On paying children to learn

A true story about bribes, beliefs, and becoming self-motivated

Hey, it’s Harrison 👋 Welcome to my publication about creativity as a tool for personal and professional growth.

If you're a high-achiever ready to make a major life or career shift through a creative project, I offer 1:1 coaching to help you navigate that transformation. If you’re interested in how I can support you, learn more about me and then send me an email and we’ll set up a call to chat.

---

On paying children to learn

“Money is a good servant but a bad master.” — English Proverb

OK. Serious question: Would you pay your children to learn?

It’s a question that’s been on my mind since reading a parent’s fascinating review of the experimental Alpha School in the U.S., which claims that its students are growing 2.6x faster than normal while doing just two hours of academic work a day.

The parent explains that Alpha is achieving these results using a special “bundle” of components, which includes mastery-based learning, teachers turned into “guides,” high guide-to-student ratio (5:1), and an innovative AI learning platform, all of which the school shares proudly in its publicity.

But there’s one extra ingredient that Alpha isn’t saying anything about publicly: cash incentives. Alpha School pays its students to learn.

When I first read this, I was fascinated, but part of me recoiled. Paying kids to learn just felt…morally wrong. But the more I thought about it, the more I realised I didn’t have a clear, satisfactory reason why. So, this essay is an attempt to figure it out.

The secret ingredient Alpha keeps hidden

The student incentive system at Alpha works on multiple levels.

Primarily, students earn “Alpha Bucks,” about 10 bucks a day for completing their daily “minimums” (8-12 lessons), with bonuses for extra lessons, finishing on time, and reaching weekly targets. Each buck is worth 10 cents, so they can earn roughly $2 a day.

Importantly, kids can only earn bucks if they complete lessons AND score 80% or higher (not mere completion).

Kids spend their bucks at school stores on personalised prizes like bouncy balls, protein bars, LEGO sets, and Taylor Swift sweaters.

There are also “Dojo Points,” which are rewards for non-academic behaviour like teamwork, respect, and perseverance. And daily “Dojo Masters” win keys to potentially unlock prize boxes on Fridays.

Finally, there are summer programmes that pay kids $1 per lesson completed (10x the normal rate) to prevent regression during the summer holidays. This can amount to as much as $5/hour for 2 hours of daily work. For the kids, this is real money, not pocket change.

The results from Alpha are startling. The guy’s 8-year-old middle daughter, for example, in her first year at the school, “has completed more than two full grades in ~20 weeks (60% of the school year) and shows no sign of slowing down.” And the same lessons that were “a fight every time” became engaging once they were embedded in Alpha’s structure. His kids even ask to do extra lessons.

According to Alpha, 100% of its students meet projected RIT scores (a standard U.S. measure of student progress in reading and maths), and their growth rates exceed projections by 200%, a feat managed by less than one in a thousand schools across the United States.

There is certainly something remarkable going on at Alpha School. But I want to know more about these incentives. The fact that Alpha keeps them hidden, and the fact that I bristled at the thought of them, told me they were worthy of an essay.

The economist who paid kids to learn

Arguably the person with the most authority to speak on student incentives is Professor Roland G. Fryer, born in Daytona Beach, Florida, in 1977. A MacArthur Genius Award winner, a recipient of the prestigious John Bates Clark Medal, and the youngest black professor ever to be given tenure at Harvard (he was 30), Fryer has been described by his peers as “one of the premier empirical economists of his generation.”

In the 2007–2009 school years, Fryer ran four large trials in New York City, Chicago, Dallas, and Washington D.C. and his programmes spanned ~250 schools, involved ~36,000 students, and paid them ~$9.4 million in cash incentives.

Fryer’s personal mission was to close the racial achievement gap. In 2009, results from the National Assessment of Educational Progress showed that about 4 in 10 white eighth-graders were proficient in reading and math, compared with about 1 in 10 black students. Fryer himself flunked school and ended up in gangs. And it was only a talent for sports and an uncompromising grandmother that spurred him to get back on track.

As an economist, Fryer was interested in testing scalable ways to get ROI from what he describes as the “demand side” of education. There is already important work being done on the supply side of the equation (through institutions, educators, and pedagogy, etc). Fryer wanted to supplement this with solutions that centred on learners.

