STANFORD
UNIVERSITY PRESS
  



Hive Mind
How Your Nation’s IQ Matters So Much More Than Your Own
Garett Jones

BUY THIS BOOK


Introduction

The Paradox of IQ

This isn't a book about how to raise IQ: it’s a book about the benefits of raising IQ. And a higher IQ helps in ways you might not have realized: on average, people who do better on standardized tests are more patient, are more cooperative, and have better memories. But while dozens of studies by psychologists and economists have established these links, few researchers have connected the dots to ask what this means for entire nations. And since average test scores vary across nations—whether we’re talking about math tests, literacy tests, or IQ tests—an overall rise in national test scores likely means a rise in the number of more patient, more cooperative, and better-informed citizens. This in turn means that higher national test scores will probably matter in ways too big to ignore. And if education researchers and public health officials can find reliable ways to raise national test scores, productivity and prosperity will rise where poverty and disease now flourish.

You can get a sense of how big these effects are by looking across countries: nations that do the best on standardized tests—nations such as Singapore and Finland—usually have governments that are reasonably free of corruption; have decent roads and bridges; and have plenty of private investment in office buildings, factories, and homes. China does well on standardized tests, and particularly in the post-Mao decades, the nation’s economy has grown rapidly. The high test scores in these countries are a sign that their citizens have the cognitive skills, the human capital, to take on the complexity of a modern economy.

By contrast, nations where test scores are average or lower tend to be the kinds of places where people have to bribe government bureaucrats to get things done, whether it’s the school principal, the bureaucrat at the driver’s license office, or the congressman’s brother-in-law. And even if you don’t have to bribe the government, it’s a good bet that the government will be inefficient, sluggish, less than competent. Nations with lower average test scores are usually tough places to take on complex, costly private investment projects, since skilled workers and twenty-four-hour-a-day electricity are hard to come by. Lower-scoring nations aren’t places that appeal to international investors, and so private investment tends to drift away. The long-run result of lower test scores? It’s often a mixture of rickety bridges, decrepit buildings, slower Internet connections, and less prosperity. On average, nations with test scores in the bottom 10 percent worldwide are only about one-eighth as rich and productive as nations with scores in the top 10 percent.

Outside of a few countries with abundant natural resources, the most important productive asset in each nation is the human mind. And while standardized tests can’t tell us everything about how productive the mind is, the tests can tell us more than you might think. Boosting broad mental skills boosts a nation’s prosperity, and while standardized tests are obviously not perfect—no statistic ever is—they are a good way to measure those skills.

Think Win-Win

One of Professor Steven Covey’s “seven habits of highly effective people” was “Think Win-Win.” In the Covey view, a key to success in business and in life is to look for ways to find pie-growing solutions rather than to just focus on grabbing the biggest slice of a fixed pie.1 Pie-grabbing and pie-growing are both rational actions—they both make you better off, at least in the short run—but nations where people tend to grow the pie will have more pie to eat. So what predicts pie-growing skill? Researchers at Vanderbilt University ran two experiments to find out. One study looked at what traits predicted skill in haggling over the price of an industrial commodity—so one student is “selling” tin for as much as possible to another student who is “buying” the tin for as little as possible.2 This was a simple pie-slicing game. The researchers collected data on personality as well as scores on the business school General Management Admissions Test, or GMAT, a test that’s much like the SAT and similar to some IQ tests.

The second Vanderbilt study put two students into a free-form negotiation game: a mall developer haggling with a potential anchor store (like a Macy’s or a Bloomingdale’s) over the details of the contract. The second bargaining situation was more multifaceted, more “integrative,” as management professors like to say. Could the anchor store sublease space to cosmetics companies? Could it keep different hours than the rest of the mall? Could it sell the same products as other big stores in the mall? Who pays for the escalator? Issues such as those were addressed, along with the usual haggling over the price. Clearly the second study was more complicated than the first. And one thing that IQ-type tests predict is the ability to handle complexity, to keep multiple facts in mind.

So did test scores matter or didn’t they? As so often in the study of humans, the answer is “it depends.” Student test scores didn’t help at all to predict haggling skill in the first, simpler study: students with higher GMAT scores weren’t better than others at buying low or at selling high. But in the complicated mall negotiation study, it turned out that the average test scores of the pair of players did a moderately good job of predicting success. Pairs with high average GMAT test scores were more likely to increase the overall value of the project by getting the details of the contract right, and they were less likely to overlook ways to build overall value: they were more integrative. But what about personality traits, how much did they matter for success? The personality measures used in the study—extraversion, conscientiousness, and the rest—all did a worse job of predicting pie-growing behavior than the GMAT score.

