Sunk Costs

When I started this entry, I intended to write an essay about keeping resolutions. After all, it was still close to new years and people did this sort of stuff at this time. After three weeks of procrastinating, I got around to writing the thing. The essay boiled down to focusing on the whys over the hows because given the latter, a sufficiently motivated person can always figure out the how. Essentially, it was another "know thyself" piece. If this sounds familiar, it's because I also write about the topic (publish)here, (publish)here and (publish)here.

Though this new article was different, I realized that I was in danger of creating my personal redwall sequel (a fantasy series set in the realm of redwall inhabited by talking animals; in every book, an evil warlord takes over the land and the good animals must band together and go on a perilous quest to overthrow the warlod; this is exciting the first time but the author rehashes the theme for another dozen books). Besides, it was a new year so it seemed fitting to explore a different topic.

So I committed myself to write about something different. But instead of starting from a new article, I tried editing the original piece, trying to shape the conclusion into something radically different. I loathed to start over since I already spent so much time on the original post. But the more I worked on changing the old article, the less I liked it (but the more invested I became in not abandoning it). After a third shoddy rewrite, I stopped and forced myself to see the essay for what it was at that point - a sunk cost.

A sunk costs is a loss that has already been incurred and cannot be recovered. In economics where we assume the presence of rational agents, sunk cost would not factor into a transaction because a rational agent would disregard that which has no ties to the present situation. Unfortunately, people are not rational agents and it is extremely difficult for us to accept a sunk cost. The greater the sunk cost, the greater the tendency to dig down and try to recoup the loss - this is how gamblers gamble away their life savings instead of walking away when they are still able to do so (with pants still on them).

Accepting my original essay as a sunk cost, I scrapped it and started writing something new. To my surprise, I found that the words came easy this time in a rapid stream of consciousness that soon filled the screen with words. Compared to working on the old piece, it felt like I moved from maintaining a legacy java spring enterprise Frankenstein to building a new service from scratch in python (aka everything was about 100x better).

I finished the new entry in less time than I spent revising the old one. Had I not spent so much time trying to repurpose the old article into something new, I could have saved myself even more grief. But vision in hindsight is always better and persistence (or stubbornness as my parents would call it) has always been core to my character. This is why I find it hard to let go and accept a sunk cost - it feels too much like surrender. But there's a fine line between not giving up and being obstinate and I fear of having been on the wrong side of that line more times than I would care to admit. Like in poker, there's no shame in folding a hand in order to play another round.

Of course, this idea of accepting sunk costs can also be pushed too far the other way. Sometimes you need to give ideas more time to develop and not lose faith too soon. Just ask any seasoned javascript developer about the "standard way" of developing web apps to see the dangers of starting over too many times. Sometimes wheels are nice and don't need to be re-invented (at least not until the release of the consumer hovercrafts).

Like with most things in life, there are no categorical rules about dealing with sunk cost so one must do it intelligently. Where would we be if Steve Jobs wrote off Apple as a sunk cost after being forced from the company? Or if a certain British female author called it quits after her book about a boy discovering he was a wizard was rejected by the first dozen publishers?

I want to now take a brief moment to talk about the cousin of sunk costs - mistakes. In improv, the secret about mistakes is that there are no mistakes, only missed offers (aka its not a bug but a feature). Mistakes are only mistakes if you don't commit. For example, if you're in a scene and you mess up someones name, then your character becomes the person who can't do names. You will proceed to purposefully call your partner by a different name every time and the audience will love you for it.

Thomas A. Edison was one of the most prolific inventors ever, with over a thousand patents to his name. "The Wizard of Menlo Park" wasn't without his share of do overs and when asked about his thousands of failed attempts at creating the light bulb, Edison replied: "I have not failed. I've just found 10,000 ways that won't work". Like mistakes, sunk costs only become so if you let them sink in. Seen the right way, they can be incredibly insightful. So the thing about sunk cost - don't be afraid of incurring them. Starting over is not necessary a bad thing. Sometimes, that's the only way to move forward. In the process, as it goes in the Pocahontas song, you'll "learn things you never knew you never knew".


  1. thoughts