But it wasn’t just about supply and demand. Fryer was also driven to test incentives because he saw how gangs were incentivising many kids to go in the opposite direction, especially in poorer inner-city schools. “If you fundamentally believe that these kids have the same potential as your kids, and that they are rotting away in these horrible schools,” Fryer said, “then you should not be able to sleep without trying to make a change.”

I felt torn learning about Fryer. He wasn’t some privileged academic who was far-removed from the realities of public school life, or some corporate cowboy chasing efficiency. He was someone trying to rescue kids just like his younger self.

When you read his work and hear him talk, it’s clear he is left-leaning in his social sympathies. In one of his podcast appearances, he shares a story about how his grandma rang him on his first day at Harvard and demanded to know, “I don’t care how big your office is. Do you know the janitor’s name?”

I was surprised that such market-driven and economically provocative experiments could come from someone with his background. I was hooked.

What were these experiments exactly? How did they come about? And what could they tell us about children, motivation, and learning?

Cash, classrooms, and controlled trials

The seed of Fryer’s experiments began in 2003 whilst he was working with a school in The Bronx where he’d spent time as a 12-year-old. At the end of each week the school would tell him whether or not the kids had behaved and, if they had, Fryer would pick up the phone and order them pizza.

It proved so successful at enhancing student behaviour that in 2006 the chancellor, Joel Klein, asked Fryer if he could find a way to scale-up the pizza parties across the whole of New York City.

Realising that the logistics of city-wide pizza delivery wouldn’t work, Fryer pivoted to more scalable cash incentives instead and, after several small pilots, reached multi-school trials in the city districts of New York City, Dallas, D.C., and Chicago.

But it was no hop, skip or jump. He was thrown out of New York City three times (literally ordered to get in his car and leave mid-way through school visits). And furious parents and teachers even began picketing outside his house.

He persevered, and in 2007 Fryer and his team finally pressed go on their first large-scale student incentive programmes in New York City and Dallas, with D.C. and Chicago following shortly after.

Fryer’s first experiment was to pay kids for outcomes, such as test scores and grades.

In New York City, pupils were paid for interim assessments. 4th-graders (9/10-year-olds) received a $5 base rate plus up to $25 per test (maximum $250 per year), while 7th-graders (12/13-year-olds) received a $10 base plus up to $50 per test (maximum $500 per year).

And in Chicago, 9th-graders (14/15-year-olds) were paid for their report-card grades. $50 for an A, $35 for a B, $20 for a C, and $0 for a D or F, with earnings capped at $250 per report card and $2,000 per year.

While these experiments were underway, Dallas and D.C. went live. For these Fryer wanted to introduce some “experimental variation.” Rather than paying kids for outcomes, he paid them for inputs, things like reading books, completing homework, and attendance and behaviour.

In Dallas, 2nd-graders (7/8-year-olds) were paid $2 per book, with an average payout of $13.81 and a maximum of $80.

In D.C., 6th- to 8th-graders (11- to 14-year-olds) could earn up to $100 every two weeks (up to $1,500 for the year) for meeting targets in attendance, behaviour, and three performance metrics selected by the schools.

All of Fryer’s experiments were randomised control trials, meaning there was a treatment group that received the rewards and a control group that didn’t.

Fryer even created bank accounts for every student, which provided visibility on their saving and spending patterns.

All money was paid directly into students’ accounts except for the first payment; for this Fryer visited the schools and personally handed each student a cheque, partly because he wanted to meet them, and partly because he wanted to convince the doubters he was for real, especially the inner-city kids who were understandably wary—“You ain’t giving anything away for free!”

At this point, I wasn’t sure what to hope for. If the incentives failed, or made things worse, then it would be bad for the kids but at least I could leave my moral instincts intact.

But if the incentives worked? Well. I knew I’d have to examine some of my deeply held beliefs.

---

---

When the data and the gut disagree

There were essentially three big insights that came out of Fryer’s experiments.

I

First, paying kids for outcomes (good grades or high test scores) simply didn’t work. In New York, where kids were paid for their test performance, the scores barely budged. In Chicago, where 9th-graders were paid for their report-cards, there was a small improvement in GPA (the overall average of students’ grades), but nothing meaningful in their actual test results.

