Just A Handshake

“Between you and me, my friend, a handshake is enough.”

A recent economics journal article says one might reasonably avoid complex formal contracts to show you trust your associates:

This paper shows how the fear of signaling distrust can endogenously lead to incomplete contractual agreements. We consider a principal agent relationship where the agent may be trustworthy (dedicated to the project) or not. The principal may trust the agent (i.e. have a high belief of facing a trustworthy agent), or distrust him. The proposal of a complete contract, including fines and other explicit incentives, is shown to signal distrust. When trust is important in some non-contractible part of the relationship, a principal may prefer to leave the contract incomplete rather than to signal distrust by proposing a complete contract. Contractual incompleteness arises endogenously due to asymmetric information about how much one partner trusts the other side.

There are literally hundreds of papers out there showing how signaling can or does explain various details of human behavior.  In fact, fifteen years ago my Ph.D. thesis advisor tole me not to write such papers, because there were already so many of them that they weren’t very interesting.

Yet people keep complaining everytime I mention a signaling explanation of something, that I’m too free with such explanations.  So I’m stuck between an academic discipline that considers such explanations too obvious to be worth publishing, and an audience that finds them too implausible to believe, even when backed by such publications.

GD Star Rating
Tagged as: ,
Trackback URL:
  • I think the reason is that people do a lot of signaling without explicitly intending too. So when you say “signaling explains why people do X”, it sounds like “people do X because they want to signal Y, and they have explicit decided to signal Y by doing X”.

    Also, when you have this situation where signaling appears to be the “real” explanation for why people do X, but to the people actually doing X don’t believe they’re doing it to signal, the signaling explanation isn’t really satisfying, because it doesn’t do anything to explain why people don’t know they’re signaling. It’s sort of like explaining that the sense of smell works because in the ancestral environment the critters with acute smell were less likely to eat something poisonous. It’s sort of a valid explanation, but it doesn’t explain how the nose actually works.

  • Grant

    There is that Larry, but sometimes there are signaling and non-signaling reasons to do something. Sometimes the non-signaling reasons seem more important than the signaling ones (at least to me; e.g. health care consumption).

    The abstract Robin posted seems reasonable to me. Provided the people involved in a contract expect to work with each other again, I would think there would be positive benefits to signaling trust.

  • Andy

    I haven’t read the paper, but the abstract sounds like something that’s been known for at least the 15 years you mentioned. Baker, Gibbons, and Murphy’s 1994 QJE paper make the point that enforceable contracts can be welfare decreasing. It’s not really a signaling argument though, just people having rational expectations on trusting each other.

  • Jess Riedel

    I think the situation is very similar to the “just so” stories of evolution. Even if we are all confident in the theory of evolution–and even if it is backed by mountains of data–that doesn’t mean we should accept any explanation of a phenomenon which traces back to selection pressures. It’s extremely easy to say “women prefer manly, rugged faces because strong rugged men were better able to provide for them in the ancestral environment”, but it’s also very easy to say “women prefer feminine, delicate faces because caring, delicate men were more likely to care for women in the ancestral environment”. (And, of course, women actually prefer feminine faces, but I’m not sure if anyone knows why). And if you can explain any data, it’s not worth anything.

    Very often I’ve read posts here where a signaling explanation is given for particular behavior when there are many alternative explanations which seem more plausible (at least to me). Perhaps this is just an example of my ignorance, but in that case you may simply need to educate your audience.

    And again, this doesn’t mean I necessarily doubt that signaling drives a large fraction of human behavior.

  • If you expect to work with someone again, it is especially important to be clear up front about exactly what is expected in an agreement. You need a clear written description of what is to be done and when, and how much and when you will be paid, even if it is not an actual contract.

    Failure to be clear about these is the most common cause of problems I’ve found in remodeling and landscaping work – and I would expect similar problems anywhere you are doing significant work before getting paid, or paying out significant money before getting the work done. Pay as you go, getting paid, or paying, as the job is done, usually by the week, is different in that you can adjust expectations or work as you go.

