The journey begins...

My ‘discovery’ of the micro-oriented approach began in the heart of Southwest China. For my dissertation research project, I followed a somewhat unusual way of generating a puzzle: I used a straightforward strategy of using the best (and limited) data from China available—GDP per capita, poverty rates, all on the provincial level—to model the classic conception of the relationship between economic growth and poverty reduction. Most mainstream economists would expect that as an economy’s economic growth increases, the income of the poor should also increase. As one influential 2001 World Bank report put it (in its title, no less), “Growth is good for the poor.” This paper went even a step beyond, arguing not merely that t economic growth and poverty reduction were causally connected, but also that,

Most scholars agree that this relationship is generally valid, although nearly all would admit to exceptional cases. But most mainstream economists will dismiss these exceptions as noise. So instead, I wondered, what can we learn from these exceptions? From the provinces where poverty reduction significantly outperformed economic growth, could we uncover insights that might make economic growth more pro-poor? And from provinces where poverty reduction did not move much despite rapid economic growth, could we learn what not to do?

In 2002, when I was arranging this research, comparable provincial-level poverty data was extremely limited—indeed, such data were a state secret. But the World Bank published four years of aggregated versions of that data, though one of the report’s authors confided that only two years of that was actually reliable. So armed with these two years of data—spread five years apart—I compared the change in provincial-level data with each province’s growth rate. This simple model was also statistically significant—economic growth. But instead of just accepting that conclusion and walking away, I applied a simple calculation to identify four exceptional provinces: two that experienced economic growth that was pro-poor and two where economic growth was the opposite.

And the two most extreme of these provinces were actually neighbors. And as it turned out, these neighboring provinces—Guizhou was the good one; Yunnan was the bad—were quite similar. Given this, I identified a research site with natural controls! Both provinces were largely agrarian, grew tobacco as a primary product, were extremely mountainous, had lots of rainfall, were home to significant numbers of ethnic minorities, were blessed with vast reserves of natural resources (especially coal), and were both in the Southwest, which means they both experienced similar treatment from central government policy.

Now, China’s provinces are huge and complex places. And though by then, I had already spent years in China, I had never been to either of these provinces. Yunnan, in the southeast extreme of the country, was more famous. It had been since the 1980s a popular spot for backpackers seeking authentic experiences with rural ethnic minorities—such as the Dai people of Xishuangbanna, which bordered both Myanmar and Laos, such as the Bai people of Dali, like the Naxi people of Lijiang, and more recently, the Tibetan culture of Diqin, (later erroneously renamed Shangri-la). As a border province, Yunnan enjoyed lots of international trade, and was also an important region militarily. So it was not surprising that Yunnan’s growth rate was pretty good. Given rural tourism, its tobacco industry, its trade, and its natural resources—these added to the confusion. If these were the basis of Yunnan’s growth, why didn’t these translate to much poverty reduction?

Guizhou province, by contrast, was famous for three things. First, the province hosted several historical sites of the Red Army’s long march, part of Mao’s revolutionary soldiers’ 5,600-mile attempt to escape from the clutches of Generalissimo Chiang Kai-shek’s pursuing Nationalist army. It was in a series of meetings in Zunyi, in northern Guizhou, that Mao made his final ascent to become the undisputed leader of China’s Communist Party. Second, the province has, reported for thousands of years, produced Maotai, the potent sourgum-based alcohol reserved for toasting honored state visitors, including Nixon in 1972.

And third, Guizhou was famous for poverty. The dire conditions in Guizhou in the 1940s inspired historian Joseph E. Spencer to write, “Nowhere in several provinces of central and southern China has the writer seen so many abandoned farms and farmhouses, nowhere is there the same pinched, dulled, barely alive look on the faces of so many people, almost nowhere are the villages, towns, cities and the whole countryside so devoid of those cultural adornments characteristics of the rest of China” (1940, 162). A diverse range of notables, including Zhou Lin, the province’s first party provincial secretary (“all the nationalities [there] lived a frightful existence”); Wang Xiaochuang, party propagandist (“the people wore rags...and their shabbiness was indescribable”); and explorer Alexander Hosie in the 1880s (“dangerous, barely cultivated, and economically backward”) agree on this point (see Goodman 1986, 33-34).

