Uber's problems are not Airbnb's problems, and I find this continued trend of always mentioning the two in the same breath when it comes to regulation quite baffling.
Opposition to Uber is a clear case of regulatory capture. Every negative outcome that regulation is meant to mitigate (accountability, safety, reliability, discrimination, non-predatory pricing) is better served by Uber (and two-factor location tracking) than traditional cabs. I have not heard anybody raise compelling objections to Uber's model based on objective harm to consumers or market non-participants (negative externalities).
Airbnb's model, on the other hand, is rife with negative externalities. The most significant of these are the degradation of communities that were never meant to accommodate de facto hotels (noise, parking, transient traffic), and the drying up of affordable housing stock for people who actually intend to live in the homes they purchase.
"Every negative outcome that regulation is meant to mitigate (accountability, safety, reliability, discrimination, non-predatory pricing) is better served by Uber"
No they're not. Try getting an Uber as a handicapped person.
"I have not heard anybody raise compelling objections to Uber's model based on objective harm to consumers or market non-participants (negative externalities)."
What's the current story with uber and liability insurance? A while back I understood it to be: 1) uber provided secondary liability insurance while a passenger is actually in the car, and 2) they nominally encouraged drivers to get effective (read commercial) insurance for all other times but didn't take any steps to make sure they actually did.
So you ended up with a bunch of drivers that were driving around with non-commercial policies that the insurance company could easily disclaim in the event of a serious accident, as commercial use would invalidate the entire policy.
I like the way Kim-Mai Cutler ties them together: techcrunch.com/2015/07/22/uber-airbnb-and-the-conflict-between-policys-ratchet-effect-and-techs-accelerating-speed/ -- the connection being the strain they both put on public infrastructure, and the extent to which they test existing regulations which were designed for another era
> Opposition to Uber is a clear case of regulatory capture. Every negative outcome that regulation is meant to mitigate (accountability, safety, reliability, discrimination, non-predatory pricing) is better served by Uber (and two-factor location tracking) than traditional cabs. I have not heard anybody raise compelling objections to Uber's model based on objective harm to consumers or market non-participants (negative externalities).
You are aware Uber skimps on background checks that result in real harm to real people [among other problems] right? They also actively fight taxi-equivalent background check requirements?
I think the problem is people really do not understand why the vast majority of these regulations exist in the first place. The vast majority are not instances of "regulatory capture" but basic safety like "Hey, don't hire ex-cons in a line of work where they can trap you in a car" and those are what Uber focuses on repealing/fighting.
> And in California, Uber and other companies like it helped kill a law that would have required drivers to undergo a background check by the state’s Justice Department, as is required of taxi drivers.
> Uber champions its “industry-leading standards” for vetting its drivers. On its website, it describes its background checks as “often more rigorous” than those in the traditional taxi industry.
> But in statehouses across the country, Uber has fought against legislation requiring background checks as strong as those demanded of traditional taxis. Other ride-sharing companies like Lyft and Sidecar, Uber’s chief rivals, have also pushed against the laws, but supporters of stronger background checks say Uber has been by far the most aggressive.
> The driver, Duncan Eric Burton, 57, is an ex-con. He'd spent 14 years in federal prison on drug charges and was released in 2012, according to the Houston Chronicle. And he had cleared Uber's background check.
> How does that happen? The city of Houston believes Uber's background checks aren't thorough enough. That's why Houston is among the few cities to require every Uber driver to be licensed by the city and undergo FBI fingerprint checks. But while the city requires it, Uber doesn't -- allowing people to still drive for the ride-hailing service as long as the authorities don't catch them.
> Case in point: one applicant who cleared Uber's background checks had 24 alias names, five listed birth dates, 10 listed Social Security numbers and an active warrant for arrest, according to a report released last week by Houston's Administration and Regulatory Affairs Department.
> However, Pando has since learned that the driver -- 28-year-old San Francisco resident Daveea Whitmire -- has a criminal record, including felony and misdemeanor charges, and at least one felony conviction involving prison time. How, or why, Uber missed -- or ignored -- this criminal history is unclear.
Sorry but there a bunch of reasons to dislike Uber, from how they use the information they gather of you and your location, to all price manipulation it does with surge pricing.
Surge pricing is so intransparent that Im 100% sure that has been gamed against the consumer and the drivers in favor of Uber. Not only that, Uber doesnt tell you how much it costs while its going which is also a huge disadvantage for the consumer. Who knows what happens between what you paid to Uber and what the driver received?
Wait, what? I don't want to have a flamewar over Uber here, but most of this isn't even a little bit true:
>surge pricing is so intransparent that Im 100% sure that has been gamed against the consumer and drivers in favor of Uber
I guess this isn't "false" since you're just stating your belief, but if it's totally opaque to you, then how can you be so sure? I mean, OK, I guess, but... [citation needed].
