Wednesday, October 10, 2007

Pacaran is sunk cost

I said in the last post that the guy who was left by his girlfriend after what he did (he said he had done everything -- presumably good things in his view) was just not good enough for the girl. Well, if I were my student, I would have given that answer a B-minus. A better answer is: because the costs incurred during dating or pacaran are sunk.

Again, dating is a two-to-tango game. So, when the guy said he had given everything, chances are she had given everything, too. When they both were giving each other, they incurred both costs and benefits -- of which I do not want to dwell into details.

In the story (Samuel Mulia, my apologies, I'm using your article here again -- by the way, keep up good writing), one person is left out: the girl. The decision that is relevant to see why "she left me after all what I had done" is the girl's decision to leave him, not the guy's fate to be left alone. That is, what matters is what probably triggered the girl to leave. So let's focus on her.

In the period of dating, the girl incurred costs: makeups, fancy skirts, perfumes, even the effort to walk and talk "like a lady". All this she did to attract and later, to keep that lucky guy with her. Right before she did every single item above, she had an expectation: All these I would do because I expect him to give me this or that, in return. But once the activities were done (e.g. once the fancy skirt was bought), they all became sunk costs: they were not recoverable (well, she might go to second hand market selling the used skirt, but hey, she is a lady).

Fast forward. There was a quarrel between the two lovebirds. The girl was now at the margin of decision: leave or stay (or, if you don't like these terms, use: leave him or keep him). What would you think she had in mind, as factors affecting her decision? Yes the costs and benefits of leaving him. She would leave him if the resulting benefits (relief, chances to get a new guy, etc) exceeded the costs (effort to make him go and to make up good, convincing story to friends, etc). But the costs incurred during pacaran were not relevant anymore.

(If this is not convincing, try picture you are about to go mudik. You are thinking of driving your car or simply take the train. The amount of money you used when you bought the car is not relevant to your decision).

So, ladies, next time a jerk begs you to not leave him, using "Please don't go, Honey. Remember what I have given you all this time?" Tell him: "Baby, it's all sunk. Now shut up".


  1. i wonder if there's anyone out there whose mind works this way...

  2. Maybe no one, Tirta. As Warkop says: "cinte itu bute". But at least we have a framework to explain what actually happens.

  3. Aco
    "But the costs incurred during pacaran are not relevant anymore."

    "Are not" or "should not be"? Positive or normative? I'd say the latter.

  4. Arya, first, I think I made mistakes with the tenses -- my bad, sorry. Now corrected. But your question is still valid, so let me adjust it:

    Arya: "Were not" or "should not be". Positive or normative? I'd say the latter.

    Well, I think it depends on where we're looking from. If I am a person wanting to understand why she left him, then it is "were not". But if I were the girl herself in the edge of making up decision then I would tell myself that sunk costs "should not be" relevant. (Otherwise, I would probably have to keep him still).

    We observers can only see what happened afterward, ex post. The fact that she left him reveals that her sunk costs were indeed irrelevant.

  5. Aco, say that you're the man who has experienced the sunk cost. Have you ever considered to replicate your situation by doing a same thing but the other way around to the other woman? Meaning that you get your sunk cost back from that woman. Kind of portfolio replication.

  6. Anymatters, you mean if I were about to approach someone else? By its very definition, sunk costs are sunk... They are not recoverable.

    But when I were about to approach someone new, I would consider all the expected benefits and costs. But then they are not sunk costs. At least not yet. Sunk costs are sunk only after they are incurred.

  7. Basically, the larger our effort to maintain a relationship, the more sunk costs incurred, and the larger the loss if the relationship ends up amounting to nothing. So, to minimize the loss from failed relationships, we should cut down on the sunk costs as much as possible, which means we should cut down on the "costs" of the relationship, i.e. our effort to maintain it (dressing up, eating out, buying gifts, etc). But, the lower the effort, the larger the likelihood of the relationship failing.
    Conclusion? As a single 20 year-old still very much in the dating game, I'm screwed, aren't I?

    for Anymatters: I would love to find a girlfriend who would be willing to incur all the costs while I sit back and relax, so that I can regain all that I've lost from previous relationships. But I'm not sure one exists.

  8. By the way, wouldn't the answer to the "Why are you leaving me after all I've done for you?" question be more like "Because those were your costs, not mine, and thus are not relevant to my calculations"?

  9. Aco, sorry I've given a different meaning. If it's sunk, then nothing. We can only replicate the situation with hime/her in different time in the future maybe. Such as, asking your ex become your custormer/client or business partner or whatever meaningful relationship now, considering the enjoyment and happiness in the past. The different is you're quite taking more benefit.

    Fik, I think in the feminist society whose the women have their own source of income or capitalist whose the women have their own capital. Sorry, I'm not sure if it exists but you'll find one who doesn't really care your money but love and guardianship.

  10. Fik says: "we should cut down on the sunk costs as much as possible".

    No you can't do this, Fik, because sunk cost is an ex post thing. Before it happens, it is not sunk cost. Remember the car analogy.

    On your question about her verdict, she was referring to his sunk costs, not hers.

