-defined amount of risk and it's not all on the renter.

The issue of insurance fraud is tougher though.

The other tricky thing is that they need to negotiate this per jurisdiction. They can't sell insurance directly in most markets without being an insurer, and there are all kinds of BS legacy laws about referring and reselling insurance too. (i.e. you'd need to go to classes and be a registered agent to offer it)

It would be a fun problem to solve for them.

I don't think the actual per-room-night price for, say, $10-20k in contents and $bignum in liability (i.e. my guest burns down the building and kills everyone) would be that high; on the order of $10. 2m nights is enough to have some data, and this is similar to the vacation market.

It could even just be a rider on top of existing homeowners/renters insurance. A nice trick might be to sell renters insurance at the same time, and make money off that (which is IMO something everyone should have anyway, even if they don't use airbnb), with free coverage for airbnb use.

At least circuses have low potential for fraud -- nobody gets eaten by a lion just for the cash. If airbnb started insuring (directly or indirectly) against damage caused by renters then there's quite a few ways for unscrupulous folks to try to rip that off.

I'm not saying it's not possible but it is expensive (high risk = high cost). What's AirBNB's margin? They charge 10% but after expenses, can they make money paying the insurer X%? That's the fundamental question and I suspect it's "No".

You're missing the point. It may be that the rewards possible from renting out your house for a few nights might not be worth the risks in letting strangers into your home (at least under the current terms offered by Airbnb).

That would mean that AirBNB doesn't have a viable business model, or that it's a much smaller market than previously thought.

Saying "insurance will handle it" isn't an answer. AirBnb needs to come up with some creative and effective policies to solve the real underlying problems first. A good start would be better identity verification of renters.

Once they've turned it from a bad risk into a good risk for a large market of people, then they can outsource the details to an insurance company.

Lloyd's will insure it if you can think of it, basically.

AirBnB is not that far out there that they couldn't get at least three special risk companies to give them a decent quote.

I'm in the insurance field btw.

Can you even get insurance on an illegal activity. It seems clear that many people on Airbnb are either illegally subletting their rental apartment or in places like NY where Airbnb is violating laws meant to stop illegal and unlicensed hotels.

Sub-letting is more often perfectly legal, but yes, I'd imagine if you are in contradiction of either your rental agreement and/or local laws you would also be voiding your insurance contract.

Why isn't acquiring this type of insurance the responsibility of the renter?

Sure, if you install a surveillance system or have a doorman/reception area, you can get a decent insurance rate. Also, you have to buy insurance in 6/12 month chunks, even if you rent out the place for 3 days a month.

All of a sudden the rates that greedy hotel chains charge don't seem so greedy.

OK, so they're an insurance company, just like ebay. And Google isn't really a search company but an advertising one. So what?

Your underlying premise seems to be that a responsible business model isn't viable. Is your argument that therefore companies should be irresponsible where there is profit to be made?

I would argue they become an insurance company, or make it clear that the householder assumes all the risk and makes it possible for them to minimise that risk i.e. by removing the obstacles they provide to investigating the other party.

This incident is suggesting that instead they mislead the customer regarding the risk, deliberately reduce their ability to manage that risk, in order to maximise their profits.

Setting aside the moral and legal aspects of such a strategy, I doubt that's good for business. Who wants to trust their home to a company with a reputation for cutting corners to make money?

Your underlying premise seems to be that a responsible business model isn't viable. Is your argument that therefore companies should be irresponsible where there is profit to be made?

Geez, that's a leap. I thought I made myself clear: they either need to change their company completely to deal with this problem, or they will cease to exist. I'm not certain that they can completely deal with this problem - the risks are much more open-ended than with ebay and PayPal. So I think it's possible the business model cannot survive - which I thought I made clear when I said it was not sustainable.

I'm arguing that their company may cease to exist because of this problem. I don't know how you connected that to advocating for companies to be irresponsible.

Reply on news.ycombinator.com