Sunk Costs Fallacy Fallacy
I have a problem with never finishing things that I want to work on. I get enthusiastic about them for a while, but then find something else to work on. This problem seems to be powered partially by my sunk costs fallacy hooks.
When faced with the choice of finishing my current project or starting this shiny new project, my sunk costs hook activates and says “evaluate future expected utility and ignore sunk costs”. The new project looks very shiny compared to the old project, enough that it looks like a better thing to work on than the rest of the current project. The trouble is that this always seems to be the case. It seems weird that the awesomeness of my project ideas would have exponential growth over time, so there must be something else here.
The “evaluate future expected utility and ignore sunk costs” heuristic works well for life-planning where people get status quo bias and financial-utility things where utility is easy to calculate consistently. In fact it seems like generally good decision theory, except that I’m running on corrupted hardware. My corrupted hardware seems to decay the perceived value of a project over the time that I know of it, or inflate it when it seems new and exciting, which of course throws off the sunk costs hook’s assumption that I can evaluate utility *consistently*.
So my sunk costs hook has a bad assumption. I don’t want to go modifying the hook in a way that would break its applicability to economic and life-planning situations or its theoretical correctness, so I’ll just add “this assumes consistent utility function”. This of course doesn’t actually help me on the project planning case, I need to put a hook on evaluating the utility of a project that makes the utility function consistent.
Some things that might de-skew my evaluation of exciting new projects:
If the project is new, it probably looks shinier than it is. I should wait a while or try to correct for this before evaluating.
I should take into account my record of defecting halfway thru a project to discount the utility of the new project.
The latter half of a project is relatively untouched territory, full of valuable new experiences. I will have to work thru the first half of the new project to get to this, but I am already at the threshold on the current project.
Maybe there’s more?
So I’ll see how this works out.
I think this situation is probably not unique. Many of our debiasing hooks are formulated to combat specific biases but might catch situations outside their domain. In this example, the sunk costs bias is a real thing, but the hook to catch it was also catching a situation where sunk costs was not the primary bias, and actually ended up contributing to bias.
It might be valuable to think about what other situations a hook might catch, and modify it to not screw things up before we install it. Also, other biases may act in the opposite direction, and only hooking one of them might make things worse.
Anyways, that’s my thoughts on a specific bit of debiasing. Maybe you all have some other examples of this sort of thing, and maybe this will be useful for people who have the same problem.
- Hammertime Day 8: Sunk Cost Faith by (5 Feb 2018 1:00 UTC; 28 points)
- [Link] “Proxy measures, sunk costs, and Chesterton’s fence”, or: the sunk cost heuristic by (8 Aug 2012 14:39 UTC; 13 points)
- 's comment on Can group identity be a force for good? by (4 Jul 2021 21:58 UTC; 12 points)
- 's comment on Is Sunk Cost Fallacy a Fallacy? by (6 Feb 2012 2:24 UTC; 0 points)
- 's comment on SotW: Check Consequentialism by (4 Apr 2012 3:39 UTC; 0 points)
- 's comment on Thoughts on moral intuitions by (4 Jul 2012 5:27 UTC; -2 points)
- Visions and Mirages: The Sunk Cost Dilemma by (20 May 2015 20:56 UTC; -13 points)
It sounds like you only take the utility of finishing both projects when deciding whether to switch and disregard the effort required for finishing. Rather than deciding to work on a project with the highest final utility, pick the activity with the highest ‘utility production rate’. So if your project A is three-quarters finished then you won’t switch even if the new project seems twice as shiny.
You could try to make your existent debiasing habit conditional. For example allow it to run only when your circumstances changed in a way that can significantly affect the utility of finishing your current project or when the new project idea depends on some novel information or insight that you wouldn’t be able to come up with when you originally embarked on the first project. In other cases happily embrace the instinctual flinch against ‘wasted effort’.
I think you’ve hit the nail on the head.
It’s basic economics. We should work to maximize marginal utility per unit input cost, or bang for the buck. Once a project is half completed, you’ve got half as much work to do to collect on the value of the project. Nowhere in the analysis do I see any cognizance of the cost of completion, and actually delivering the goods.
The sunk cost issue is a red herring distracting you from your actual error.
Just to be thorough, one should not forget about future discounting. Therefore, a project that is half-finished might be worth more than a newly discovered project that would take the same total amount of time and produce slightly more than double the payout, on account of the first project finishing earlier and earning interest.