>Uber doesn't tell you how much it costs
What? Yes it does - it tells you up front the multiple you'll pay, and even makes you manually type it in to confirm before your ride is started.
>Who knows what happens between what you paid to Uber and what the driver received?
The driver gets the same 80% that they get during non-surge times. That's why surge pricing works: it summons more drivers to the street by paying them more.
This is all very publicly known stuff; I don't understand your complaints here. By all means, don't ride it if you don't like it, and criticism is fair game, but try to get the basic facts straight.
> I guess this isn't "false" since you're just stating your belief, but if it's totally opaque to you, then how can you be so sure? I mean, OK, I guess, but... [citation needed].
You really think a for-profit company that doesn't have to or doesn't disclose the process to do surge pricing, having the knowledge and capacity to use surge pricing to maximize profit, will sacrifice profits in favour of the consumer?
Its opaque on Uber/Lyft, it's not opaque on taxis, so thats definitely a reason to dislike Uber(Sidecar does disclose prices before you get on the car)
> What? Yes it does - it tells you up front the multiple you'll pay, and even makes you manually type it in to confirm before your ride is started.
A multiple is not the same as the number you are going to pay. A multiple of what? I don't see a ticker like I do in a cab that tells me they are not gaming the numbers.
You say the driver gets the same 80%, but I dont know how much it cost me until I'm out of the car. If Uber tells me it cost me 10 dollars, and tells the driver that trip was billed 8 dollars for him, I would never know, and neither would the driver.
Wait, Uber's surge pricing is actually very, very good for users because it increases the supply of cars especially at odd hours. If it's 2 AM and you need a ride, thanks to surge pricing, you might find some drivers up late. If you need a ride home from a crowded football game, you'll have to pay a bit more but you'll definitely get a ride soon. Without surge, you'd be waiting a while.
Sure, what we are missing in the middle is surge pricing when there are plenty of cabs around, and its done to earn more revenue, pay more to drivers, and stiff consumers.
Its specially confusing to try to guess if 25%+ Uber, 50%+ Lyft, Taxi prices are. At least if they all had the "N per ride, M per time or distance" you would be able to compare.
This is a nonstarter for Airbnb because if they gave the type of data suggested here you'd be able to extremely easily analyze it and find all the places where illegal sublets were happening, which is a major part of their business. There is no way around Airbnb's problems without changing zoning legislation.
> Nearly two-thirds of the city apartments recently listed on Airbnb were being offered in violation of the law, an analysis by state authorities has found.
> The study of data from the subletting service’s own Web site showed 64 percent of its 19,500-plus offerings for Jan. 31 covered an “entire apartment,” says an affidavit from the state Attorney General’s Office.
There is no way to really know without being AirBnb + knowing which apartment complexes sublet illegally.
Yep. And then you need to multiply that by the fact that most of their business undoubtedly comes from major metro areas, in which almost all have laws which make AirBnB illegal.
This essay has some interesting analysis contrasting the differences between the model of licensing/guild vs crowdsourced reviews.
However, analyzing that aspect misses the point.
The heated issues around Uber is economics and how the pieces of the pie are cut up. Explaining how new technologies allow a different trust model to emerge is irrelevant to the French taxi driver throwing bricks at Uber cars or DiBlasio's proposed cap to curb congestion.
In other words, this essay is more relevant to a potential passenger of Uber who wonders how he/she can trust a car that doesn't have a government medallion.
One needs to write a totally different type of essay to appeal to angry taxi drivers and angry politicians.
There are issues on a number of levels. One is "public safety" which is essentially the licensing issue. Another is externalities like traffic, which aren't about licensing for safety but more about regulating for other impacts. Then, there are the economic / labor issues, which deserve their own consideration.
Yes, I agree with that and also the analysis in your essay.
My point is that your content (though intersting) does not live up to your ambitious (and generic) title, "A Solution to the Uber Problems".
Talking about "real time data sharing" with mayor of NYC is interesting but the "congestion" problem, which could be perceived or real, is not what comes to mind when people discuss "Uber problems." Traffic congestion caused by Uber is a relatively minor debate so far in relation to other problems it has.
The following google search for "Uber congestion" is 504,000 hits:
Because one can type "Uber" into the HN search box and read all those threads. Look specifically at the angry posts and frequent rants.
What is the recurring theme in all those angry posts?
It is not about people wondering about trust. The nastiness is all about economic unfairness.
"Trust" is orthogonal to "Economics".