II

Second, paying for inputs rather than outcomes did make a difference. In Dallas, 2nd-graders earned $2 every time they read a book. On average, those kids ended up about 2.5 months ahead in reading comprehension compared to the control group. That’s huge! And for just $20 per child too!

In D.C., the programme was broader: middle-schoolers could earn up to $100 every two weeks for good attendance, behaviour, homework, and classwork. The results there were smaller but still good, a gentle improvement in reading and maths scores.

III

Third, the benefits faded over time, as expected, but they did not disappear completely. In Dallas, a whole year after the payments had stopped, the reading boost had only roughly halved.

So, why did the inputs work when the outputs didn’t? Fryer’s conclusion was that most kids don’t know how to “get good grades.” “Work harder” is vague; whereas “Read this book” is concrete. In other words, the closer the reward was to the actual skill being built, the better it worked.

When the kids got their money, they spent it on what you’d expect: ice cream, video games, shoes, etc. But many also saved more AND opened their first bank accounts, which Fryer saw as a healthy side effect of the experiment.

From an ROI point of view, the experiments were impressive. The cost of paying kids to read books was a fraction of what schools spend on other reforms that achieve similar gains, such as extending school days or high-dosage tutoring.

Fryer’s verdict was that paying kids isn’t a magic bullet. ”Will it close the racial achievement gap in America? No.” But his experiments had shown that well-designed incentives can have a huge impact on student achievement, and do so at a price that makes policymakers look silly for ignoring them.

That’s not how the rest of the world saw it, though. Even with the results of his experiments in-hand, Fryer faced a backlash at every step of the way.

The revolt against paying kids (values versus outcomes)

As the parent pointed out in his review of Alpha School, a 2010 Gallup poll showed that only 23% of Americans supported paying kids to learn, meaning that 76% were against the idea (1% were undecided).

“There aren’t many things that 76% of Americans agree on,” he wrote, “Only 69% of Americans believe another Civil War would be a bad thing. Only 78% agree that American independence from Britain was the right choice.”

What explains such deep-seated unease about incentives? Why was Fryer thrown out of New York? What was going through the minds of those angry picketers outside his house? And why did I find myself instinctively bristling at it?

After doing some journaling on these questions, as well as reading commentary from parents and looking at some of the psychological literature, there appear to be a few recurring themes in peoples’ arguments against cash incentives:

• “I don’t want my kids to love money more than reading!” This is a values worry. Parents are concerned that cash payments will shift the moral meaning of learning from culture and identity to commerce and transactions.

• “My kid already loves reading. Paying her is pointless. It might even ruin reading for her altogether!” This is a concern about withdrawal, and it is backed up by studies such as those carried out by Lepper, Greene & Nisbett (1973) where a bunch of kids who loved drawing were split into groups: some were promised an award for drawing; others got no award. Later, when awards were taken away, the promised-award group lost interest in drawing.

• “Rewards could stop my kids developing their own intrinsic motivation!” This is a worry about developmental pathways. I’ve written before about Deci & Ryan’s Motivation Continuum, with pure volition at one end and coercion at the other. The concern would be that kids become motivated by coercion alone, never achieving self-direction or identification as learners.

These are the predominant worries I picked up both in my own thoughts and feelings, and in the commentary around Fryer’s work.

A quick query of ChatGPT produced a list of 15 additional arguments, most of which were rehashes of the three arguments above, but some of which are definitely worth considering:

• Gaming worry (Goodhart’s Law) - “My kids might optimise for what’s paid (pages, easy books, short cuts) instead of for comprehension and curiosity.”

• Peer dynamics & status worry - “What if the kids start to compare themselves and become competitive over their earnings?”

• Equity worry - “What if unplanned top-ups and side-payments create new inequalities?”

• Slippery-slope worry - “If we pay kids for reading, then why not for kindness, curiosity, appearance, etc etc?”

• Norms-at-home worry - “Incentives do not sit well with our family’s values around money, chores, and learning!”

It was starting to become clearer to me why incentives were such a contentious topic.

After reading critiques from parents and seeing some generalised ideas from an LLM, I wanted to hear from at least one academic voice that was directly opposed to the idea of incentives and hear what his or her arguments were on the matter.