On average the high-scoring pairs tended to be pie growers. Other psychology and economic experiments have backed up this basic finding: players with higher scores on IQ-type tests are more likely to take the pie-growing approach, especially in complex situations. One might imagine what that could mean for entire nations.

Your IQ Doesn’t Matter That Much

But if test scores matter a lot for nations—a claim that economists and psychologists alike have made in recent years—then we’re left with a puzzle. I’m sure you know a lot of people who are big successes in life—people with good jobs and nice kids—who don’t do well on standardized tests. I know people like that too. And we both know people who are failures—bad habits, constant legal troubles, no money—who do great on IQ-style tests. And these aren’t just one-in-a-million cases: every day we meet smart people with no money and we meet slow learners with good jobs. At the individual level, test scores just aren’t a great predictor of success in life.

Economists have known this fact for decades: when it comes to the link between test scores and wages, exceptions are the rule and the link is moderate at best. So knowing a person’s IQ—or her SAT score, or how well she does on vocabulary tests—just doesn’t tell you all that much about how much money that person earns. Later, we’ll have plenty of opportunity to survey what IQ means (it’s short for “intelligence quotient”) and what it doesn’t mean, and we’ll survey some of the imperfect techniques used to measure it. We’ll also see that people with higher IQs tend to be taller, quicker to react to a flashing light, and more patient. But for now, consider this one fact: your own, individual IQ score isn’t that good at predicting how much you’ll earn over the course of your life.

That fact is half of what this book is about, and I’ve covered it in two paragraphs. Now set that next to another fact—that nations with the highest test scores are about eight times more prosperous than nations with the lowest scores—and you can see the paradox of IQ.

What I’ll do in the remaining pages is explain how both facts can be true: how your personal IQ matters little for you as an individual, but how at the same time your nation’s average IQ matters enormously for determining how much you earn, how much you produce, and how good your life feels. I’ll also explain what policymakers and public health professionals may be able to do to raise a country’s average IQ score. After all, if IQ is as important to a nation’s prosperity as I claim, then raising a nation’s average level of mental skills should be a top priority. Fortunately, here there are grounds for genuine hope: IQ isn’t perfectly malleable—at least with the science of today—but neither is it etched in stone.

As you read the book, every so often you may start thinking, “But I know someone who does great on standardized tests who’s done terribly in life, so how can IQ really matter that much for a country?” Remember: that puzzle is what this book is all about.

Do Your Nation’s Test Scores Matter More Than Your Own?

Is there really evidence that nations with higher average scores on math and science tests are richer, more productive, faster growing? Do these test scores tell us more than if we just knew the average years of schooling in a country? That’s what economists Eric Hanushek and Dennis Kimko wanted to find out.3 For decades, international agencies have measured math, science, and reading skills in countries around the world. In recent years, the best-known of these testing programs have been PISA (Programme for International Student Assessment), TIMSS (Trends in Mathematics and Science Study), and PIRLS (Progress in International Reading Literacy Study). These are the widely publicized test scores that make the news every few years to let Americans know that their students are falling behind, and the test scores have been used by hundreds of scholars around the world. In 2000, Hanushek and Kimko published an influential study that drew on an earlier, related set of international math and science tests. Hanushek and Kimko’s key finding has been replicated many times since then by scholars using a variety of international tests: test scores do a better job predicting an economy’s performance than do years of education.

Perhaps that’s no surprise. In many less developed countries, “years of schooling” don’t include a lot of actual schooling. Economists and education researchers alike have found that in poor countries, students often don’t have textbooks and teachers are often absent. That’s not a recipe for learning. But there’s more to the story than that. Another factor is that students in less developed countries come to school in a weaker position to learn. Children in these countries are more likely to have faced malnutrition and disease, both before and after they’re born.

But Hanushek and Kimko did more than just find out that test scores are better predictors of national prosperity than years of education. They also found out that higher test scores have a much stronger relationship with national economic performance than with individual economic performance. Looking at how individual student test scores predicted those students’ wages later in life, they found that individuals with higher test scores earned only slightly more than average within a given country, but nations with higher average test scores grew exceptionally fast. Here again is the paradox of IQ: standardized test scores—whether we call them IQ tests or math tests or something else—predict big national differences but only modest individual differences.

So yes, if you do better on math tests you’ll probably earn more in life. But if an entire nation has higher average scores on math tests, it can probably look forward to a substantially more prosperous future. Nations with high test scores are usually either already rich or growing quite quickly, such as either France or China. The fact that average national test scores do a great job of predicting national economic performance has been found again and again in the economics and psychology literatures, but an explanation for why this is so has been harder to come by. Hanushek and Kimko themselves seem to think that the high test scores aren’t really causing much extra growth: they say, “the [national] growth equation results are much larger than the corresponding results for individual earnings. . . .” and they also say, “the simple estimates of cross-country growth relationships appear to overstate the causal impact of” the test scores.4

But perhaps it’s not an overstatement; perhaps it’s a sign of a mental multiplier, a positive side effect of cognitive skills. Perhaps one person being good at standardized tests is just “interesting,” but a whole nation of people who tend to be good at standardized tests is world-changing. In this book, I’ll walk through some reasons rooted in mainstream economics, psychology, and political science to explain why the skills that standardized tests try to measure have bigger payoffs for nations than for individuals.

