More Evidence That Paying for Grades Isn't Easy

As you may have read on this blog, the economist Roland Fryer has done quite a bit of research on bribing kids — i.e., offering financial rewards for good grades. A new working paper from Josh Angrist, Philip Oreopoulos and Tyler Williams examines the effect of financial rewards on performance among an older cohort: college students. First- and second-year college students in Canada were offered cash rewards for grades above 70 and were also contacted by upperclassmen advisers “trained to provide advice about study strategies, time management, and university?bureaucracy.” The authors found that “[t]he intervention increased the number of courses graded above 70 and points earned above 70 for second-year students, but there was no significant effect on overall GPA. Results are somewhat stronger for a subsample that correctly described the program rules.” Their findings are consistent with previous research indicating that the incentives are less successful for older children. [%comments]

Drill-Baby-Drill drill Team

One of the more progressive programs for bribing for grades was the HOPE scholarship in Georgia began in 1993. If a student maintains a 3.0 GPA in college they get FREE tuition at a public institution plus fees and allowances for books.

It encouraged not just enrolling in college but achievement through college. It has funded nearly a million scholars.

I would like to know the long term achievement of these graduates compared to let say Alabama. or South Carolina where they do not have such a program.


The financial incentive caused them to shift their utility and spread it more evenly, rather than increase their utility. The college students decided that two C's were worth more now than a B and a D. I would think that, although this doesn't increase overall GPA, this is a desired outcome.

Michael Radosevich

Paying students for good grades does not work. Paying higher taxes for better public schools works.

How many "failing schools" are in wealthy school districts? None. There was a recent study involving California public schools, and not a single school in a wealthy district was a "failing school".


The fact that students are being paid for getting grades above 70 causes them to spread out their time studying for all courses instead of mostly one. If a student used to get really high grades on one course and really low on another he or she now focuses on both in order to receive the money. The marginal benefit of receiving the money is larger than the cost of changing the way of studying in order to get grades above 70's. This can be seen because the students are choosing to change their study habits. The reason why this works better with younger children is because they need the money more, when older there are other ways to get the money and other things such as spending time with friends instead of studying become more important. So, the marginal cost of changing study habits for younger kids is higher then the benefit of the money.


One possible explanation for the higher success rates in affluent schools than in poor ones is that in neither case is the school actually providing an adequate education in and of itself, but in affluent areas, parents provide so much additional education that the failure of the schools is masked. Between help with homework, private tutors, various kinds of enrichment activities, as well as the normal reading aloud and going to museums that affluent parents provide, parents are essentially home-schooling in their spare time.

Additionally, a story on public radio today talked about the difference in sheer volume of words spoken by middle class parents compared to poor parents to their children, which accounts for the vocabulary that kids arrive at school with. Kids in poor schools have many disadvantages that are not easy to overcome with more money paid to the school itself.


@#3, let's not continue to blindly pump more money into the abyss that is public education. Your anecdote, assuming it is even true in the first place, doesn't account for a number of factors beyond tax dollars, such as parental involvement/oversight, education level of parents, etc. Correlation isn't causation.

The problem with these studies about financial incentives being insufficient motivators is that in many cases the carrot simply isn't big enough. Offering a college student $100 isn't going to get you very far. Upwards of $1,000 may show a different outcome.

Bindi Papadum

@michael - you are making an assumption that would make the freakonomics authors smirk (correlation vs. causation). You assume that wealthy school districts have better schools because of higher taxes, but there are more likely causes for this. For example, the (relatively) higher education (and ambition) levels of the parents is a better predictor of good schools than tax rates. The wealthy school districts have more smart, ambitious parents, and this raises the level of normal achievement for the kids and schools. Tax rates don't cause high achievement, smart parents do.



I wonder if it might be more effective if students were paid incrementally. For example, pay students $5/hour to 10 hours tutoring sessions per week. As we know, students are procrastinators, so the thought of earning $700 16 weeks down the road may not motivate them. But, if they could earn $50/week attending tutoring sessions starting the 1st week of the semester, this immediate payment will entice them to attend. Slowly they will replace the hard-to-kick habit of "not studying" with the good habit of "studying" and their performance would increase.

Leon Palafox

Wouldn that be related to the fact that older kids, specially college students, are supposed to be on a creative environment.

College is full of creative tasks, and thus giving incentives to get good grades might focus the students on the money rather on the imagination needed to solve the task at hand, thus utterly failing.

Elementary school and such, needs little imagination or actual brain work, to get good grades you just need to do mechanichal works that work well with motivations such as economic ones.

Mike B

Incentives break down if there is no guaranteed outcome. I think there is a misconception that people can just go out and get good grades. If someone lacks natural talent...and in this case it is the sort of book smart talent that translates into consistent good grades, no incentive will matter because the payout is far from guaranteed.

If you were to offer me $100 to get an A, but if I know that if I spend hours and hours staring at a textbook I'll still only have a 20% chance of getting an A the expected payout is only $20 and with cognitive biases factored in it might as well be $0. School performance is not something a child necessarily has all the tools to succeed in. Offering someone an intensive to do something that they literally have no ability to do will probably only result in depression or other negative outcomes. These systems need to go after things that the child does have in their power to affect, such as attendance, book reading, participation, etc.



My daughter had a lower grade point at university, but it's not because she didn't work as hard or because she is less intelligent. She took one of the most difficult and demanding degrees out there.

During her time at university, she had many friends who had been awarded contingent scholarships (ie they would lose them if their grades dropped.) These students invariably shied away from taking anything difficult and filled their schedules with "basket weaving" classes.

