December 22, 2007
Dear Economist: How can i reduce my queuing time at the bank?
Dear Economist,
My time waiting in a bank queue is vastly longer than standing in a supermarket one. How can I reduce my queuing time?
Ken
Dear Ken,
On a personal level, your options are obvious. You could take a collection from those behind you and use it to pay those in front of you to leave. Or you could simply bring a slim paperback and put your queuing time to good use.
But what really matters here is the cost to society. Queues are enormously costly. Imagine a bank queue in which one customer arrives per minute, and one customer per minute is dealt with by staff. All it takes is a cashier on a cigarette break, or a sudden rush of customers, and you could have 10 people in the queue. At that point, each person has to queue for 10 minutes, even though people are leaving as quickly as they are arriving. Somehow the queue must be disposed of.
The solution is elegant and unexpected: new arrivals should go directly to the front, to be served immediately after the current customer. Queues would then be very short, because once a customer was pushed back a couple of places he or she would give up and go home. The economist Refael Hassin has shown that this rule can be socially efficient, while the economics writer Steven Landsburg has advocated its introduction for telephone queues.
This makes sense. In both cases, the same number of people get to use the bank. But under the Hassin-Landsburg rule, queues are much shorter and we all spend less time waiting in line. All that remains is to encourage banks to enforce the system. Since it is perverse and counter-intuitive, they may find it very appealing.
Questions to economist@ft.com











There must be something missing in this explanation, because under the proposed system, fewer people use the bank on a given day and all the people that give up and go home apparently have to come again (thus their wasted time is quite substantial). It could probably work for telephone queues, because the capacity is fixed (so some people MUST give up, if demand rate exceeds capacity), but not for banks, where the capacity can be varied (by opening/closing windows). A bank where the maximum capacity is lower than the demand rate would not survive long.
Posted by: Jirka Lahvicka | December 22nd, 2007 at 7:20 am | Report this commentWhy would the customer go home? If I was pushed back a couple of spaces, I would leave the bank for two seconds, rejoin the queue, and be the next one served.
The bank would have to invest a lot of resources to:
-explain to customers why this queue system is good
-prevent people who leave the line from re-entering the bank
-deal with conflict generated by a system that violates common sense notions of fairness
I suspect it would be cheaper to hire more tellers.
Posted by: Dan | December 22nd, 2007 at 12:33 pm | Report this commentWhat’s to stop people in the queue walking out of the bank, turning around, and walking back in again?
Imagine the chaos when everyone joins in with the game and tries to co-incide their entrance with that of a teller becoming free.
Posted by: Ian | December 22nd, 2007 at 4:09 pm | Report this commentThe way you describe it, you do not seem to count in the disutility to the people who leave. (They have to come back or forego what they were in the line for. There may also be regret for having waited in vain.)
Correct?
Posted by: LemmusLemmus | December 22nd, 2007 at 4:48 pm | Report this commentUm.
“once a customer was pushed back a couple of places he or she would give up and go home”
I wouldn’t, I’d go out and come back in again. Straight to the front of the queue.
But hang on a minute, so would everyone else. We’d all be hovering in the doorway, aiming to “come in” just as a counter is available.
That’s (a) not a queue, it’s a scrum, (b) a nightmare for the poor security guard, and (c) probably a breach of fire regulations.
Banks aren’t interesting in “solving” queuing, because most of the time it doesn’t cost them anything, only their customers.
Customers might be interested in this solution (assuming that they’re somehow compelled to “really leave”, and not come back), but only if their estimated cost of the risk of being sent home (and thus not paying their bills) is less than their estimated cost of queueing for 10 minutes.
Good luck with that.
Posted by: Steve | December 23rd, 2007 at 7:51 pm | Report this commentIf the banks perceive their main problem to be having an excess of customers, then this is a great idea.
Otherwise, it’s really stupid.
Even if the problems others have mentioned are “solved”, I don’t think anyone would prefer to have an unknown chance of not being served at all.
At least if a normal queue is too long, you can see that as you enter and decide to leave if you want. With this system, you would never have a good idea of how long you might have to wait, because it depends on a factor (the future rate of customers entering) that you can’t see.
Any bank that tried this would lose its customers very quickly. Probably to a competitor that made fun of the stupid bank in its advertisements.
Posted by: Gil | December 27th, 2007 at 7:24 am | Report this commentUnder the current system the amount of time a person is willing to wait in line determines whether or not they will be served. Since time most certainly has value this means that the person who is willing to pay the most for the right to use the bank will be able to. Under the new system there would be no way to ensure that the person who was willing to give up the most was rewarded by getting to utilize the bank. Doesn’t seem like a great economic solution…
I wonder if anyone has found a way to determine who most desires to be at the front of the line without making everyone stand around…
Posted by: STP | December 28th, 2007 at 12:13 am | Report this commentWe’ve always complained of foreigners, who “don’t know how to queue”.
Posted by: JonA | December 31st, 2007 at 10:25 am | Report this commentNow we realize that it was they who followed the rules all along.
How embarrassing!
If only Northern Rock had paid more attention to its queueing systems. Keeping queues short its customers might have inferred that demand for withdrawals was low, supply sufficient and the situation under control.
Posted by: Claire | January 3rd, 2008 at 12:55 am | Report this commentIf only Northern Rock had paid more attention to its queueing systems. Keeping queues short its customers might have inferred that demand for withdrawals was low, supply sufficient and the situation under control.
Posted by: Claire | January 3rd, 2008 at 12:56 am | Report this comment