I think it is simply a planning fallacy in action. More precisely: you are already in the middle of the old project, so you already see many problems, and the planning fallacy is weaker there. In the new project you haven’t encountered any problem yet, so the planning fallacy is strong there. This is what gives you an illusion that the new project is much simpler. Maybe it is not on congnitive level, but on emotional level—the old projects feels like “a lot of hard, boring work” and the new project feels like “exciting low-hanging fruit”. Am I right?
At least for me this effect is rather strong, sometimes absurdly strong. I’ll give you an example:
I was writing a book, hundred pages already finished, hundred more pages waiting to be written. My emotions gradually went from excitement, through “flow”, to boredom, progressing to hate. So I stopped writing and started thinking: perhaps I should give up on this book and do something else instead. But what? What are my long-term goals and what are my dreams? So I imagined some possibilites, weighted their emotional appeal, made some connections like “this would help me do that” or “this and this could be made at the same time”, and finally it became obvious to me that the coolest thing I could do would be writing a book, more specifically, exactly the kind of book I was writing right now. And the feeling was so strong, my emotions really wanted to stop writing the old boring book, and to start writing the new exciting exactly the same book; logically it did not make any sense, but my emotions felt that this time it would be perfect, because the imaginary book is always more shiny and easier to write than the real book. So… It’s easy to see the mistake when your emotions are telling you that “X is much better than X”. It’s more difficult if your emotions, following the same algorithm, come to conclusion that “Y is much better than X”. But the essence is that when you compare real with imaginary, the imaginary always wins. The new project seems better because it is imaginary yet.
If all of the writers I know are to be believed, this actually happens with every book—halfway through, no one wants to finish. They’re usually glad they do, though. I wonder what that’s all about.
The obvious solution that comes to mind is to write along a smooth fitness curve, where at any given point you have a valid book.
This is probably not a good idea, but might be worth trying.
I wonder how one would do that for works of fiction, where half of a work of fiction is not a work of fiction that’s half as long.
By not writing in chronological order: Start with a brief plot summary, then just start filling in details in random places. When a section grows to long for a single chapter, split it in half, etc.
Ah. That method would be workable for people who write from an outline, assuming they don’t end up changing it too much along the way, but very unworkable for many writers.
Maybe writing is different from every other form of art I know of, but if it isn’t if you want to be a good author you really should learn to write both with and without outline, and both how to stick with it and how to be flexible about it, at will. Not until you have tried all the combination can you really tell which works best for you in the long run, as opposed to which you happened to learn first due to historical accident or the like.
Thanks for the information—now I feel less stupid. :D
When I write, my mood goes through waves. Sometimes I feel skilled and full of energy, sometimes I feel tired and unable to do anything good. I guess everyone starts writing book at their best point,but this will not last long enough. Later it feels like I started something that I am unable to finish, but I must go on anyway, and I feel like I am doing something totally worthless. Also I guess many people underestimate the time it will take them. So in addition to feeling incompetent there is a feeling “this was supposed to be finished already; am I going to write this forever?” Also I gradually change opinion on what is the best way to write; but I cannot start rewriting already finished parts, because then I would really never finish the book.
When I wrote short stories, I always started with a good idea and I thought “this will be my best story ever”. I wrote it in one day (when I stopped, I was usually unable to finish the story later). Next day the story seemed completely stupid. So I just put it away, and a week later when I read it again, I concluded that it is actually not bad. With short stories these mood swings don’t matter, because when they come, the story is already finished. With book these mood swings relate to the previous chapters. It is like “the first part of the book is great, but this chapter I wrote yesterday is completely horrible, I really should rewrite it, but I can’t, because I am already too far behind the schedule, I just hope today I will write something less horrible, but I don’t really believe it.”
I suggest spending some time in the internetspace inhabited by writers and either asking or just absorbing answers. There are a lot of articles on hollylisle.com, which is geared toward fiction, but if you don’t know where to start...
yes you are. I’ll add this to my considerations.
I have struggled painfully with the problem of abandoning projects midway. In my case at least, it’s not necessary to postulate fancy add-ons like the planning fallacy or the sunk cost fallacy. I think I’m just lazier than I wish I was and simply cannot hold onto a project for very long.
I expect this particular type of failure is more common than a fake cost-benefit analysis which is actually rationalization. I don’t even come up with excuses for not finishing what I start. I simply don’t finish things, no matter how hard or how publicly I precommit to doing so.