The authors essay is not a "solution" to the issues people are actually debating. Also, the specific issue with DiBlasio was congestion and vehicle caps and not "trust".
While the issues of economic unfairness and trust are both important, indeed weighty enough to merit much more substantive discussion, the issue that Nick Grossman is debating in the article is regulatory oversight. The same issue is expanded on in a recent a16z podcast.
The author doesn't need to address all problems at the same time. He's not missing anything just because he doesn't discuss what you and HN want to discuss.
As my other comment to the author states, his essay in isolation is fine.
The mismatch is that his article's content (which is ok) underdelivers on the title which it oversold as "a solution to Uber's problems".
In other words, I'm not asking for his article to solve all of Uber's issues and solve world hunger. I also have no desire to discuss Uber labor. My point is that the article doesn't match the title. A more accurate title could have been, "A solution to let NYC monitor freelance driving services". (Admittedly, that type of title wouldn't attract as many clicks.)
I'm a little confused here. The article says Uber and Airbnb are quite effectively regulating themselves, and never bothers explaining why they're a problem.
I think he means they're good at regulating problems that affect them directly but they're bad at/don't care about regulating externalities. There are laws in place to protect people who stay in hotels and AirBNB does a good job of emulating those, but there are also laws to prevent hotels from messing up society as a whole (e.g. zoning laws) that AirBNB doesn't emulate well or at all.
Right, the regulations are meant to protect both the parties to the economic activity and people who aren't parties. The Uber equivalent, I guess, would be that their system can eliminate drivers passengers don't like, and passengers drivers don't like, but they have no incentive (except bad PR) to make sure, for example, that drivers not carrying customers are properly insured should they hit pedestrians. The self-regulatory mechanisms don't give the pedestrians or other non-parties a lever to advocate on their own behalf.
Why wouldn't those apply anyway? Anyone can rent out a room, condo, or a house, just by posting a classified ad, and landlord-tenant law applies. Nobody has to share information with the state under normal circumstances, but if there's a dispute, the courts apply the appropriate law. The same sort of framework could apply to Airbnb rentals.
I agree, it could work. The problem people have is that AirBnB is unregulated. Normally you do have to share rental income. There may also be local regulations prohibiting rental, sublets, short-term rentals, etc.
The absence of any data to keep Uber and Airbnb accountable is the problem. The current regulatory climate effectively black boxes the entire situation. Uber can cut corners in legal grey areas because the governments have no clue what actually is going on said areas. If they did, then regulators would be able to act (fine, issue injunctions to stop service, and etc), but obviously they can't because of what I've mentioned.
I agree up to a point. It's pretty clear that regulations in certain spheres like car and food safety have largely created a culture that doesn't universally cut corners. Then again, killing off a good portion of your customer base would do the same thing I guess as no one would want to buy your products if they have a good chance of dying from them.
Ahh, yes, the Randian ideology of the self-regulating data-driven, no-bullshit enterprise, because governments are Big and Bad and will stifle the hardworking man in their attempts to do right for society.
Regulation is based on social goals, like "people in wheelchairs should be able to get a cab as easily as people without wheelchairs."
Real time data can certainly help with policing compliance, but the data is not the basis for the regulation. We don't need real time data to discover that disabled folks have a right to get around too.
Yes. There's no way a data-driven company would give out its crown jewels for regulation just like that. Hell if I had a data-driven company I would definitely not trust anyone with the secrecy of said data. And the receiving end of said data should have little reason to trust me either.
I think governments still have a lot to do as far as taking decisions based on collected data. However, we're slowly getting there. But the politics surrounding the data will still be around forever.
Most industries are not self-regulating. If they were we wouldn't need the equivalents of the EPA and OSHA. Or the car regulations that force a minimum standard of safety. And so on and so forth.
I can see why no one would like this sort of regulatory scheme as it would put everyone's relevant data in the open. Everyone would be able to look at it whether you're the affected company or not. Effectively, it gives more data to competitors to figure out where a market is being underserved or underdeveloped which would mean existing firms would have to deal with potential losses of economic profit. Honestly, I prefer this approach despite being an anti-capitalist. This approach could be part of the solution to cure the inequities of capitalism.
Hm, why would that be necessary? Are there reasons such an ideal wouldn't just create one-to-one channels with individual companies providing data directly to regulators? Perhaps this article oversimplifies, it's not something I've ever thought about.
The city argument is they need the data for several purposes, for example aggregate information for demand projections (Uber's own data shows it slows down Manhattan traffic 9 percent) as well as entity-specific information to track compliance with applicable laws and licensing (criminal record, etc.).
Airbnb and Uber are correct when they say they need to keep some information confidential. Cities have terrible records when it comes to this. Indeed, if you look across the pond to the UK, where the HMRC is building the next Big Brother all in the name of tax collection, one can only shudder when looking at their track record of securing large quantities of data.