  • Psychohistorian

    I don’t really see where people find signaling explanations too implausible to believe, except where there’s neither clear evidence nor clear reason showing signaling to be a primary motivator.

    I totally agree with this example. In fact, I’ll bet if you asked someone, “Why not make breach-of-contract more explicit?” He’d says something about it insulting the business partner, or not wanting it to be so impersonal. (I’d bet that’s the motivator behind the signaling – you want business relations to occupy more of the social sphere than the “money sphere,” and too many lawyers make it feel like impersonal “money.”)

    On the other hand, calling sleep a signaling behaviour seems to stretch the meaning of the term a wee bit far. Just because it explains a whole lot of human behaviour doesn’t mean it explains all human behaviour.

  • diogenes

    I basically agree with Jess and the other posters here. The fact that “signaling” explains a part of human behavior is boring and irrelevant. You can make up many “just so” stories in evolution — and this is a gigantic and known problem in field of animal behavior in general. A medical equivalent would be suggesting disease X has a “genetic component”. This has absolutely no value — everyone knows this already (all disease have a genetic component, all human behavior has some signaling component).

    How much variance does the genetic component explain — or in your case, how much variance does human signaling explain in every day decision making. The references you post just prove that signaling plays *a role* — they really do a poor job of fleshing out how much variance signaling plays in every day decisions.

    As many other people mentioned, a lot of the activities where you describe a signaling explanation — there are many other very plausible alternative explanations. You really need to have evidence showing that the signaling explanation is the dominant source of variance (or even a significant source) — which doesn’t seem to be forthcoming.

  • Jaffa_Cakes

    “Between you and I, my friend, a handshake is enough.”

    Where is that from? I have been looking at it for a couple of minutes and I’m pretty sure that grammatically it should be “Between you me, my friend, a handshake is enough.”

    The handshake is the subject and “you” and “me” are just objects in the sentence, aren’t they?

  • Larry, I agree that we eventually want explanations at all levels, and that we especially want to explain why folks aren’t aware of signaling.

    Jess, yes it can be hard to determine what causes what; all we can do is try.

    Jaffa, I made it up, and may well have the grammar wrong.

  • Stephen

    I don’t think anyone wants to explain everything with signaling.

    But it would help if signaling was considered before embarking on a search for brand new explanations. That is like reinventing the wheel, with corners.


    “you” and “I” are objects of the preposition. Prescriptively, “I” should be “me” and the entire noun phrase could be reduced to “us.” Conversationally, nobody cares.

  • jess:
    how masculine or feminine a face women prefer varies over the course of their cycle. women prefer more masculine faces while fertile.

  • Cyan

    … an audience that finds them too implausible to believe, even when backed by such publications.

    It seems to me that some pointers into the literature are just what your audience needs. You can hardly blame them for finding signaling explanations implausible when (i) humanity’s very nature blinds us to the signaling we’re doing, and (ii) they’re unaware of the data, and thus cannot update to where you’re at.

  • Chuck

    This is another example of the paradox of this site.

    For humans, overcome bias is like overcoming oxygen dependency. It is fundamental to who we are.

    Post-human is a really useful concept.

  • John Maxwell IV

    Do any of those hundreds of papers have experimental data to back up their claims?

  • Pingback: Interessantes woanders (2009.05.27) › Immersion I/O()

  • Pingback: Les liens du matin (48) « Rationalité Limitée()

  • Signaling models do suffer from the ‘just so story’ problem, which is one reason that they are not all that popular in mainstream economics anymore. I am not sure how much sense it makes to rely heavily on them in more speculative areas, like human social behavior.

    I was inspired by this post to write my own thoughts here about the decline of signaling models in finance and accounting…Ron Dye of Northwestern wrote a comment that alone is worth clicking on the link.