It didn’t take long to find the work of Alfie Kohn, an American writer and lecturer, born in Miami Beach, Florida, just two years before Fryer, who is best known for his books on human behaviour, education, and parenting (his ideas draw heavily on Deci & Ryan’s Self-Determination Theory).

Kohn studied at Brown University and the University of Chicago where he earned a master’s degree in social sciences. He’s not an academic researcher in the traditional sense like Fryer. Instead, he combines classroom experience with psychology, education research, and cultural critique, producing popular presentations and essays in a strong, polemical style. Kohn’s a big champion of progressive education.

Having written extensively on the topic of incentives, Kohn’s stance is that all external rewards (money, stickers, grades) undermine intrinsic motivation and real engagement. He makes the point, which I’d never considered, that rewards are punishments in disguise; for a reward that isn’t given merely becomes a punishment, and a punishment that is avoided functions like a reward.

If Kohn was critiquing Fryer’s experiments, he would likely say that rewards produce compliance, not curiosity. That even if kids read more and their test scores moved in the right direction, Fryer and his team were teaching the wrong lesson.

Put simply, for Kohn, and for many of the parents and educators who have rejected Fryer’s findings, it does not matter if students are highly motivated if they are running on the wrong type of motivation.

The right type of motivation is engagement for its own sake. Engagement that is sparked by genuine interest. I find it hard to disagree with that.

What Fryer’s work has changed in me

On the whole, I am emerging from this essay feeling more open to the idea of incentives than I was at the start. I’ll say why.

But first I want to say that there are still things that worry me about them:

1. I know it’s dramatic and unlikely, and I know there are lots of ways to help kids nurture a healthy relationship with money, but there’s still a part of me that’s worried they’ll turn into mouth-foaming capitalists who view every action as a transaction and treat people and the planet badly.

2. I do think there’s something to the slippery-slope threat too. Like, how do we in all seriousness tell kids that it’s fine for them to earn cash for studying but not for, say, being kind or for making their bed?

3. And my biggest concern is still, “What about the pleasure of learning itself?” If the purpose of school is to foster a love of learning, then I want the fostering part to include a lot more than just cold hard cash; things like curiosity, authentic curiosity. And creativity and collaboration and all that good stuff.

But here is where my perspective has shifted:

There will always be some subjects or activities that only become intrinsically rewarding after putting in considerable unpleasant effort. In these cases, without incentives to lure us in the beginning, we may give up (because we’re too young, too distracted, too irresponsible) and therefore never get to appreciate the depths of a discipline or the full range of our resilience, like wells we never draw from.

Writing essays has taught me this lesson. I have not always loved writing essays. They are gruelling and confusing. And when I started writing in 2022 I sought motivation in the promise of extrinsic “glory” (readers, job opportunities, paid subscriptions). Today, I love writing essays so much that not even Fryer could pay me to stop. And I don’t care nearly as much about status as I used to. But I strongly doubt I would’ve developed intrinsic drive or pleasure without those early incentives.

If this “delayed bloom” of intrinsic motivation is what’s holding millions of kids back from cracking subjects open and discovering the love of learning for its own sake, then that’s a big problem, and I’d be prepared to consider anything—including incentives—if it had a chance of addressing it. Simply expecting or hoping kids to automatically love learning does not seem realistic.

And, to be totally honest, when I zoom out, I struggle to see how all of this is that different from other rewards we dish out to kids unselfconsciously, like dessert if they finish their dinner, or pocket money if they do their chores. In the adult world, we use incentives all the time, don’t we.

But I actually think the BIGGEST concern I would have about dismissing Fryer’s findings would be the risk of making education even less equitable. This is because, if ideological squeamishness about money keeps some people from using incentives, whilst affluent schools like Alpha adopt them, then a tool that has the potential to narrow achievement gaps could actually end up widening them.

I do not think cash incentives should ever become THE story we tell children about why learning matters. But it would be a mistake if we rejected incentives just because money makes some of us uncomfortable, only to watch well-resourced schools use them to help their kids excel, while kids in mainstream or less-funded schools fall further and further behind.

⬥

---

---

Previous post