This is the perverse effect of this financial incentive. Even if GPAs did appear better on paper, the quality of education received was definitely worse. It rewards students for taking watered-down degrees.


If there was no improvement in overall GPA, it would seem that these students, in order to obtain the reward, worked harder in the particular class...and let another one slide. It's almost as if a student can (or will) only do so much. He can do enough for an A in 4 classes and an F in the fifth...or he can perhaps make a B in three class and C's in the others.

I know in my first term in college, I took a full load. I made A's in all courses but one: English--in which I made an F (thank you, Professor Leochler--the teacher who took me from F to A!). From that point on, I avoided full loads. I just didn't have the time, energy, firepower, intelligence, or commitment--something!--to make A's in all courses.

Maybe the key is to make students EARN the right to take a full load. That is, ONLY those who excel in three classes can take four; etc.

Annoyed Teacher

@off-topic taxation and school funding thread

Zip code predicts success. It has much less to do with the schools in that zip code than with the parents there. Unfortunately, holding parents accountable for their poor parenting, often stemming from an unfortunate socioeconomic status, isn't tenable politically or ethically. It is still correct.

On topic,

Paying students who are already well-educated enough to be in college doesn't work, and that shouldn't be surprising. Setting the bar as low as 70 isn't going to improve average GPA. Many programs won't accept grades below C for people to graduate with a particular major.

Only students who would be getting terrible grades are going to change their behavior at all for this program, and if they were getting bad grades at the college level before, why would they aim much higher than what the administration deems laudable - 70 in everything.



Reading some of the comments, one might conclude that all students have the same talent and only teachers, parents and hard work account for the difference in grades.

It is at least worth considering that some students are more talented than others and that any incentive program must be based on an improvement that seems achievable to that student rather than an arbitrary one such as achieving 70%. If, for example, you offered an incentive for a student to bring her History mark from 75% to 80% while maintaining her 90% Math mark, you might see some impact.

For any incentive to work, the incentive has to have more perceived value than the effort to achieve the goal, and the goal has to have a reasonable probability of success in the eyes of the incentee. I think I learned this in economics class long ago.

John C Bower

Reward or financial incentive is a form of recognition of student progress/achievement. Often humans need evidence of their ability to move towards competency in the subject matter before intrinsic motivation can take hold. Evidence is comes through parental recognition, peer support, self awareness and community awareness.

Spending resources automating relevant recognition systems that are integrated with the LMS's and SIS's will make it easier to offer students frequent, consistent proof of incremental progress toward their academic goals.

With the growth in online learning, relevant feedback loops that promote communication and social learning become even more important to encouraging course completion and persistence.

John Bower
CEO, uBoost


This is a product of a sad, yet realistic truth. Money evidently serves as a better motivator than personal success. Clearly, students get more utility from getting paid for their work than if not. This incentive offered is a good way for students to figure out that the amount they are getting paid is greater than the opportunity costs they pay by studying harder and showing more dedication towards school.
However, an interesting fact is mentioned here where bribes or incentives work more with younger students. I place this reasoning on the contrast of ethical values between young minds and older ones. Younger students do not receive many opportunities that involve them getting paid, so when the situation comes, it is a no brainer for them. On the other hand, older students are more wise and mature when making this decision. Not only do they have a greater sense of honesty and self-respect.
Evidently, the opportunity costs of older students is higher than that of younger students, which is why most decide to pass on the bribe. They have more opportunities involving payment elsewhere.



Isn't it possible that they simply aren't offering college students enough of a financial incentive? For elementary school kids, $2 is a huge amount. At that age, if you're buying candy bars as big as your fist (the tiny ones to us adults) you can probably get 16-20 of them (depending on what's on sale) for that amount -- $2 is huge and buys a lot of happiness. Additionally, young kids really have nothing else to compare that.

Now, i can't read the article without paying for it, but I'd be willing to bet that to college-aged kids, especially those in the latter half of college, the amount offered won't be anywhere near the same scale as it was for elementary school kids.

So, I don't think it's the case that financial incentives don't work for adults, I think it's simply the case that nobody can afford to try with "real" financial incentives. $50? Nice, but spaced out over the course of a semester of a 3-credit class, that's like 30 cents an hour (presuming 3 hours of classes, 9 hours of study, 12 week semesters) -- chump change. $100? Even nicer, but over the course of a semester... Try offering a few thousand dollars and see whether or not that changes


Hunter Gardner

The first thing that comes to mind is the opportunity costs framework--essentially we are paying college students for going to class and studying, which takes up their time and effort. This is fine, but consider the gap between a grade below 0-69 (D, F) and a 70 (C) and furthermore one grade higher, 80 (B). In the first scenario, the two outcomes are either "get paid x" or "get paid 0"--the incentive is one based on getting paid at all, so students will work just hard enough to get paid something higher than $0, e.g. earn a 70. To really boost from say a C to a B, 10 points, this can mean a great deal more effort on the student's part, which means maybe paying them 2x or even higher.

p.s. Drill Baby Drill there is a HOPE equivalent in SC, its called the LIFE scholarship.

Christopher Strom

@6 and 7:

Maybe better school funding doesn't cause better performance, but I don't see any wealthy parents suggesting that their kids' schools should get the same funding as poorer kids' schools.

Put your money where your mouth is.

Regardless, I think paying students for grades is a bad idea as reduces learning to a fee-for-service transaction.


I honestly think us students get paid in the future our jobs are our payoff. If we chose to mess around now we get a job that pays low. If we do good and pay attention then we get our dream job if we get paid now its motivation but for the wrong reasons.