Typical HN reaction - smug intelligent-sounding question to doubt the validity of a hypothesis backed by data and detailed analysis, by completely ignoring the details in a given linked article.

Your question is answered on page 15 (How Do We Determine the Causes of Changes in Congestion?) of the report you are happy to debunk so eagerly.

Link (PDF): https://www.sfcta.org/sites/default/files/content/Planning/T...

It's hardly rocket science - the cost of driving/being driven has gone down, so usage has gone up.

I think real issue here is that road usage is mispriced. Singapore does it well: https://en.wikipedia.org/wiki/Electronic_Road_Pricing.

A tax on just ride hailing is mad. They are cheaper than owning your own vehicle, in part, due to real efficiencies in terms of maintenance and depreciation. Why on earth would anyone want to incentivise individual car ownership?

>I think real issue here is that road usage is mispriced.

The problem is not that roads are too cheap but that the alternatives are too expensive in terms of time, money and quality. Making what is the most effective (all things considered) transit option for a huge chunk of the population more expensive just hurts everyone. It's like raising prices on bandwidth in order to reduce network congestion. The solution is more capacity. That capacity doesn't need to be for cars but you need to add capacity.

I don't think the problem is capacity so much as temporal capacity. There's a large opportunity cost in having a big road empty for most of the day. Surely a smaller road that's busier for more of the day would be better?

I don't in principle have a problem with bandwidth being priced according to congestion. What we have instead is rationing.

You could make road pricing approximately revenue neutral by reducing fuel and vehicle excise duty.

Road traffic is a particularly good case for variable pricing because there is a discontinuity in traffic jam likelihood versus cars on the road - essentially, roads get full at a certain point, so the marginal cost of extra cars increases rapidly as it approaches that level.

This would encourage traffic to spread across more of the network, deliveries to happen at off-peak times, the 9-5 work cycle to be relaxed, pollution to be spread out more evenly across the day (to times when people aren't walking around).

So, for clarity: I'm saying make it cheaper to use the roads sometimes, and more expensive at other times.

> Publishing reports like these without controlling for the economy/jobs/popularity of these cities seems to me to be an unfair attack on Uber/Lyft, where the real culprit might just be a strong economy.

I agree. That is why this report is so good. The section "How Do We Determine the Causes of Changes in Congestion?" covers this question.

" this time to systematically estimate the individual contributions to changes in roadway congestion of the factors of transportation network supply change, population change, employment change, and TNCs."

I agree that they have done a good job and that the conclusion that Uber/Lyft contributes to congestion increase is very valid and should be considered to take actions in the city.

At the most basic level: YYou take your congestion per citizen metrics before the market entry of uber, which should be linearly related, and extend that line. Then you draw another line with the metrics after uber. If you see a difference and you can't explain those away wwith other causes, you're confident it's ddue to uber.

The study very likely threw some more statistical tests, controls And models at it, but that's the gist.

Not an unexpected rresult either: personal vehicle use means the car is on the road twice per day, uber twice per passenger and the distance between passengers.

"and you can't explain those away wwith other causes" feels very ambitious to be sufficiently confident that there couldn't possibly be another cause, but I guess that's the best we can do

How exactly would you control for all the rest? Where is there another San Francisco that doesn’t have Uber that we can go measure?

I think these estimates pass the sniff test at least. Downtown SF has very little parking relative to demand, so Uber / Lyft are practically the only way to get there in a car.

"Downtown SF has very little parking relative to demand, so Uber / Lyft are practically the only way to get there in a car."

So, wait... Your argument is that there is a natural disincentive to go downtown with a car, unless one uses Uber or Lyft... But you also somehow doubt that the car traffic that is downtown has been caused by Uber and/or Lyft?

So is everyone an idiot because they just drive around downtown, give up, and go home? Or they aren't in on this Uber/Lyft secret that nobody has heard about?

Or... And I'm just spit-balling here... Maybe you just argued against yourself very effectively.

My point was actually that the traffic is already relatively self-regulating due to the lack of parking, most people don’t get there by driving. Sure it’s plausible that Uber or Lyft increased the traffic, but it’s difficult to assert that in an alternate universe without Uber or Lyft that more parking would not have been built (attracting more cars), or that you’d have more traffic from cruising (people hunting for parking). Development plays a huge role as well, perhaps Uber has allowed more development than would have been feasible without it, or perhaps all the new development would have happened anyway, either way it’s a complex adaptive system that you can’t easily just say “ah, THIS is the one and only cause.”

The fact that it’s extremely difficult or impossible to control for these things doesn’t mean that any analysis which doesn’t control for these things is valid or useful.

>Downtown SF has very little parking relative to demand, so Uber / Lyft are practically the only way to get there in a car.

Why do you need to get there in a car?

Is this a trick question?

I use Muni and Caltrain.

I guess your assumption is that every one gave up on buying car and instead every day they drive to work. The impression i have from my friends is different: they use lyft when they cant take car or want to drink while with strong economy the can afford having car as well. I dont know any single person who would sell their car because they can travel with lyft to work every single day.

Really? My experience has been extremely different, albeit both of us only really have antidotes.

My friends/coworkers who got jobs in the city directly after college all don't have cars. Think 2013-2018 graduates. The only ones with cars, either grew up in the bay, or have worked outside the city at some point. Mind you these are all people who have a license in their prior state/country.

How many of these people would stay in SF if Lyft/Uber were banned entirely?

At first most if not all, but it might play a part in their long term goals. Hard to say though.

You meant "anecdotes", right? Autocorrect, probably.

Lets go with that ;)

Agreed. At least also post the increase in number of employees in the same area.