As Beijing-bound flight CA1220 bobbed up and away from Yinchuan in northern Ningxia, some of the group settled into the China Daily, which led with the latest on the deadly clashes between Han Chinese and Muslim Uighur groups in Xinjiang province in the far west.
By July 9, the violence had taken 156 lives. It rarely got so bad these days, but people in this vast country hurtling into the future were often snapping over disparities somewhere. And they were routinely getting their faces rubbed in it by boasters making claims that only underscored the gulf between haves and have-nots. The day's features section, for instance, carried a story about Beijing's "boom" in dog cemeteries—20-plus apparently—and how this entrepreneur with a nose for the city's rising population of coddled canines plans to open 20 more.
The story provided a sharp counterpoint to a nine-day field trip through some of the poorest counties of Ningxia autonomous region, home to China's Muslim Hui minority, and neighboring Shaanxi province—places where malnutrition and anemia appear epidemic. On the journey were staff and supporters of the Rural Education Action Project—a joint initiative of Stanford and key Chinese research centers—who'd come to check in on REAP's latest nutritional and educational efforts to give rural kids a better shot at life and lift communities up in a generation. Millions of rural Chinese have only a third the income of urban dwellers.
Yet getting to China's poorest villagers had us using bright, new airport terminals, airlines equipped with young Airbus and Boeing fleets, and expressways that were marvels of engineering. Even the secondary roads in the poorest counties felt freshly paved. In Ningxia, our van rolled through valley after valley, thrashing wheat occasionally as we went. (Although not officially allowed, farmers spread their crop across the road so passing traffic will crunch the grain off the shafts.) Many valleys were terraced from end to end, another government initiative, aimed at turning steeply pitched ground into more productive cropland. In Shaanxi, we zigzagged up mountains on a road so narrow we were often no more than a few feet from ravines dropping hundreds of feet. If you were confident enough to look out into the distance instead of down, the views could verge on, well, celestial.
As elsewhere, you can find the forgotten stashed in places that otherwise might be judged splendorous—and sometimes perversely close to lavishly funded government projects. With China holding an astonishing $2.3 trillion in foreign currency reserves, Beijing is spending a ton on some things. Infrastructure investment is one of the more visible signs of its effort to close gaps between urban and rural economies, between the poorer west and the richer coastal provinces. If the government isn't splashing out enough on schools in the countryside, it's not out of a reluctance to spend on education. China has opened 800 new universities of some description in the past 10 years or so.
But few rural students get that far. Left unattended, severe iron deficiency in children leads to stunted physical and cognitive growth and sharply diminished opportunities down the line. And tuition fees, largely unknown elsewhere in a free K-12 world, make China's high schools the least accessible in the world, putting university even farther out of reach.
REAP director Scott Rozelle, a senior fellow at Stanford's Freeman Spogli Institute, brings an economist's tool kit to the plight of China's rural poor. Formerly with Stanford's now-defunct Food Research Institute, Rozelle has years of experience working with Chinese agencies on agricultural policy. And it is policy that REAP programs seek to shape.
Rozelle says REAP projects often aim at "proving the obvious to others," piling up local evidence to make a case that the international scholarly literature has already made: Better nourishment leads to better school performance and increased human capital. Close the human development gap and kids will be better prepared for a radically changing world.
To assess the impact of new initiatives, REAP uses randomized controlled trials, splitting a homogeneous population sample into groups who receive interventions and groups who carry on as usual. So in Shaanxi, for instance, REAP selected 66 schools in November 2008 and did a baseline survey of all fourth graders to test for hemoglobin levels. The survey revealed that 38 percent of them were anemic, not the 5 to 10 percent teachers and principals guessed. Researchers also conducted psychological and math tests to help establish everyone's starting point. Then they selected 24 schools to receive daily vitamin and iron supplements; 12 to receive only letters advising parents on anemia risk and what to do about it; and 30 as control schools for no action.
Last May, REAP conducted its evaluation and found that two-thirds of the kids who initially tested anemic in the intervention group had improved iron levels and licked the problem—and scored significantly higher on math tests. Just as important, REAP found no change in children in the "letter" or control schools. The trial was one of the largest conducted on anemia in rural China, highlighting a problem suspected of pulling down as many as 30 million Chinese. In a current trial, the group is targeting 8,000 fourth- and fifth-grade boarders at 146 schools in Ningxia, Shaanxi and Qinghai province. (Its baseline study showed anemia in more than 50 percent of the Qinghai students.)