> 2) In exchange for that freedom to operate, companies will need to share data with regulators — un-massaged, and in real time, just like their users do with them. AND, will need to accept that that data may result in forms of accountability. Just like user data informs accountability measures within their platforms:
The problem with this is it imagines a world where the regulators don't need to perform any kind of audit/inspection.
The reality is, companies cut corners which will cause bugs and "bugs" in the data transfer process. The only way for a regulator to discover these issues is....to do what they do now.
It also ignores the fact that companies like Amazon explicitly tell people who interact with the reporting portion of their MWS business that they cannot report information in real time. And I'm pretty sure that is due to costs. This regulatory model would likely increase compliance costs for businesses with delayed reporting mechanisms because they don't need to guarantee processing in real time. The same is true of numerous companies I deal with at #DayJob. There are numerous remote sites with low bandwidth connections including in other countries that regularly have replication delays for reporting purposes.
There are soooooo many holes that are going to require the "expensive" part of regulation (boots on the ground to inspect/examine) to continue to operate the "old" way anyway.
Can we stop referring to business models which choose to simply skirt or ignore the law as "innovation"? I'm pretty sure the Mafia has been doing that for decades.
But what's also true is that we will continue to see things -- things that are perfectly acceptable to society, or at least will come to be that way -- that are out of sync with current laws and regulations. So the question is how do we get there. Clearly every innovation is not positive, but many are, and we want to give ourselves the opportunity to try things and learn from them.
As I said in another post, most of these things are not new. They've been done. Maybe not "on a computer" or "with an app", but they've been done, and are the reason why the regulations were enacted. The startup that tried to sell parking spaces was one such example. There were regulations on the books making it illegal to sell a public parking space for a long, long time. Why? Because someone tried to do it before, and it was decided that it should not be allowed.
Polymet is about to get permission to open one of the first copper sulfide mines in Minnesota. This mine will be directly over the Superior Watershed, one of the largest fresh water aquifers in the world.
We have plenty of data about how destructive copper sulfide mining is yet it looks like it will get rammed through. So the idea that "asking permission" is a thing, well, I'm not sure it really is anymore.
The solution seems to gloss over the fact that these businesses (with their relaxed licensing requirements and greater freedom to operate) would have to build a real-time government data sharing system before they could operate. Sounds almost like a way to keep smaller competitors out of the market via onerous regulatory requirements, almost the opposite of the stated goal.
One step towards the proposed idea would be for a tech-savvy government to build a platform and offer APIs. Just imagine: an API to file your tax return, to vote, to register marriage, to open a business, to register a child, to get government statistics generated from all this data and of course an API to authenticate.
"I am not sure there’s a near term alternative to this process — new ways of doing things will never see the light of day if step 1 is always “ask permission”. The answer will nearly always be no, and new ideas won’t have a chance to prove themselves."
Usually the regulations are in place for a reason. Someone has already tried those things. Maybe not "on the internet" or "with an app", but they've been done. Trying to skirt lodging and rental regulation by using your apartments as hotel rooms, bypassing both hotel taxes and rent control is not new.
Businesses have used real-time data for decades to optimize their processes, and that's all we're seeing referenced in this post. Setting aside the question of whether Uber drivers are "employees," they are certainly inside the business model, and so is the data about their performance.
But the reasons we have regulations is that business optimizations can be counter to social goals. And knowing about a bad thing doesn't allow one to go back in time and fix it.
This article makes an assumption that regulation is a tool used to protect consumers. While the rhetoric may be to make it sound that way, most regulation exists to protect incumbents who stand to lose money. This is just how governments have always worked; which is why driving accountability in government is so hard.
When incumbents have invested capital specifically to comply with regulations, it makes sense that the incumbents will demand that the regulations are applied equitably.
Regulators tend to be held highly accountable to the community they regulate.
The graphics should have been split in half. The left half is the pre-approval process and the right half is the post-approval process. The two halves compare the relative differences in the compliance burden in the two models.
Opposition to Uber is a clear case of regulatory capture. Every negative outcome that regulation is meant to mitigate (accountability, safety, reliability, discrimination, non-predatory pricing) is better served by Uber (and two-factor location tracking) than traditional cabs. I have not heard anybody raise compelling objections to Uber's model based on objective harm to consumers or market non-participants (negative externalities).
Airbnb's model, on the other hand, is rife with negative externalities. The most significant of these are the degradation of communities that were never meant to accommodate de facto hotels (noise, parking, transient traffic), and the drying up of affordable housing stock for people who actually intend to live in the homes they purchase.