Using mental models

[JBird]

I live near a restaurant that has valet, and recently the valet was robbed at gunpoint. In response to this, the restaurant owner hired an off-duty police officer to stand around the valet box in the evenings and basically just patrol. Additionally, the valet’s started to bring personal weapons to work. After a week without further incident, the restaurant owner ended the officer patrol. After another week the valets stopped carrying weapons.

 

This sort of incident has brought up an issue of wisdom I’ve considered over and over. After an event such as this robbery, there is almost always an increased awareness and sensitivity to danger. Subsequently, there’s a period without incident that leads to a desensitization of risk where security standards are again relaxed.

 

Running down the checklist of psychological models of misjudgment, it seems that few of them apply. 1) Availability-misweighing bias. 2) Bias from insensitivity to base rates 3) Stress-misinfluence.

 

I’ve tried to look at the problem from the perspective a risk manager. If the real probability of a robbery is 1/1000, it seems reasonable to spend the time/effort on security that is commensurate with that probability. If circumstances change and risk goes up to 1/500, increase security by the appropriate amount. I just wonder in this case, if the risk of robbery went up enough to warrant the hiring of a police officer (or bringing weapons to work), how did it go down enough over the next 7 days to warrant ending that service?

 

The question of course is, is this rational behavior? Have you observed this phenomenon?

 

P.S. Credit to Shane from Farnam Street for helping me identify the applicable mental models.

[LC]

I’ve tried to look at the problem from the perspective a risk manager. If the real probability of a robbery is 1/1000, it seems reasonable to spend the time/effort on security that is commensurate with that probability. If circumstances change and risk goes up to 1/500, increase security by the appropriate amount. I just wonder in this case, if the risk of robbery went up enough to warrant the hiring of a police officer (or bringing weapons to work), how did it go down enough over the next 7 days to warrant ending that service?

Here's my thought: What would you do in cases where your options do not match the risk profile?

 

So if the risk of a robbery occurs in 1/500 instances, and you have two options: hire a security guard vs. install a security camera.

 

Now the security guard prevents 99/100 potential robberies and the security camera only prevents 1/100 potential robberies, which do you choose? Where is the line between being overly-conservative and overly-risky?

 

I just made these probabilities up (obviously) but I hope you understand the question I am posing!

[JBird]

I don't have a good answer to that, LC.

 

What I'm curious about is if others have observed this type of situation and how they explain it.

 

[LC]

I don't have a good answer to that, LC.

 

What I'm curious about is if others have observed this type of situation and how they explain it.

I guess the answer to my question is one of the fundamental aspects of the insurance industry...

 

In regards to your curiosity, my gut reaction is that it is human nature. People only pay attention to what is fresh in their mind. I observe it every day at work: if we have an issue with a large client, we develop some new initiative to solve the problem. Two months later? It's forgotten.

 

To make permanent change is very difficult, especially when dealing with problems of large magnitude but do not occur frequently.

[ItsAValueTrap]

After an event such as this robbery, there is almost always an increased awareness and sensitivity to danger.

I think that this is a good heuristic/shortcut.  There are probably many situations where the frequency of an event may appear in "bunches" or the frequency of an event escalates.

 

Suppose you are in Paypal's business and somebody runs a scam where Paypal is on the hook.  Thankfully for Paypal, it *really* started paying attention to its fraud problems.  Because eventually the fraudsters kept doing more of what they were doing.  Paypal is one of the very few survivors in the online payments business.

 

2- If I was working there, I would feel a lot better if the owner cared about my well-being.  The owner's cost of treating his employees well (hiring a security guard temporarily) doesn't seem to be that high.

[alwaysinvert]

The restaurant owner is signalling something towards all his employees and customers by hiring the security guard: "I care for your safety, I care for you". This may very well increase the operations of the business sufficiently (or avoid decreases, think what would have happened to employee morale and the reputation of the restaurant if the owner appeared to do nothing) to warrant the cost. This is not a straight up weighting of probabilitites and doing a cost-benefit analysis in a vacuum is not too helpful, and hiring a guard at the time when the valet is under emotional duress may very well not be due to some sort of immediacy bias on the owner's part.

[twacowfca]

The restaurant owner is signalling something towards all his employees and customers by hiring the security guard: "I care for your safety, I care for you". This may very well increase the operations of the business sufficiently (or avoid decreases, think what would have happened to employee morale and the reputation of the restaurant if the owner appeared to do nothing) to warrant the cost. This is not a straight up weighting of probabilitites and doing a cost-benefit analysis in a vacuum is not too helpful, and hiring a guard at the time when the valet is under emotional duress may very well not be due to some sort of immediacy bias on the owner's part.

 

Good perspective.  However, beyond the motivation to demonstrate an action that shows the owner cares about his staff and customers, it's not irrational to assume that the risk of another robbery happening there is greatly increased.  Robbers like targets of opportunity and tend to revisit the scenes of successful crimes, especially if protection isn't visibly increased.

[JBird]

Same ideas I'm talking about, articulated better coming from a Harvard Law professor.

 

http://www.bloomberg.com/news/2013-04-22/why-people-stay-scared-after-tragedies-like-boston-attack.html

[boilermaker75]

Same ideas I'm talking about, articulated better coming from a Harvard Law professor.

 

http://www.bloomberg.com/news/2013-04-22/why-people-stay-scared-after-tragedies-like-boston-attack.html

 

My daughter was flying to NYC for business two days after the Boston marathon bombing. When I got home the evening of the bombing, my wife was worried that my daughter shouldn't go. I told her not to worry that this is the safest time for our daughter to go to NYC because of the heightened awareness.

[LC]

Same ideas I'm talking about, articulated better coming from a Harvard Law professor.

 

http://www.bloomberg.com/news/2013-04-22/why-people-stay-scared-after-tragedies-like-boston-attack.html

 

My daughter was flying to NYC for business two days after the Boston marathon bombing. When I got home the evening of the bombing, my wife was worried that my daughter shouldn't go. I told her not to worry that this is the safest time for our daughter to go to NYC because of the heightened awareness.

 

Oh yes, I can tell you from experience that NYC was quite safe that day. Lots of undercover cops in yellow cabs and in the subway.