Test Scores Around the World

Let’s start of by looking at the relationship between these test scores and income per person. Psychologist Heiner Rindermann and his coauthors have taken the widely used international PISA, TIMSS, and PIRLS test scores and converted them into a single national average “cognitive ability score” for each country.5 Since national scores change over time, this is just a snapshot, but it will give us a sense of how test scores are distributed across countries. And of course, these averages hide a wide variation: there are high scorers in every nation.

Income per person differs massively across countries—by a factor of thirty to fifty between the few richest and the few poorest. So while pundits get worked up over the 30 percent differences in living standards between northern and southern Europe or between Japan and Taiwan, the real issue for human well-being is the 3,000 percent difference in living standards between the United Kingdom and some countries in sub-Saharan Africa. And just as it’s challenging to measure cognitive abilities across countries, it’s at least as challenging to measure national income across countries: neither are perfect measures. And keep in mind that the measurement problems for national income are worse in poorer countries, where government bureaucracies are more likely to be strongly politicized or where even well-intentioned officials just don’t have the resources to make accurate measurements.6

Figure I.1, in which each dot represents a nation, shows that countries whose citizens tend to do well on these tests tend to earn more income per person in a year. The figure uses economists’ favorite proxy for national income per person: gross domestic product (GDP) per person. GDP has the added benefit of focusing our attention on an economy’s productivity, how many goods and services people actually produce in a year. As the figure shows, there are exceptions to the scores-predict-income rule in today’s world, but most of the exceptions are modest. If you look at the high end of the scores, you’ll notice that the average person in East Asia tends to do quite well on these tests, and you’ll also see that most countries in East Asia (as well as high-scoring Singapore) have been economic success stories for the past half century.

Take a look at the lower-income countries in the graph, and you’ll see a general rule along with some exceptions. The countries in the bottom half of the graph have low average test scores with a few exceptions: the poorer nations with scores in the low 90s tend to be former communist countries, perhaps still recovering from decades of economic mismanagement. And some of the wealthier countries with lower average scores tend to be rich in natural resources.

FIGURE I.1   Average cognitive ability estimated in 2009 from earlier PISA, TIMSS, and PIRLS international exams and estimated 2005 GDP per person Sources: Rindermann, Sailer, and Thompson, “The Impact of Smart Fractions,” and Penn World Tables 7.1.

This brings us to the lower-left-hand corner, countries where poverty is all too common and where worker productivity is currently low. On average, people in these countries have done poorly on standardized tests. Of course, there are high-scoring people in every country, but on average people in the poorest countries tend to do poorly on standardized tests.

Test Scores: Shaped by School and Much Else

While I’ve spoken about math and science tests, from here on I’ll mostly focus on IQ tests and other similar tests. IQ-type tests—sometimes called intelligence tests, though my fellow economists often wisely prefer the term cognitive skill tests—are good choices for a few reasons. For one, IQ tests have been given in many more countries around the world than have the PISA and TIMSS tests: elementary schools around the world have IQ test scores already sitting in filing cabinets. Second, many of these IQ tests are purely visual pattern-finding tests that don’t obviously draw on school learning, so compared to a science test these IQ tests might be more fair and less biased; we’ll return to the issue of test bias in Chapter 2. And third, once we start discussing IQ rather than generic “test scores,” we can draw on a century of research by psychologists, infectious disease specialists, nutritionists, and others on ways to raise IQ scores and the skills these tests attempt to measure. IQ tests are the standard measure of cognitive skills across many disciplines, and it’s good to stick with the standard. “Math score” gets most people thinking about good teachers, good books, good educational psychology—all important ways of improving cognitive skills. But we can’t forget that people’s skills are shaped well before they begin school and are shaped by forces outside of school. By talking about IQ, it will be hard to ignore life outside the classroom.

Test Scores and Prosperity: Just a Coincidence?

Should you put much weight on these results so far? After all, rich countries and poor countries differ in countless ways, and average test scores are just one of the differences. I could just as easily have shown a graph that shows that rich countries have more cell phones per person, or that people in rich countries eat at restaurants more often or are more likely to have vacation homes. Does Figure I.1 really tell us anything about what drives the wealth of nations?