To be sure, Guizhou also had rural tourism, but its reputation dwarves that of the neighboring Yunnan, which ranked sixth in China for tourist visitors, compared to Guizhou ranked a moribund 24 out of the 28 comparable provinces. And Guizhou was either last or next-to-last among all provinces for GDP growth at the time. Thus, for these many reasons, Guizhou’s record of poverty reduction outstripped its growth rate was really surprising. (In fact, I fretted that the data were incorrect, though subsequently, I managed to verify and triangulate the pattern in a number of ways.)

OK - now the puzzle is set up. Two similar neighboring provinces, both exceptions to the idea that economic growth is linked to poverty reduction. And the one with a worse reputation and much slower growth rates experienced much more poverty reduction!

How could this be? Did Guizhou offer more welfare to its poor than Yunnan? Did Guizhou implement market liberalization policies in ways that allowed the poor to benefit? Did Yunnan experience a natural disaster (Yunnan is earthquake-prone! Or maybe the province experienced a drought?) Was culture a factor—like, were the ethnic minorities in Guizhou have a cultural attribute that made them more dynamic, but in a way that only reduced poverty? Did Beijing send Guizhou more aid? Or perhaps Guizhou implemented the Kerala model, a development approach based on empowerment and education. Instead, did Guizhou end up following the advice of Nobel Prize-winning economist Amartya Sen, to increase the capabilities of poor people?

I tested each of these hypotheses and more! Based on secondary and primary sources, interviews with experts from around the world—everything from the World Bank to Beijing-based social scientists) and nearly a year’s worth of fieldwork (thanks to the Fulbright folks!) wandering from village to village interviewing officials and farmers—none of these hypotheses proved true. I tested around 100 hypotheses related to economic policy, politics, culture, demographics, even geography and the weather! For nearly all of the tried-and-true pathways to poverty reduction, the evidence from Guizhou and Yunnan proved inconsistent with the patterns that we would expect—and in many cases, the data was actually in the opposite direction!

I kept at it. Gradually, I found factors that proved important. Guizhou’s rural tourism was much more modest than Yunnan’s. But its tourist areas had already emerged from poverty by the mid-1980s—further growth did little to help the poor. Moreover, by the late 1980s and early 1990s, Yunnan’s tourism policy shifted away from its backpacker model. These rural areas started putting in airports and fancy hotels and restaurants. The sites were closed off to the local community, and open only to tourists and official employees. Expensive, exclusive resorts became enclaves, and often didn’t even purchase local agricultural products. Lots of money was spent—generating growth, but little went into the pockets of the poor.

By contrast, Guizhou’s much smaller rural tourist industry—also in ethnic minority areas—was organized in a bottom-up way. A tourist arriving there would stay with a local resident, who would cook for them and even serve as a tour guide. The locals would sell handicrafts and perform cultural dances and songs for tourists. Less money was spent—less growth in those guesthouses—but more money went directly to poor farmers.

I could start to see the disconnect: economic growth was not reaching the poor in one case, whereas more modest growth managed to pull many out of poverty.

Similar patterns were seen in a range of areas. Both provinces invested in infrastructure, though Yunnan’s funds were in more expensive highways that radiated from the capital of Kunming to link itself to other provincial capitals and foreign countries. More highways, more trade. More highways to ferry tourists around, more quickly and more comfortably. These highways often charged tolls and featured fences that kept out the locals. And in any case, there were at first only six such highways during this initial period. Even as they generated GDP, these highways could not benefit the poor—even those who lived nearby—in any meaningful or systematic way.

Guizhou also built roads—some highways, though not nearly as extensive as Yunnan’s.

The province mainly focused on building roads that linked villages and marketing towns. Such roads—often made of cement, but often just well-constructed dirt roads—did not generate much GDP. However, these roads did link farmers to nearby markets, helping them access more information about everything from market demand to migration opportunities. Outside intermediaries could also use these roads to purchase agricultural products, and bring in more necessities. These roads were so extensive that by the mid-1990s, the province ranked third in the density of these ‘non-graded’ roads.