REAP has many collaborators and financial supporters from alumni, philanthropic, corporate and academic quarters. Rozelle calls REAP's annual summer trips "mobile retreats," a chance for everyone to break away from their individual projects, cast a critical eye on the group's big initiatives and map plans for the year ahead.
July's journey took us to primary schools, some new, some old, their classrooms decorated with faded posters of Marx, Mao and other old hall of famers. We'd shuffle into schools with crowded rooms of sometimes 50 faces that creased with curiosity, if not smiles, welcoming the interruption. We met with principals and teachers and smartly dressed party secretaries, some still in the dark about the extent of anemia, or even what it is.
We visited "dorms" for boarding students—charmless concrete spaces often filled wall-to-wall with metal-framed bunks, a single narrow passage giving the only access to perhaps five or six beds. Mattresses were bedding scrounged from home thrown over sheets of plywood; extra clothes and towels hung on nails above the beds. Personal food from home—mostly pickled vegetables and buns with cruelly short shelf lives—was stowed along with eating utensils, some still encrusted with the remains of the last meal, in wooden boxes (often old dynamite carriers scavenged from local mines).
We heard of kids going four hours into their busy day before getting anything to eat. We visited kitchens, some clean, some not, their common denominator a menu that promised an endless regime of starch, usually noodles, occasionally brightened with a drop of protein in the form of soy sauce. (Boarders used to bring grain to school to keep larders stocked. Today they pay fees, leaving the schools to worry about supplies.)
These boarding schools are not the product of an educator's vision, but a jerry-built response to problems thrown up by heaving demographic shifts. China's one-child policy, combined with increasing migration to towns and cities, emptied classrooms to the point where thousands of village schools were forced to close. The only option for children who stayed was to attend consolidated central schools in the nearest towns. Those who live too far away for daily travel, usually on foot, must board during the week, a misery many used to endure for a whole month between home visits.
There was a lot we didn't see. Because schools were letting out for the summer, we didn't get a sense of the stench from bed wetting in baking, crowded, unattended quarters or from rotting food or, by dint of the season, poisonous fumes of leaky coal-fired heaters. Indeed, we missed the hell of a north China winter stabbing through the cracks and gaps around loosely fitting single-glazed windows. And we could only imagine pint-sized boarders, far too young for even afternoon survival games, spending the winter scurrying in the cold to an outside water pipe to wash up or running to squat freezing in outside latrines. The only compensating thought was that many often faced little better when they went home for weekends, after walking 10 or more miles to get there.
On a good day, REAP proposals can end up in the inbox of the State Council, the cabinet of the central government chaired by Premier Wen Jiabao. The work is trailblazing both in scale and impact, potentially life-changing for tens of millions of children across whole regions of China.
The path to China's policy establishment begins at the Center for Chinese Agricultural Policy, REAP's partner within the Chinese Academy of Sciences, headquartered in Beijing. "The Chinese Academy of Sciences has a unique channel for voicing researchers' opinions to the top leadership in the State Council," explains Linxiu Zhang, CCAP deputy director and director of REAP-China. That doesn't mean projects can't be stopped at the council's door. Many are, she says, but the few proposals that pass screening can really go places, even the premier's office.
In January, REAP's email network crackled with some big news. Rozelle reported that REAP's policy brief on nutrition and education was approved by the State Council and would serve as a basis for policy action. What's more, the approval bore the personal stamps of Wen and Vice Premier (Education) Li Keqiang. Rozelle heard the news from Beijing on December 31. "Interestingly," he wrote, "on the same day the Shaanxi provincial department of education . . . asked us to design an intervention that would help them create a sustainable program to eliminate anemia (and address other health, nutrition shortcomings)."
Months earlier, Shaanxi's governor had mandated that an egg and milk be given daily to primary school kids across the province. As China's 17th largest province, Shaanxi's population is 37 million, the same as California's.
