Chapter 8: counter-argument for the exclusion restriction in the education-earnings example

[DAlkemade]

I'm really enjoying your handbook, so thanks a lot!!

It's not really a big deal as it doesn't matter for the explanation, but I thought I might have an interesting addition. There is the following passage:

In other words, the time people are born is not an indication of their personal ability or any other factor that can cause a difference in earnings, other than the effect on education. A good way of doing that is to say that the quarter of birth is as good as randomly assigned when we are thinking about it’s impact on earnings. (It isn’t random. There is evidence that people tend to conceive around the end of the summer or around some sort of holiday. But I can’t think of any good reason that this pattern also affects income in any way other than through education).

I recently read a news article (in Dutch, sorry) which discussed the fact that relatively many professional soccer players were born in the first quarter. One hypothesis for this is that these kids are always bigger and stronger than their peers in their year group, which means they'll be selected more often to play in matches and have other benefits. It could be that in your example, a similar effect could happen. In the Dutch education system, for example, kids at some point are divided into different education levels based on their performance. Maybe kids born earlier in the year could more often be selected for higher levels because they are slightly ahead of their peers due to their early quarter of birth? Which would then lead to higher earnings later in life.

Once again, it doesn't really matter for your explanation, but it could be interesting to mention this as a counterargument to using quarter of birth as an instrument variable.

[DAlkemade]

Here's a chatgpt translation of the article, if you're interested:

Title:
"Had Football Talent, But Born in December? That's Tough Luck"

Background:
Football Talent - Young footballers born later in the year are less frequently chosen for professional clubs and usually don't catch up. What can be done about this 'birth month effect'?

Authors:
Joris Kooiman

Published on:
March 10, 2023

Reading Time:
7 minutes

Article:
Mees Berbee (13) never expected to be scouted. He's a skilled footballer, technically adept, and always plays in the select teams of S.J.C. Noordwijk, the club he joined at six. Although it's not a club competing at the highest youth levels, each year one or two teammates receive invitations from regional professional clubs, usually Feyenoord or Sparta, to trial. "Of course, I hope for that too," says Mees. "But they never came for me."

His father, Harry Berbee, understands why his son isn't "on the radar" of professional clubs. Mees isn't a "pure standout" in his team, he says. He's small for his age. Plus, with a November 2 birthdate, Mees is one of the youngest players in his team based on the January 1 cut-off date. On Saturdays, Mees often plays against boys who are much stronger. Sometimes, he gets intimidated, his father observes.

Yet, Mees Berbee was scouted on a Saturday in November. Not by Feyenoord or Sparta, but by the KNVB. He was selected to participate in a new initiative by the football association for youth players who show potential but remain unseen by professional clubs because they are less developed than their teammates. And so, on a cold Tuesday evening in February, Mees runs down the right flank of a synthetic field at VVH/Velserbroek in a white KNVB kit, under the watchful eyes of scouts and trainers in long, blue coats from the association.

The practice match between two regional KNVB teams is part of an effort to address perhaps the most persistent phenomenon in youth football: the birth month effect. Simply put, children born late in the year or who are delayed in their biological development have less chance of being chosen for select teams of amateur clubs and youth academies of professional clubs.

This phenomenon is not new, nor is it unique to the Netherlands or confined to football. In 1985, Canadian psychologists Roger Barnsley and Angus Thompson already identified the birth month effect, finding that nearly two out of three NHL ice hockey players were born in the first half of the year. Since then, the birth month effect has been demonstrated repeatedly, in sports and even in academic performance.

What makes the phenomenon fascinating is its persistence, even in football, where thousands of scouts, trainers, and agents are engaged daily in spotting and developing talent. It's a sport increasingly using data analysis and scientific insights to monitor and improve players. Clubs want to win, of course, but successful talent development also brings financial rewards in the form of lucrative transfer fees. Everyone understands that children born in December have, on average, as much aptitude as those born in January. Yet, the latter are often rated higher from an early age and receive more attention and opportunities. Conversely, autumn-born children are systematically undervalued and their talents, crudely put, neglected.

Jorg van der Breggen, manager of 'football development' at the KNVB, says the problem often starts with the youth players. "From around 7 years old, children start playing matches. This means clubs must form teams and traditionally look at who plays best at that moment. If you're born just after the new year, you're generally bigger, more cognitively developed, and have more football minutes under your belt. So, these children end up in the first team more often."