The way the two provinces developed their natural resource extraction also exhibited a similar pattern. Yunnan developed its mines more formally and with more machinery. What jobs machines could not do typically required specialized skills and some formal education—even a middle-school education was out of reach from many poor farmers. Guizhou has one massive mine in the Liupanshui area. But besides that area, most mines were mined by artisanal, part-time miners, rural residents who could mine during the off-season. Such nines were less formal—and even a small group of friends could apply for a license to mine in them. Some were unsafe (though most of these mines were surface mines, rather than deep pit mines), they were terrible for the environment, they were wasteful—and in the end, the central government banned such mines throughout China. But while they remained legal, hundreds of thousands of rural families in Guizhou could boast at least one part-time miner. But because Guizhou exported most of its coal to more prosperous provinces, and ended up—ironically, unjustly—effectively subsidizing China’s wealthier coastal provinces! Though Guizhou’s coal industry was structured in a way that poor people without much formal education could benefit, that coal ironically contributed mainly to coastal provinces’ GDP! Yunnan also subsidized the raw materials it sent each, but somehow the province ended up keeping more of its own coal, which contributed to its own GDP. Even still, its industry was structured in such a way that generated little employment for poor people.

The provinces’ migration policies were polar opposites—Guizhou sent more workers—and poorer ones—to other provinces to find jobs. They remitted most of their money, and these remittances bolstered the finances of many of Guizhou’s most impoverished families. With these remittances, Guizhou’s families accessed better food, built warmer houses and bought warmer clothes. Yet the remittances were rarely of the same scale as those of other provinces that are more famous for migration. Moreover, the province established education programs to train workers in more advanced skills. They set up offices in areas that needed workers to help migrants find jobs and secure other necessities. Transportation companies—many of them informal—traversed the rural road network to pick up workers in remote areas, and ferry them, through what highways there were, to jobs in wealthier ones. Their earnings helped their families emerge from poverty. Their labor contributed to the GDP of other places. Meanwhile, without programs and policies to train and place them, Yunnan’s rural residents did not migrate—and when they did, they stayed local.

(There were a few more important factors that didn’t get into the book: Guizhou had a terrific work program designed so that poor people would self-select into it. By participating, rural residents gained some skills, even as they built rural roads, irrigation and other vital infrastructure.)

What do these programs—tourism, roads, coal, migration—have in common? Both provinces had policies that focused on these areas. It wasn’t the programs’ existence that mattered. It was the way they were structured.

In Guizhou’s case, each of these is designed to generate opportunity—some (like tourism and coal) directly, others (like roads and migration) indirectly. But these generate specific kinds of opportunities. Artisan mines rather than industrial; backpacker havens rather than five-star resorts; small country roads rather than highways; migration programs designed with the needs of the poor in mind. Each of these focuses on the small scale. Each of these focuses on low-tech.

• Small-scale—human-scale—provides opportunities for the poor. Large-scale tends to shut them out.

• Low-tech allows people without much formal education to participate; high-tech excludes them.

• 
  

And because, in Guizhou’s case, the state was involved, a friend suggested a powerful moniker—the micro-oriented state. (Since the state is not inherently necessary, I now call this the micro-oriented approach). The same friend suggested a title for my book on Guizhou and Yunnan—“Small Works.”

A powerful idea. And one that, as I mentioned in a previous post, flies in the face of the advice of most development experts and institutions, governments and academics. Scale up! Increase in technology! Not just Dollar and Kraay of the World Bank. That’s been the advice, at least since Adam Smith. Karl Marx also agrees. Scale and technology are the very heart of progress! (That’s another post.)

The micro-oriented state. Only later, I discovered I didn’t discover it. After a visit to a small women’s college in the south, one professor—dropping me off at the airport—encouraged me to read E.F. Schumacher’s Small is Beautiful. This 1973 publication was kind of a hippie book. And while others have pushed the idea (that’s another post!), the idea still remains fringe. That said, I was pleased to discover that Small is Beautiful received 15,000 cites from Google Scholar—that’s quite a number!

It is my goal to dust off Schumacher’s idea—now a half-century later—dust it off, and give it more relevance.

As I became attuned to the micro-oriented approach, examples emerged out of nowhere. Not just Guizhou—far from it. From Barbados to Bhutan. From Trinidad to Thailand. From rural Louisiana to urban Chicago.

Urban Chicago, where a young community organizer documented his experience from the mid-1980s, reflecting what I had found in Guizhou in the early 2000s:

The organizer was a mid-20s Barack Obama. The quote is from his Dreams from My Father.

The micro-oriented approach. It is not that this is better than others mentioned above—those theories don’t explain Guizhou, but it doesn’t mean that these theories don’t work (more in another post). Instead, the micro-oriented approach is an alternative pathway to poverty reduction—one that has its pluses and minuses. One that is not universally applicable. But certainly, one that is under-researched and under-discussed.

To follow: who hates this idea? Who gets close? Whatever happened to Schumacher? And numerous other examples. Stay tuned!