3 ms·

Sunk Costs

21 points by icey 17y ago

notauser 17y ago
I'm certainly not going to argue against his point, however there's a related and more subtle one which gets ignored a lot. Generally if you are writing off sunk costs there are also some accrued benefits lurking. Before you terminate X you need to look at those. Examples are things like: - Cost of restarting things like advertising in the future from a cold start. - Continuity and loyalty benefits from suppliers. - R&D realized or in-progress. - Accrued experience. That last one is a biggie. There's a reason that failing companies get new injections of capital. Quite often their major asset is that they know where the toxic slime pits in market are - because they fell into them. So, the money is gone and you shouldn't give it a second thought. But you probably bought some assets with it (plant, experience, contacts) and you should think about which ones will diminish of evaporate the moment you stop doing something. Revalue these assets in terms of the cash flow you can extract from them, write off the actual amount spent, and then you have a good basis for making the decision.
- smartbear 17y ago
  I agree completely. In the real world costs are often NOT completely sunk. I could even add to your list, e.g. employee morale. Along your lines of argument, you might say you've INVESTED money instead of just SUNK. At the same time, we also tend to hold on to obviously fruitless enterprises. So the real lesson of course is to try to be objective about what's really going on, suspending your ego for the benefit of the project. Which I'm sure we agree on!
ryanwaggoner 17y ago
The difficult thing with sunk costs that this article fails to really cover is that very often we don't know they're sunk costs until after the fact. In the middle of a situation, it's very difficult to predict whether things will only get worse and you should abandon your efforts. How many successful companies and products got off to a rough start and would have been shut down if this kind of analysis had been applied?
- smartbear 17y ago
  I like your point, but in the specific examples given of my own mistakes with sunk costs in startups, I usually did know it was wrong at the time. I just didn't have the stones to pull the trigger.
- mikeryan 17y ago
  No this isn't true. A cost is "sunk" as soon as the money (or time) has been spent. The fact that a cost is "sunk" shouldn't be confused with determining the value of an expenditure. When you start throwing in "value" the whole concept gets confusing. Sunk costs _only_ mean that the fact that money has been spent on something should not affect future decisions. What you got out of spending that money should absolutely be considered in future spending.
MikeCapone 17y ago
If you're interested in cognitive biases, heuristics, etc, here are some resources: http://www.lesswrong.com http://www.lesswrong.com (start with the "about" and "sequences" pages) http://en.wikipedia.org/wiki/Cognitive_biases http://en.wikipedia.org/wiki/Cognitive_biases http://en.wikipedia.org/wiki/List_of_fallacies http://en.wikipedia.org/wiki/List_of_fallacies http://en.wikipedia.org/wiki/List_of_common_misconceptions http://en.wikipedia.org/wiki/List_of_common_misconceptions Also check out the books of Daniel Kahneman and Amos Tversky.

21 points by icey 17y ago

I'm certainly not going to argue against his point, however there's a related and more subtle one which gets ignored a lot. Generally if you are writing off sunk costs there are also some accrued benefits lurking. Before you terminate X you need to look at those. Examples are things like: - Cost of restarting things like advertising in the future from a cold start. - Continuity and loyalty benefits from suppliers. - R&D realized or in-progress. - Accrued experience. That last one is a biggie. There's a reason that failing companies get new injections of capital. Quite often their major asset is that they know where the toxic slime pits in market are - because they fell into them. So, the money is gone and you shouldn't give it a second thought. But you probably bought some assets with it (plant, experience, contacts) and you should think about which ones will diminish of evaporate the moment you stop doing something. Revalue these assets in terms of the cash flow you can extract from them, write off the actual amount spent, and then you have a good basis for making the decision.

smartbear 17y ago
I agree completely. In the real world costs are often NOT completely sunk. I could even add to your list, e.g. employee morale. Along your lines of argument, you might say you've INVESTED money instead of just SUNK. At the same time, we also tend to hold on to obviously fruitless enterprises. So the real lesson of course is to try to be objective about what's really going on, suspending your ego for the benefit of the project. Which I'm sure we agree on!

ryanwaggoner 17y ago

The difficult thing with sunk costs that this article fails to really cover is that very often we don't know they're sunk costs until after the fact. In the middle of a situation, it's very difficult to predict whether things will only get worse and you should abandon your efforts. How many successful companies and products got off to a rough start and would have been shut down if this kind of analysis had been applied?

smartbear 17y ago
I like your point, but in the specific examples given of my own mistakes with sunk costs in startups, I usually did know it was wrong at the time. I just didn't have the stones to pull the trigger.
mikeryan 17y ago
No this isn't true. A cost is "sunk" as soon as the money (or time) has been spent. The fact that a cost is "sunk" shouldn't be confused with determining the value of an expenditure. When you start throwing in "value" the whole concept gets confusing. Sunk costs _only_ mean that the fact that money has been spent on something should not affect future decisions. What you got out of spending that money should absolutely be considered in future spending.

MikeCapone 17y ago

If you're interested in cognitive biases, heuristics, etc, here are some resources: http://www.lesswrong.com http://www.lesswrong.com (start with the "about" and "sequences" pages) http://en.wikipedia.org/wiki/Cognitive_biases http://en.wikipedia.org/wiki/Cognitive_biases http://en.wikipedia.org/wiki/List_of_fallacies http://en.wikipedia.org/wiki/List_of_fallacies http://en.wikipedia.org/wiki/List_of_common_misconceptions http://en.wikipedia.org/wiki/List_of_common_misconceptions Also check out the books of Daniel Kahneman and Amos Tversky.