This initial selection has significant consequences. While select teams receive two to three training sessions a week from a certified coach, lower teams often have a well-meaning parent as a coach. Scouts attend the first teams' matches, not the games on field 7 where boys who didn't make the cut play. "It's a self-reinforcing effect," says Van der Breggen. "So, the differences keep growing."

Inspired by visits to Scandinavian clubs, the KNVB has been trying for a few years to convince amateur and professional clubs to change their approach to young players. Talent is simply hard to gauge at a young age, and development is unpredictable, as the numbers show. Hence, the KNVB's message is to stop making sharp and narrow selections until boys and girls are at least around 12. Provide everyone with the same facilities and focus on enjoyment rather than winning. After all, the desire to form a winning team leads coaches to select players who perform best at that moment, which isn't very telling.

USV Hercules, a large club in Utrecht, was one of the first to change course. Since 2019, Hercules stopped early selection, explains Jesse de Boer, head of youth training. Instead, the club gives its youngest members a 'rating' before each season and forms more or less equal teams. From 9 years old, the first broad selection takes place: half go to the select team, the rest to 'broad teams.' But everyone follows the same training. "We hope that boys and girls who previously trained less and worse now develop better," says De Boer. "So we can make better choices when we start selecting sharply, from around 13."

His first impressions are positive, he says, but it's too early to make definitive statements about the results (the first cohort without early selection is now under-11).

According to the KNVB, "dozens to hundreds" of amateur clubs now work in this way, more or less. However, there's also significant resistance. From club administrators and coaches who want to win, but especially from members. Some parents took their child away from Hercules because it could play in a select team at another club, says De Boer. Meanwhile, many young footballers turn to commercial football schools where everything revolves around selection, competition, and the prospect of a professional football career. Some professional clubs, including Ajax and Feyenoord, still scout from around 6 years old.

Prominent figures in the football world support this approach, arguing that great talents need the opposition they get from a young age to improve. "At the KNVB, they want to turn it into a playground until 12 years old. Just mucking around at your amateur club. Then you understand nothing," said former coach Aad de Mos in De Telegraaf. "Do they really think in Zeist that Frenkie de Jong, Calvin Stengs, Myron Boadu, or Matthijs de Ligt would be so good so young if they had played until the juniors among players from Laakkwartier, CVV, Gemert, or DUNO? There's excellent work happening at professional clubs."

It's a familiar argument. Even though scientific studies show that you can't say much or anything about young children's performances and their potential for a top sports career, everyone knows Frenkie de Jong, but not the talents that have fallen by the wayside or simply never been noticed. Moreover, it's impossible to know how Frenkie de Jong would have fared if he had joined a professional club's youth academy later.

What the KNVB does know: the persistence of the birth month effect proves that there are numerous small Frenkie de Jongs at lower amateur levels, underestimated by coaches and overlooked by scouts. The football association is now trying to discover these talents and give them the attention they deserve. For this purpose, it uses the Youth Plan, a decades-old KNVB selection program that traditionally served as a kind of safety net for youth players who were (just) not good enough for professional clubs. Since this season, the Youth Plan focuses exclusively on children born in the third or fourth quarter or who are late 'bloomers.' Boys like Mees Berbee. They receive weekly additional training from the association and play practice matches, including against professional clubs.

Not everyone within the association reacted enthusiastically, says Stanley van Kesteren, responsible for the Youth Plan in the west of the Netherlands. "For our scouts, it was 'a thing' that they suddenly had to look at lower levels. Some left. Also, it's difficult not to just pick the best player, but the one with the most potential. You have to pay attention to intentions, learning ability. Normally you look for a decisive pass, now you have to search for the boy who wants to give a decisive pass but might just lack the strength."

Whether 'Youth Plan Futures' succeeds in helping underrated talents and weakening the birth month effect in Dutch football, Van Kesteren doesn't dare to predict. It's "pioneering," he says. The first evaluation will follow after this season.

Mees Berbee, at least, is enthusiastic. "Trainings are more intensive, we do different exercises, and the level is higher than at my club." Here too, among his Youth Plan team mates on the field at VVH/Velserbroek, Mees is one of the smallest. Not for long, he hopes. "I still have to get my growth spurt. Maybe it's already started. I've been having growing pains."