Finishing things is a skill, and from anecdotal evidence I think it goes up with practice. The solution might be to do really, really small things, so that finishing something isn’t so difficult for larger things.
ya I do this too, but I noticed that even my deliberative thinking about it would abandon projects, hence this post.
If you assume that your future self will make a rational decision about whether to defect, this is unnecessary.
Yes, remember the cost of getting the new project up to the point where the current project is right now. The sunk cost fallacy says you should ignore the “sunk cost” of work on the project you’ve completed so far; however, you should not ignore the “replacement cost” of that work.
I’m unsure of this. If I missed some other source of irrationality, and didn’t include this empirical catch-all, I can imagine still making the stupid decision.
Thanks for the terminology.
It is possible that the evolutionary reason of the sunk costs fallacy is exactly this problem. Different fallacies (here: sunk costs and planning fallacy) compensate each other—removing one fallacy may be harmful if it destroys the previous balance.
Yeah that’s what I was thinking. What a kludge.
I used to have this problem. In my case I think it boiled down to the planning fallicy. I’d start a project, make a little progress, but only then make an honest estimate of how much effort would be required for a shiny outcome. Meanwhile, other ideas for shiny new projects would occur, and these were unencumbered by realistic (or, in fact, any) planning.
Part of my solution is to go ahead and do some project planning on new ideas. I find the process of (and even the ancicipation of) thinking through new project ideas, expanding them, devising solutions to forseen problems etc immensely pleasureable. By doing a fixed amount of planning I get a hit of that along with a (usually sobering) estimate of what it would take.
That’s a really good point. The first project looked to have some cost X and some benefit Y. Once you start it, all the “unknown unknowns” come out of the woodwork, increasing your estimate of the cost. Let’s say to 3X.
Meanwhile, shiny new project comes along, and you estimate X as the cost again, with payoff Y. So even if you pay attention to costs, if you fall prey to the planning fallacy, it can look like the new project has a better cost benefit ratio.
Have people noticed a potential link between the planning fallacy and akrasia before? After you fall prey to the planning fallacy again and again, you will associate working on your problems with an increase in your problems. How could you ever be motivated to work in such a world?
thanks, I’ll think about trying that.
Related: http://lesswrong.com/lw/9si/is_sunk_cost_fallacy_a_fallacy/
The way i always seen the sunk cost fallacy is that if the evaluation of the value of completed work is unreliable, and the value is re-evaluated multiple times through the project lifetime, some of those times it can be grossly under-estimated. The first underestimate may result in dropping the project and no further re-evaluations will take place. There is definitely a need not to drop the project the first time the evaluation function fails.
Furthermore, in the competitive environment the unexpected difficulties you encounter are usually also the difficulties that your competitors encounter, and the project that turns out to be harder to do than expected can also have higher than expected value.
Yes, but I wouldn’t totally discount this factor. When I look at the history of my projects for the past few years I think the more recent ones do tend to pack a bigger punch in objectively measurable ways. I can see progress from “publish obscure articles cited by no-one in local, low-prestige journals” to “publish articles of interest to a reasonable number of people in the same journals, get cited a bit” to “publish articles in international, medium-prestige journals”.
If it looks to you like you’re grappling with bigger and better stuff as time goes on I doubt that is entirely an illusion.
You might just be seeking status. You might feel like you gain status whenever you declare that you will be working on a new project, and you might feel that you won’t gain as much status by finishing an existing project that people are aware of already. At least, this may be true until you gain a reputation for not finishing projects.
I generally don’t announce because I would feel incredibly bad not finishing a project that I had publicly committed to.
So it’s not about status.
Another point to consider is the planning fallacy : when you evaluate a shiny new project, you’ll under-evaluate the amount of work required to complete it. While for the already half-done project, you have a better view of what is left to do, of the unexpected problems to solve in it, … and you’ll have an evaluation of the work that is left which will be closer to the real amount of work (but usually still a bit off).
Basically to evaluate which of project A and B to do, you’ll do “expected gain of A”/”estimated work remaining on A” and compare it to “expected gain of B”/”estimated work remaining on B”. You address the issue of overestimating expected gain of new project compared to old project, but you should also consider the issue of underestimating the amount of work required for new projects.
So maybe using reference-class forecasting on the new project would help ? “Last time I had that new shiny project idea, I thought I would be done in 3 months, but after 6 months I only did half of it...”
Edit : sorry only saw afterwards that Viliam_Bur said more or less the same, but well, since I wrote my comment anyway, I’ll leave it ;)