Thursday, June 21, 2007

Moral vs. Monetary Incentives

A few people that (thankfully) actually read what I write have asked why I specifically mention the potential effects of "insultingly small" monetary incentives. In general, I make the point that a small monetary incentive can crowd out an intrinsic moral incentive. I am careful to limit this assertion to small incentives, since economic research has shown this to be empirically true only on a small scale. This doesn't mean that there is evidence that a large monetary incentive doesn't crowd out a moral incentive, it just means that the large incentives haven't been tested, probably due to resource (or fairness) constraints. Consider the following example:

Gneezy and Rustichini (2000, JLS) performed a field experiment with 10 day care centers in Haifa, Israel. These day care centers were having problems with parents picking up their children late, which (to me at least) is not surprising since there was no specific penalty for doing so. In order to combat this problem, the day care centers instituted a fine of 10 Shekels per child if a parent arrived more than 10 minutes late. Economic theory would obviously suggest that, since the price of picking up a child late has increased, there would be less of that activity. However, what the researchers saw was that more, rather than fewer, parents started picking up their children late. It is also important to note that this higher level of tardiness persisted even after the fine was taken away (at least for the period that the researchers observed.) The authors of the study make the claim that this evidence is consistent with the crowding-out of the original moral incentive.

The authors of this study are careful to point out that the fine is "relatively small but not insignificant". They go on to mention, for example, that an illegal parking fine is 75 Shekels, and the failure to pick up dog droppings results in a fine of 360 Shekels. (I know, I know, I just coldn't resist adding that one in.) Furthermore, a baby-sitter earns between 15 and 20 Shekels per hour. My hypothesis is that the economically expected behavior would have been observed had the fine been larger. The parents likely took the fine as an approximation of the inconvenience to the teachers, and may have realized that they were previously overestimating the inconvenience to others.

As a side note, I really like the signs in Vermont that say "DUI: You Can't Afford It". Most states post signs giving specific dollar amounts for various offenses, but I think that the fear of the unknown is a much better incentive, if for no other reason than it prevents people from making an explicit cost/benefit comparison.

For a more thorough treatment of the crowding out effect, see here. (Note that the authors of this site describe a "significant monetary fine" in the day care example, which I feel implies the fine to be larger than it actually is.)


Charles said...

I am curious if outcome would have been the same if the incentive was changed from a fine (negative) to a discount (positive). On the surface, this should at least eliminate the psychological explanation that the teachers were "paid" for the inconvenience of staying later, which was never true in the first place. The teachers were never paid extra for the fines that were collected or the additional time they stayed, but this was never conveyed to the parents unless they asked, which I assume none did.

Would there be a change in behavior assuming that the discount was small like the fine and adjusted so that the daycare does not become unprofitable giving the discount? I wonder if a crowding out effect would be in play here as well. Would the parents still feel that they were paying for the extra time even though it was not a fine?

Charles said...

One of my comments might be confusing. When I wrote that the "pay" was never true, I was referring to the fact that the teachers were not receiving any additional income, not the feelings or perception from the parents.

Angad said...


I don't understand the logic behind giving a "discount" in such a case. The whole problem is that parents are "late" in picking up their children not "early."

It is understandable that a fine helps parents rationalize that since they are paying an extra amount it is okay for them to be late to pick up their kids, rather than the "intrinsic morals" that made them come relatively earlier before the fines were in place because they internally they could have possibly thought it was wrong to free-ride.

Giving them a discount is like paying someone to take a free-ride. Can you please elaborate on what you mean?

"On the surface, this should at least eliminate the psychological explanation that the teachers were "paid" for the inconvenience of staying later, which was never true in the first place."

The intuition behind the explanation appears to me is that it is an inconvenience (kind of cost) to keep the children longer, so the researchers can cook up any other explanation and give it to the parents. It is a legitimate economic problem rather than a psychological one in this context.

Did you mean to ask what whether the parent reaction would have differ having received a different explanation for the fine?

If that was the question, I don't think it would have made a difference as long as it is replaced by something reasonably plausible.

Angad said...

correction: I meant,

"Did you mean to ask whether the parent reaction would have differed having received a different explanation for the fine?"

Apologies for all the typos.

Charles said...

In theory, you should be able to use negative and positive incentives interchangeably to produce the desired behavior, which is for the parents to pick up their children by closing time.

The scenario I describe is where parents who pick up their children by closing time will receive a discount on the service. Parents who usually pick up their children late now have a monetary incentive to pick up their children on time. People who usually pick up their children on time will now have an incentive to continue doing this. This scenario might not be efficient or practical for the owner, but this is not what I am trying to observe.

In the version of the paper I have been able to find, the authors make several references to penalties and rewards. To me, they seem to imply that they are interchangeable and therefore, a reward might produce a similar result as the penalty in this specific case. However, they never actually tested this scenario and I am curious as to what the result might be. Will you see the theoretical result? Or will you see the crowding out effect again? Or will you see something completely new? What will happen when you take away the reward? This scenario might not produce any interesting data or information.

The scenario will also only be for a small reward and is comparable to the small penalty. If I made it such that the daycare is free if the parents pick up their children by closing time, I am fairly certain that I will see the theoretical result, which is the same result you would see if you made the fine very large, even though it is impractical.

Angad said...

Thanks for the clarification. Discounting makes good sense to me now, though I am still uncertain about its effects.

Charles said...

Everyone likes to cite this paper. I found the following from a link on the Freakonomics Blog today. The link is to Gelf magazine, which has "an interview with the economist Uri Gneezy, who conducted a 2002 study on the economics of splitting restaurant checks." The results are not that surprising, but it's something that everyone can relate to.

Anonymous said...

Good evening

We do not agree with this year BRITs 2010 decision.

Please attend our little web poll

Lady Gaga can not be better than Nina Hagen

Poll supported by BRIT awards 2010 sponsor femmestyle
[url=]brustvergr├Â├čerung preis[/url]

Do you have a burning question we could ask all the stars at The BRIT Awards?

Anonymous said...

My brother recommended I may like this website. He used to be entirely right.
This submit truly made my day. You can not imagine just how much time I had spent for this info!

Here is my website; Bmi chart for men

Anonymous said...

[url=]windows 7 activation key[/url] overprotective to say the least. [url=]windows 7 product key[/url] Pwbumysyj [url=]windows 7 pro key[/url]
btnfuc 901414 [url=]robe de cocktail[/url] 841769 [url=][/url]