This could also be an artifact of the metric. I call this the "Geico" fallacy.

Geico advertises that "People who switch to Geico save 15% or more". Notice that people who would lose money by switching to Geico should rarely switch. Accordingly, this metric will usually be positive.

Consider that people moving to a new city for a job may need more incentive to move.

I've always found those advertisements weird. If consumers are rational and have good information, then the average savings from switching is the same as "how much do you have to pay someone to switch to Geico?". The higher it is, the worse you should expect service to be!

Consumers don't have good information; applying for car insurance is a big pain, so they usually don't know. They probably apply once, then have the same car insurance for years... which might mean that they are paying too much.

Same basic problem as an employee who has worked at the same company for 10 years, and gotten their 3% raise every year - the market may have changed radically, but looking is a burden (emotionally, and otherwise) so they are unaware of that.

Enter a tremendous flood of car insurance advertising... to get people to actually evaluate if they are paying too much.

Most insurers actually increase premiums over time regardless of whether you make any claims. In the past they offered longtime customer loyalty discounts but as some point some bean counters looked into consumer psychology and noticed exactly what you're talking about so now it's up every year.

All the more reason that they separate from the fair market price!

In practice it's quite difficult to compare insurers in this way since it's not that often you need to make a claim and online reviews are overwhelmingly biased toward negative reviews so they're all like one-star. As far as I can tell the correlation between price and service is not that strong.

For what it's worth, when I had to make claims with Geico it was fine. But I was clearly not at fault in either case so it was not a difficult process.

I also think part of their ability to offer lower premiums is generally having lower human involvement -- no independent agents, claims routed through Web site or call centers, claims adjuster on site in body shop instead of visiting you, etc.

Exactly! "Customers pay us more than our competition. They clearly love us and we're doing something right" is a nuanced point hard to make in an ad, but it needs to be made somehow. That's probably why Statefarm et al focus on their customer service in their ads, but those ads don't sound so catchy.

According to Consumer Reports the last I checked, State Farm is near the bottom of the pack in customer satisfaction among claimants.

Oh interesting. I guess that's the benefit of not competing on price. People assume you're probably good at something if you aren't the cheapest.

I like the term you've coined, but their claim is actually significantly weaker.

"Fifteen minutes could save you 15 percent or more on car insurance."

https://web.archive.org/web/20111201164345/http://www.the-dm...

This is really insightful. Does this fallacy have an well known name?

Self-selection bias : https://en.m.wikipedia.org/wiki/Self-selection_bias

I think it’s attrition bias, a form of selection bias.

https://en.wikipedia.org/wiki/Selection_bias#Attrition

Not that I know of... the closest might be the Broken Window Fallacy (http://www.investopedia.com/ask/answers/08/broken-window-fal...), where only the effect of "seen" outcomes are front of mind and considered, but "unseen" outcomes are ignored even if they are more impactful to reality.

Sampling bias, I think

Thanks!

While I agree with the others calling this a type of sampling bias, I've seen this type of error frequently enough that I think it deserves a more specific name.

It could be as simple as Bay Area hackers being accustomed to higher salaries and negotiating higher (or not accepting lower offers) even in places with cheaper living expenses.

There is that, and the fact that in the eyes of prospective employers Bay Area experience is highly valued.

Honestly I don't think experience in Bay Area is necessarily better, there are great teams in Europe/rest of US and shitty teams in SV, but for some reason there is this Bay Area "halo".

Source: I've relocated to Europe after 3 years in Silicon Valley.

Maybe European companies see a Bay Area "halo," but I wouldn't say the same about American companies.

My company has software developers in Los Angeles, Boston, Dallas, and Sunnyvale. From what I've seen it's a lot easier to bullshit your way into a senior software role in the Bay Area than it is elsewhere, so the rest of us tend to be a bit wary of the ones we don't yet know well.

It's easy to get a senior role at a random company in SV. There are many companies looking for senior people, provably more than there are candidates. However you have to check if they stick, and what they do. It's actually fairly easy to weed out the bad ones with a 45 minute interview if you don't let yourself be seduced by their ability to talk well. In my entire career I have only once heard of someone who seemed senior doing really well at coding exercises and then turning out to be horrible. I have frequently heard of people not coding as awesome as you expected but talking a good line and then being disappointed after the hire.

It's BS but the most effective negotiation tactic in salaries is "I make $X now so I want to make at least $X+Y."

Perhaps developers in the Bay Area learn very quickly how much demand there is for their skills and that confidence translates directly to leverage during negotiations, regardless of where they look for a job. There's too many confounding factors to clearly pick out which are significant without much more precise data.

On a side note: Calling them hackers when you clearly mean run of the mill corporate programmers seems like something straight out of an HBO show.

Oh, I default to using the hacker label as a compliment to programmers here on news.yc. I tend to assume (based on reading bits of code from various news.yc contributors) that many devs (carefully not bucketing myself in this group) here exhibit a level of mastery above a run of the mill corporate programmer.

I can offer only anecdotal evidence; take it for what it's worth (not much, I know).

I lived and worked in the Bay Area for a little over fifteen years, almost two thirds of it at Apple, and almost a third of it at a successful startup called Reactivity.

I left in 2006 because of some catastrophic health events, and have lived and worked in Arkansas since recovering my ability to work.

When I'm working, I find it pretty easy to command higher-than-average pay for this area. Of course, I fairly often work remotely for people still in the Bay Area, or in Boston or Denver, and naturally they're going to tend to offer me more than Arkansas companies would. But I can also command a pretty good premium in Arkansas.

I don't know how much you can take from my example. I'm an old-fart developer now, with a long resume that has a bunch of famous names on it. People who are impressed by those things will naturally tend to offer more money to get them. I've seen some employers in this area become visibly more enthusiastic because of some of the people I've known and worked with, or that I can offer as references. My assessment is that some of that enthusiasm has been respect by association.

On the other hand, there may be some real quality differences in engineering culture, and some fraction of the premium may reflect that. Some of the markets I've worked in outside Silicon Valley have significantly different standards and practices, and the differences are not often for the better. In some cases they've verged on infuriating or clown-car laughable. Maybe some fraction of the premium is employers paying for people they expect to know better, and maybe they're not entirely wrong.

A lot of employers look at previous salary to determine a base point for a new job. Coming from somewhere with a high salary makes it much easier to "justify" asking for more.

How would they know your previous salary?

they ask, and some people stupidly answer.

It's not a big mystery. They ask for higher salaries, because that's what they are used to and end up getting paid more. It's not because their skills are more valued.