REAP intern Max Kleiman-Weiner, '09, remembers hearing that news. "The excitement I felt while analyzing the data, which showed that our nutritional supplements had significantly lowered the level of anemia in a few thousand fourth graders, was dwarfed by the announcement that the Shaanxi government is providing eggs and milk to every student, in major part influenced by REAP's reports and policy briefs."
Kleiman-Weiner, a Marshall scholar studying neuroscience at Oxford, says a response like this is unknown among nongovernmental aid organizations. "Instead of merely dispensing services and aid, which hopefully help people, REAP can develop and encourage large-scale programs with empirically proven impacts."
Rozelle can point to other briefs acted on in the past year. One focused on the paucity of rural preschool programs—missing out on them, it's argued, can leave kids behind over the duration of their schooling. As a result, Rozelle says, the Ministry of Education created a division of early childhood care and education and started plans to extend state support to kindergartens in poor rural areas.
REAP also documented the dismal performance of migrant schools that cater to families leaving the countryside for work in the cities. China's millions of migrant families have it rough, but the kids really take it on the chin. They're either left behind or they go along for the miserable ride, often ending up in makeshift schools, sometimes just around the corner from better public schools that traditionally barred migrants. REAP showed that, compared to their peers back home, migrant kids consistently scored lower in standardized math and Chinese tests, leading field researchers to conclude that poor teacher quality and facilities were largely to blame. "In response," Rozelle says, "the central government is using our evidence in support of new initiatives to provide financing for making public education mandatory for all students—including migrants—who live in urban areas."
Teacher, scholar, fundraiser and taskmaster-cum-part-time barista, Rozelle doesn't abide any time wasting, unless you count joining younger field trippers for a late-night karaoke session. After a dawn-to-dusk day of school visits, he seems energized by the thought of corralling everyone for two hours of brainstorming to get the day's takeaways and compare them with REAP's experience to date, the project roadmap and the relevant literature.
Just when normal mortals might reasonably be lulled by the thought of napping under a good book, he wants everyone to focus hard on REAP's crowded to-do list. That will probably entail more critical discussion of the cost or timing or logistics of an intervention; or the soundness of a proposed survey; or challenging the group to come up with incentives for school principals to really drive nutrition programs.
Or it might mean hearing Rozelle out on the nuts and bolts of collecting stool samples in school latrines in the farthest reaches of China. "You just spread this paper out across the opening and there are these disposable tweezers to pick it up with," he says, to wondrous looks. There are also special jars to carry the specimens—and first-class air tickets for interns who sign up for the task, he adds less convincingly.
"Anyone need coffee?" he asks between agenda items. Rozelle refers to himself as a "mobile Starbucks," a conceit clearly not intended to signal lolling in easy chairs.
Sometimes, just as debate winds down and agreement seems within reach, someone makes an observation that forces everyone to take a step back. Grant Miller, an economist at the Stanford School of Medicine, did it enough to earn the moniker "party pooper." If it's been a long day, Rozelle's usually hopeful smile can morph into something close to a grimace. "He splashes water on our project, that's my first reaction. But then I think, that's a good point," he says. "You need party poopers to make yourself better."
Miller came to get a close-up look at REAP. Impressed by the network Rozelle has spent a career building, he seems to be warming to its new Paying for Progress project that would see how incentives might push anemia rates down. "You can't do this work without phenomenal collaborators who know the setting," he says. "So to me it's marvelous that I can insert some ideas that we can try out together through those relationships." Some ideas batted around on the trip have been more complicated than he'd imagined, but he says that's not unexpected when plotting new projects.
One complication that seemed to jolt people on this retreat was how long it took a teacher to hand out vitamin tabs to 50 fourth graders: first, two paper cups, one for water, one to hold the tablet, had to be handed out; then vitamins; then water (after boiling it): three torturous rounds filling little fidgety hands. It was not the quick morning routine some had assumed.
By October, Miller was signed up as a principal investigator, alongside Linxiu Zhang and Jennifer Adams, assistant professor in Stanford's School of Education. In an email, he said there had been a lot written about the promise of directly rewarding socially desirable outcomes, "but little is known about how they actually work." The concern, he explained, is that the people implementing and managing projects such as an anemia-reduction program have relatively weak incentives to actually do what funders hope to achieve. "They receive their compensation regardless of success." Base compensation on success, one theory goes, and they might try harder.
