Money is everywhere. Our societies are run by it and the pursuit of money is to many people an integral part of life. An interesting question, given the great dependence modern societies have on this particular form of exchange, is, what does money actually do to us? There are two related literatures that connect to this question that have interested me, namely that on motivation crowding, and the by now highly contested phenomenon money priming. Given that these two literatures deal with a similar problem, it might be interesting to see how they stack up against each other. This post will touch on these phenomena. What are they and how are they different? How are they thought to work? And most importantly: do they really exist?

Such a discussion has to depart from an even more fundamental one, namely: why do we do the things we do, and for whom? As microeconomic theory traditionally posits, many of our motives are purely self-interested. We often judge the desirability of an action mainly by how we ourselves would benefit if it was taken (relative some alternative action). Another motivation may be to benefit others who are close to us, because we are happy when they are happy – one could perhaps call it other-regarding self-interest. And then, there are those cases where an action is deliberately taken with the motivation to benefit a complete stranger (or perhaps a group of strangers), and such an action is often called altruistic.

The effects of money on behavior that is already motivated by raw self-interest is pretty straight-forward, and not really interesting. The law of incentives is pretty solid in these cases: the higher the financial payoff of a certain behavior, the more likely one is to engage in it (with diminishing returns). But what about behaviors that are not solely motivated by self-interest? How do money factor in then? Money is, to begin with, merely a tool used to obtain certain goods – whether those goods benefit oneself or others. It is, however, by pure conditioning if nothing else, overwhelmingly associated with benefitting ourselves. If we were to add up all the occasions where we have used money for our own sake, and see how it stacks up against all the times we have used money for someone else’s sake, the latter pile would be dwarfed by comparison. This could potentially have consequences for how our pro-social behaviors are affected by money.

One phenomenon that has been looked into is often referred to as motivation crowding, and sometimes as over-justification effects (depending on whether we look at the behavioral economics or the cognitive/social psychology literature). An early overview of the phenomenon can be found in Frey & Jegen (2001). Some will undoubtedly have encountered it in a highly entertaining form in Levitt & Dubner’s book Freakonomics. In the book, the authors take as an example a study from Haifa, Israel, conducted by Gneezy & Rustischini (2000). The study looks at the effect of introducing economic incentives to ameliorate problems with parents picking their children up late (that is, after closing time) from the daycare center.

It turns out that when introducing economic incentives in the form of a penalty fee, parents were more, not less likely to pick their children up late. One interpretation of this is that when economic incentives were introduced, the decision to come late was transformed from something that was mainly based on concern for the social consequences (the staff will have to work late etc), to a simple economic transaction. Despite all the other consequences still being there, and worse – these was now also a monetary penalty for the parent – coming late was now instead perceived as something one could simply buy one’s way out of. In a second study by Gneezy & Rustischini, they tried giving monetary incentives to high-school students collecting money for charity, with the consequence that those students collected significantly less money than a control group who were not compensated at all. With the incentive, the calculus suddenly became “is it worth the money I’m getting?” instead of “am I doing something good?” and the incentive simply wasn’t large enough to compensate for the loss of the pro-social motivation the students had had before.

Put simply, we appear to be highly challenged when it comes to taking more than one aspect of a decision into account at once. The consequence is that when a new motivation is added, the former ones are easily forgotten – they get crowded out. This logic appears to work mainly in one direction: an added pro-social motivation rarely crowds out a self-interested motivation, but only vice versa. To give an intuitive example: imagine you’re sitting at a café enjoying a nice espresso. At the next table is seated some random stranger. The stranger turns around and says “sorry to bother you, but could you perhaps get me a glass of water?” Depending on who you are might agree or decline. Now imagine if the stranger instead would have said “sorry to bother you, but I’ll give you a penny if you get me a glass of water.” Would you be more or less likely to agree than in the first case? Chances are, you’d be substantially less likely to help, because a penny is simply not worth it – even though a penny is more than nothing at all, which is the monetary compensation in the first case!

This type of phenomenon has been shown in several studies, and although there are reviews of the existing evidence, there are (as far as I know), no high-powered replication efforts to check how reliable the effects are. As with any field, publication bias is likely an issue. We simply don’t know yet whether we should trust it (and there will be some more on that in another post).

So much for motivation crowding – what about the literature on money priming? A number of studies have purportedly shown effects of priming people (that is, giving them very subtle reminders) with money related concepts. The priming stimulus used in this literature can be, for example, a dollar sign screen saver on a computer in the corner of the lab in which participants are tested, or concepts related to money in word descrambling tasks. Studies have claimed that subtle reminders of money make people less likely to act altruistically, and more likely to endorse free-market policies, among other things. The thought is that merely having the concept of money somewhere in the back of your head – even without conscious awareness – will have profound effects on how you evaluate a certain decision calculus. It is a proposition that, certainly, is very exciting – if it is true.

The problem with this literature is that there are now large replication efforts that have essentially shown null results. At this point, it appears that money priming is either not real, or at the very least a very brittle phenomenon with very little real-world relevance. If it is real, the effects are certainly not as large as the early literature found, and are likely also highly dependent on very specific contextual constraints.

Given that these two phenomena both concern social effects of money, it is reasonable to ask how they stack up against each other. A priori, money priming seems to me to be less credible than motivation crowding, replications aside. Social priming is, to begin with, a highly contested phenomenon, and I think it is as of now uncertain whether very subtle reminders of anything can actually cause any measurable differences in behavior, one way or the other. This, I think, stands in stark contrast to the phenomenon of motivation crowding, where the stimulus is far from subtle. We have quite firm reasons to believe that tangible money rewards influence peoples’ behavior (it is, after all, the whole foundation of economics). It is therefore not unthinkable to believe that monetary incentives could also affect the motivations for our actions in ways that in certain situations could make their effects have the opposite sign to what we usually tend to expect from them.

I would argue that a reasonable way of making sense of these two separate literatures is that it is plausible to assume that money can also have negative effects on certain types of social behavior, but that the required stimulus is likely much stronger than that utilized in the money priming literature. That said, I would like to see solid, pre-registered replications of motivation crowding before making up my mind.