  11. While the costs are sunk, can they be considered as investment or (capital) expenditures that can be depreciated?

    That would encourage her/him to find a better return, quicker break-even, or greater IRR - somewhere else. Not good for true objective of pacaran itself...?

  12. Bleu, it's perfectly fine to treat dating expected costs as investment. But they are "irrecoverable investment" - yes, sounds like an oxymoron.

  13. Tirta,

    well spotted!
    Following prospect theory, much evidence from psychology and neuroscience have suggested that people's decisions are often made against those predicted by economics theories such as expected utility.

    And so we, representing the psychology camp, will keep questioning if these rational cost-benefit calculations that economics have suggested for decades really apply when we examine the human mind and behaviour at the individual (one person) level, or in a dyadic (two-persons) interaction.

    I am sceptical that even an economist would actually weigh-up issues in pacaran this way if they were HIS own issues?

    Why use the framework if it may not work at this level?

  14. aco,

    i can see how this analysis is interestingly applicable to the issue of pacaran, but i can't help myself with this following quibble:

    if no mind works this way, then how can what happens -- according to the sunk cost framework -- be 'actual' (referring to your response to my first comment)?


    frankly, i don't see any tension between economics and psychology here. regardless of the assumptions and framework, i can't see any fruitful prediction about pacaran offered by both camps.

  15. Aco:
    Along the same line of Tirta's and Irma's comments, that's why I asked whether it's "are not" or "should not be".

    Although economic rationality normatively suggests that sunk costs should not matter, in our daily lives, it plays a big role. From a utility-maximizing point of view, this surely seems irrational (I don't know if it is).

    But haven't all of us (except maybe some very hard-core economists) regretted our past spendings on things? If this is so, Tirta's question above is spot on.

    PS: Her decision to leave him might not have disregarded sunk costs after all. The average cost of staying might be greater than the marginal benefit.

  16. Tirta and Arya, I see now. I read again my post and I think what triggered Tirta's question in the first place was this sentence: "What would you think she had in mind, as factors affecting her decision? Yes the costs and benefits of leaving him".

    Thanks to your question, I've been thinking and well, I guess now that my quick answer to your first question was wrong. Because the more I think about it, the more convinced I am that actually I was .. right.

    I mean, the moment somebody makes a decision, he or she has engaged himself or herself into a simple benefit-cost calculation. The reason why this is so controversial (and even some attack it as unethical) is because lots of people(even economists) always think about "benefits" and "costs" as represented by some money amount. Nothing can be more wrong.

    When somebody is seeing a date, any decision he/she makes also comes from a b-c calculation. Otherwise, why don't we see a guy gives her girlfriend everything so as to leave him with nothing at all?

    Arya, yes, I've seen regret modeled as a component of non-rational behavior. But I think it's a matter of finding another equilibrium when you expand your local neighborhood and realize that the decision you have made was suboptimal. As for your p/s note, are you comparing average cost and marginal benefit? How's that?

    Irma, I know nothing about psychology, so I'm sorry I can't contest between the two. Yes, I've heard about prospect theory and all (in line of Kahneman's or Tversky's works), but all I can see is that psy and econ complement each other rather than compete -- I might be wrong. Would be helpful if you can elaborate more for us here about the "much evidence in psy and n-science against econ predictions" if any.

  17. aco: "The reason why this is so controversial (and even some attack it as unethical) is because lots of people(even economists) always think about "benefits" and "costs" as represented by some money amount. Nothing can be more wrong."

    I always find this kind of argument is not useful. If everything under the sun can be measured, quantified, and calculated as c-b, then you can explain everything. Then it becomes really boring; there's nothing to discuss because you already know everything.

  18. aco,

    "...the moment somebody makes a decision, he or she has engaged himself or herself into a simple benefit-cost calculation...When somebody is seeing a date, any decision he/she makes also comes from a b-c calculation."

    are you saying there is no room for instincts, impulses, knee-jerk reactions, and emotions? even for pacaran?

  19. This comment has been removed by the author.

  20. Aco:
    Comparison of average cost and marginal benefit implies the role of "memory".

    Marginal cost exceeding marginal benefit is not the only explanation for why the girl leaves guy: It may also be that the average cost (which is always higher than the marginal cost) is greater as well. An outside observer cannot infer this decision is due to MC > MB.

    On regret: Now, this is why I was saying that this is a normative model (what should be done), instead of a positive model (what is is). Behavioural economics suggest how regret affects decision -- if a person loses a $20 ticket, s/he is less likely to purchase another ticket than had s/he lost a $20 cash previous to buying a ticket, and so on.

    It appears memory matters in decision-making. Whether, optimally, it should is a matter of debate.

  21. "If everything under the sun can be measured, quantified, and calculated as c-b, then you can explain everything"

    Roby Everything with price can be measured. Price being the existence of alternative(s). But even if all can be measured, we have not explained everything -- the world is not just about benefits and costs. I was saying that you can predict someone rational's behavior using a cost b-c approaches. But someone rational's behavior is not the only thing you want to know about the world. To date I still hate myself because I don't know anything about quantum physics. But I know how to apply a b-c approach to some projects.

