5 ms·

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

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.

evanpw 10y ago
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!
- kirubakaran 10y ago
  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.
  - emodendroket 10y ago
    According to Consumer Reports the last I checked, State Farm is near the bottom of the pack in customer satisfaction among claimants.
    - kirubakaran 10y ago
      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.
- JimboOmega 10y ago
  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.
  - emodendroket 10y ago
    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.
    - JimboOmega 10y ago
      All the more reason that they separate from the fair market price!
- emodendroket 10y ago
  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.
qntty 10y ago
This is really insightful. Does this fallacy have an well known name?
- sbuttgereit 10y ago
  Not that I know of... the closest might be the Broken Window Fallacy (http://www.investopedia.com/ask/answers/08/broken-window-fallacy.asp 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.
- jraines 10y ago
  Sampling bias, I think
- evincarofautumn 10y ago
  I think it’s attrition bias, a form of selection bias. https://en.wikipedia.org/wiki/Selection_bias#Attrition https://en.wikipedia.org/wiki/Selection_bias#Attrition
- sjbp 10y ago
  Self-selection bias : https://en.m.wikipedia.org/wiki/Self-selection_bias https://en.m.wikipedia.org/wiki/Self-selection_bias
- harterrt 10y ago
  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.
mrisse 10y ago
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-dma.org/moty/mostrecenthonorees.html https://web.archive.org/web/20111201164345/http://www.the-dm...

evanpw 10y ago

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!

kirubakaran 10y ago
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.
- emodendroket 10y ago
  According to Consumer Reports the last I checked, State Farm is near the bottom of the pack in customer satisfaction among claimants.
  - kirubakaran 10y ago
    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.
JimboOmega 10y ago
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.
- emodendroket 10y ago
  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.
  - JimboOmega 10y ago
    All the more reason that they separate from the fair market price!
emodendroket 10y ago
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.

qntty 10y ago

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

sbuttgereit 10y ago
Not that I know of... the closest might be the Broken Window Fallacy (http://www.investopedia.com/ask/answers/08/broken-window-fallacy.asp 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.
jraines 10y ago
Sampling bias, I think
evincarofautumn 10y ago
I think it’s attrition bias, a form of selection bias. https://en.wikipedia.org/wiki/Selection_bias#Attrition https://en.wikipedia.org/wiki/Selection_bias#Attrition
sjbp 10y ago
Self-selection bias : https://en.m.wikipedia.org/wiki/Self-selection_bias https://en.m.wikipedia.org/wiki/Self-selection_bias
harterrt 10y ago
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.

mrisse 10y ago

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-dma.org/moty/mostrecenthonorees.html https://web.archive.org/web/20111201164345/http://www.the-dm...