"So through our project, we will learn how giving them incentives to innovate and create more effective programs leads (or doesn't lead) to larger reductions in anemia," Miller said. "Some social projects in developing countries have tried rewarding process measures thought to lead to desirable social outcomes, but as far as we know, we are the first to try to directly reward the ultimate objective." In November, field teams fanned out to train school principals for the trial, which comprises three school groups: one receiving only information on proper nutrition; another receiving both information and subsidies to procure food and supplements; and a third receiving information, subsidies and cash incentives.
"My biggest takeaway from the summer field visits was the same reaffirmation I get every time I go to the field," REAP managing director Brian Sharbono emailed later. Before coming to Stanford in early 2007, Sharbono coordinated and evaluated projects in India, Mexico, Guatemala and Brazil and had been a research analyst at the Beldon Fund. "'Good' ideas get their real test when put into practice," he said. And like other trips, this one was a chance to retrace steps and do some follow-up grading. "What really happened and why? How might we improve upon our anemia and education project going forward?"
Eric Hemel is on his second retreat, as a donor and consultant. He says he has to see a project on the ground before he supports it, and REAP comes up trumps for intellectual honesty. "Scott's a fantastic collaborator and I think that's been reinforced on this trip in just how much he cares about what other people think," he says. "Academics tend to have preconceptions and I'd describe Scott as truly, truly open-minded—and the other guys are, too. They do not have an ideological agenda."
Hemel, '74, MBA '77, PhD '80, does have a bit of an agenda. He suspects intestinal worms might be a contributing factor to poor academic performance, and he'd like to see REAP take up the issue. He's been peppering the group with questions about doing fecal tests in different provinces. "My aim is cost effectiveness," he says. "My hope is a relatively small expenditure could highlight a problem and lead to an intervention costing hundreds of millions of dollars—basically stemming from my expenditure." In October, Rozelle reported that a worm count would go ahead, with consulting help from Paul Wise, a pediatrics professor in the School of Medicine and senior fellow at the Freeman Spogli Institute's Center for Public Health. REAP's scientific collaborator: the Center for Disease Control, Intestinal Worm Division, Shanghai.
Often prefacing interjections with "I'm not arguing with you, I'm just saying . . ." Hemel says he brings an analyst's skepticism to the table. He's not shy about giving generously, just careful. He'd risen to the top of equity research on Wall Street when life took a sharp humanitarian turn after a visit to Vietnam in 2003. He and his wife, Barbara, were stunned to discover kids weren't going to school because their parents couldn't afford the $50 fees. So six years ago, Hemel set up a trust earmarking half his income for charity, committing some of that, as well as management assistance, to two scholarship programs covering 5,000 children and another program supplying medical equipment for at-risk newborns. He describes his commitment as a half-time job.
But China through REAP's lens still holds its attraction. "What's really cool is travelling in the most rapidly changing society in human history with world-class experts who can explain to me what's happening."
Rozelle sees a dearth of development research capacity across China generally. "Surprisingly little formal, rigorous evaluation work is going on jointly with the government," he says. "In practice, experimenting relied on a single 'model project,' informal evaluation and scaling up."
Government agencies are under the gun to put money to work, he explains. But because officials are constantly reassigned, there is little incentive for incumbent office holders to make sure projects make sense. "More of the evaluation effort is put into counting inputs—how many miles of roads built, how many schools constructed, how many scholarships issued."
New projects also carry risk of failure, so officials tend to put them in locations where the capable leaders are, not where randomization studies would more logically put them, Rozelle says. To get off this treadmill, he says, China's leaders will have to call for rigorously designed evaluations to be part and parcel of policy implementation.
It's not that the leadership isn't interested in doing so: It is, Rozelle says, which is why it is easy to get government agencies to help clear the way for studies by nongovernment groups. "The officials who help and participate in this way are protected from failure, since it is the NGO or the research organization that is carrying out the project. [But] they can take credit for successes—and REAP encourages them to do so, of course."
Whatever works.
JOEL McCORMICK is a journalist based in Menlo Park and Hong Kong.