    P/s. Thanks for the link.

  22. "Are you saying there is no room for instincts, impulses, knee-jerk reactions, and emotions? even for pacaran?"

    Tirta, of course there is room for all that. Find any advanced micro text and you'll find it easily. And guess what, they are explainable and still within the realm of b-c framework -- call it extended if you like.

    But let me again stress it here that economists assume rationality not because they believe that irrationality does not exists, but because they can safely predict the behavior of people who are rational. Even if from 100 persons there are, say, only 50 rationals, at least we can predict half of that population. The rest, maybe the field for other discipline (psychologists?). So far, economists have been trying to relax the rationality assumption to accommodate impulses and all, bit by bit. Not all successful, of course.

  23. Arya, if AC >> MC and MC > MB then the latter inequality suffices as an explanation. Yes, we don't observe this, we predict it.

    Thanks for the explanation on regret (and hence your point of normative-positive). Got it now.

  24. Anyway, economics is exact science while pacaran has its definite goal. So, cost-benefit analysis is still appropriate. once the couple enter the wedding prep, they may incorporate their analyses.

    With basic concept of money consciousness and from a holistic point of view, money is thought often of as energy that is not a thing to be possessed but an action to transact, transfer and exchange. The psycho-spiritual view of this concept is how human being uses money for purpose and finds satisfaction by not looking at the amount but through assessing the positive things affected when achieving, aligning and balancing the goals.

  25. aco, i have no problem with the working assumption of rationality, at all. also, i find your referring to any textbook for an extended account of those non-cognitive stuff (emotion et al.) very interesting -- so will try to hit those books.

    however, being ahead of myself reading the books, i would be quite dissapointed should emotion et al. be satisfactorily explained by the b-c framework. because it means we don't even need those concepts in the first place, for any human-like decision, orthogonal to the rationality-irrationality assumption, is just a matter of b-c computation ('cognitive', in the word of psychologists).

    to date, i believe that emotion et al. are primitive concepts (i.e. non reducible to any other primitive concept such as the b-c idea). see if i change my mind per your suggestion :-)

  26. Aco:
    You previously wrote:

    "We observers can only see what happened afterward, ex post. The fact that she left him reveals that her sunk costs were indeed irrelevant." (10/10/07)

    We observe a phenomenon (she leaves him). Does it mean that MC > MB? Maybe. But, if memory and her sunk costs matter, the decision may be driven by AC > MB, although MC < MB. Her decision to leave does not mean that her sunk costs are irrelevant.

    Suppose we redefine:
    MC' = MC + regret(her-sunk-costs).

    You can also change the sign and redefine regret as -joy(his-sunk-costs). In the latter case, she will leave if her MC >= MB + joy() [average benefit?].

    The point being that her leaving reveals little of her considerations of her (or his) sunk costs. It only does so if we impose the strict (economic) rationality condition on her.

  27. Arya, we do not observe the benefits and costs incurred to the decision maker. We just observe what happens after the decision is made and an act is done. Then, confronted with similar problem, we use that knowledge to predict what is going to happen. That's what we do.

    On your MC/MB/AC analysis, I guess it is similar in apporach to the way public economists extend private MB and MC into social MB (SMB) and SMC.

  28. Tirta, if you're interested rational addiction literature (Becker, Stigler, etc) is a good read - sorry I don't have the complete ref handy now, but you should be able to locate them in JSTOR. Currently, a book by Bryan Caplan (The Myth of Rational Voter) is addressing the issue again in the context of voter rationality (or lack thereof) -- I haven't got the book, I just read some reviews and comments on it. For more technical approach, there's an ample lot of lit on bounded rationality -- some of which was criticized too. Of relevant read, please find Ariel Rubinstein's works on the pluses and minuses of behavioral economics.

  29. Anymatters, thanks for the link! Psyco-spiritual -- interesting!

  30. Hmm....if only the game had been kept at 'dating' and not 'pacaran'...there will be no sunk cost *wink wink*

  31. Interesting analysis of "dating". But I have a problem with your analytical approach. I am sure you know that the concept of "sunk cost" refers to "investment" decision, right?

    It's the relationship between sunk cost and investment that makes me wonder if you are serious in using investment principle to explain "dating" process. Is dating really is an "investment" activity? My view it's not!

    If you seriously think, and insist that, dating can give you some forms of "returns", and you can measure them, perhaps you elaborate more on your approach and educate me. I will congratulate you on and then happily rest my case.

    Otherwise, please get serious: dating is simply one form of social activity and never perceived as investment, because normal dating doesn't follow "cost minimization principle" (well, unless one of them was a "stingy bastard" and use his girlfriend for his own advantage. But then, the analysis should proceed through utility maximization principle, rather than cost minimization principle). What I mean is: your dating expenses are all simply referred to daily expenses.

    Having said that, if a guy is on the verge of cutting off his relationship with his girl, my advice is not to panic or even trying to justify himself. Just simply say, calmly: "sorry babe, I like you, you are great, but I don't love you anymore..."