One preliminary check is to see whether a nation’s average test scores still do a good job of predicting a nation’s level prosperity even when you know a lot of other things about the country: this approach is known as multiple regression. It amounts to asking, “Even if I already know that your country has a high savings rate, a lot of mineral wealth, or whatever, does knowing your nation’s average test scores still help me to predict how productive your country is?” In a paper I coauthored with psychologist Joel Schneider, we used rough estimates of each nation’s average IQ score to predict prosperity. We found that in hundreds of statistical tests, even if you knew which region of the world the country was in, even if you knew its religious background, even if you knew how open it was to global trade, it turned out that a nation’s estimated average IQ was a good predictor of a nation’s level of prosperity.7 National test scores predict national productivity.

You might be wondering why some countries have higher average scores and others have lower average scores. That’s an important question, but it’s a question this book isn’t designed to answer. This isn’t a book about where a nation’s IQ comes from, it’s a book about where a nation’s IQ takes it.

I have a good reason for focusing on where IQ can take a nation: it’s an important overlooked topic. Also, there are already a lot of books and articles of varying quality about why IQ and other test scores differ from person to person and from country to country. I will discuss some of this research, partly because it is so fascinating and partly because I want to encourage more people to think about the topic. If higher average IQ is important to national economic outcomes, then it’s important to find ways to raise these skill levels for each and every nation. If higher average IQ is important, it’s time to start reading the high-quality books about raising national test scores and national cognitive skills.

There are good reasons to believe that national IQ scores can rise—and the best reason is because they already have. In the rich countries, IQ scores have consistently risen over the twentieth century, a phenomenon known as the Flynn Effect, after the philosopher—the philosopher, mind you—who discovered it. We’ll talk about Flynn’s world-shaking research in Chapter 3, but keep the Flynn Effect in mind when people tell you that IQ can’t be changed.

Collective Intelligence: Each Nation as Hive Mind

Animal researchers, computer scientists, and occasionally social scientists sometimes use the metaphor of “collective intelligence” or a “hive mind” to explain group actions. Why do animal herds run together to avoid predators, dodging this way and that? How do honeybees and ants share information and collaborate to build vast structures and complex economies that no one insect could ever build alone? Indeed, human society in every nation today is a form of collective intelligence, in which the accumulated knowledge of the past makes its members richer today, and in which the many small, daily cognitive contributions of millions of their neighbors—in offices, in factories, in the halls of government, and elsewhere—help to make their lives better as if by Adam Smith’s invisible hand. Those millions of small cognitive contributions are what create each nation’s collective intelligence, each nation’s hive mind. Members of society all draw on that collective intelligence, they all get benefits from the hive mind that they never pay for, benefits that, by my lights at least, they don’t deserve. And it’s typically better to be the less-skilled honeybee in the highly productive hive than to be the highly skilled honeybee in the less-productive hive: your neighbors have an important influence on what you can accomplish. That the hive mind exists for every nation is almost obvious. The key question is whether a nation’s average IQ scores are an important driver of the hive mind.

It Pays to Be Around the Cognitively Skilled

In the next three chapters we’ll review the basics of modern IQ research: brain scans and job performance and nutrition and whether we really can compare average test scores across countries. Then in Chapters 4 through 9, I’ll lay out five major channels for how IQ can pay off more for nations than for you as an individual:

1. High-scoring people tend to save more, and some of that savings stays in their home country. More savings mean more machines, more computers, more technology to work with, which helps make everyone in the nation more productive.

2. High-scoring groups tend to be more cooperative. And cooperation is a key ingredient for building higher-quality governments and more productive businesses.

3. High-scoring groups are more likely to support market-oriented policies, a key to national prosperity. People who do well on standardized tests also tend to be better at remembering information, and informed voters are an important ingredient for good government.

4. High-scoring groups will tend to be more successful at using highly productive team-based technology. With these “weakest link” technologies, one misstep can destroy the product’s value, so getting high-quality workers together is crucial. Think about computer chips, summer blockbuster films, corporate mega-mergers.

5. The human tendency to conform, at least a little, creates a fifth channel that multiplies the effect of the other four: the imitation channel, the peer effect channel. Even a small tendency to conform, to act just a little bit like those around us, to try to fit in, tends to quietly shape our behavior. If you have cooperative, patient, well-informed neighbors, that probably makes you a bit more cooperative, patient, and well-informed.

Of course, test scores don’t explain everything about the wealth of nations: I’m only claiming that IQ-type scores explain about half of everything across countries—and much less within a country. If your next-door neighbor earns a lot more or less than you, it’s not mostly because of IQ. And across countries, differences in geography, culture, and the long reach of colonialism are surely shaping the wealth of nations. The hive mind is no single-cause theory of prosperity. It’s just a story that almost no one else is talking about.