Measuring local economies
IN A poor country, under a regime that pursues a programme of authoritarian capitalism and keeps unreliable statistics, no one is terribly surprised when the government reports that the economy is gleaming and things are generally looking up for the people. Sifting the truth from the propaganda was never rocket science. But in the past it did require a visa. To estimate how and how much people earn, and how this has changed over time, an observer needed access to eyes and ears on the ground.
Those days may be ending, or so a pile of new research suggests. Souknilanh Keola, Magnus Andersson and Olla Hall are a trio of scientists who specialise in monitoring economic activity from space. They have devised a way to estimate changes in economic activity that does not rely on sending out an army of impartial pea-counters. Instead they feed a variety of economic models with some basic facts about land use and then add fine-grained data about night-time emissions of light, gathered from satellite images.
The scientists are remarkably ambitious about the results:
We show that it is possible to estimate agricultural and non-agriculture growth of administrative units of virtually any shapes and sizes with radius more than 1km. The implication of this is large and broad because such small administrative units, particularly in developing countries, do not have official figures of growth. Our framework can therefore potentially be used to analyse economic impact of…public or private investment [or] disasters, in extremely fine scales.
One of their first subjects of investigation has been Laos. Rapid changes in land use—in the form of deforestation, mining and dams—would appear to make the country their ideal guinea pig.
They make use of two satellites that were already in orbit around the earth, circling the planet and picking up lots of useful information. The images they beam back to earth can be used to generate data about land cover, surface temperature, vegetation and evapotranspiration. And these satellites have been filing their reports with admirable regularity, covering the world’s entire surface every day or two since 2001. The resolution of their images can be as fine as 250 metres per pixel. Until not so long ago, access to the complete picture pool cost tens of millions of dollars. Now it is free.
Based on this data the scientists are able to make their estimates of economic activity in Laos at the district level. So far their technique appears to work extremely well at spotting the impact of foreign cash from space. The districts with the fastest-growing rate of non-agricultural activity in the period between 2003 and 2010 are: Vilaboury, the location of the country’s largest gold-and-copper mine (with an annual rate of 23%); Pakkading, home to Laos’s largest hydroelectric project (at 22%); Xayboury, the conduit for trade from Vietnam on the Thai-Laotian border (21%); and Ton Pheung, which a Chinese-backed casino plans to make into a “Macau on the Mekong” (it’s no Macau yet, but paddling away at rate of 20%). Their attention to agricultural economic activity has produced results that are no less intriguing. For example, the scientists estimate that four out the seven districts in the province of Oudomxay, which borders China’s Yunnan province, have among the fastest-growing rural economies in the country.
Indeed, something dramatic is going on there, and it merits closer inspection still. From the passenger window of a plane, 700km lower than the satellites’ orbit, the naked eye reveals the extent of poor Oudomxay’s deforestation.
(Coincidentally, the garish green clearings of the landscape below now match the seats of the plane. It’s an Airbus 320 that Muammar Qaddafi once ordered; the colour of the seats was made to match the colour of Mr Qaddafi’s Green Book, published in 1975, the year Laos’ civil war ended. The late colonel ran out of time to take delivery of the A320, which subsequently fell into the hands of Laos Airlines, the state-run carrier.)
Even farther down, at street level, anyone can observe that the jeeps roaming the deforested valley of Oudomxay bear licence plates from China and Vietnam. Chinese firms have secured concessions covering 30,000 hectares of Oudomxay. Tens of thousands of Chinese workers will one day work here, tapping the rubber trees that are to be planted at some stage. For now the endemic forest has been hauled away, biodiversity is dwindling and the soil’s nutrient levels are just beginning their long decline. Despite the fact that 85% of the agricultural investment in Laos is happening south of Vientiane, a 20-hour bus ride away, it is relatively easy to see signs of it here, by the border with Yunnan. To date most of the “investment” in Oudomxay has consisted of clearing timber and natural growth; the tricky bit about cultivating the land has yet to begin.
Laos still has more forest per square kilometre than any other country in Asia. Razing its forest to prime the land yields an immediate uptick in economic activity, and this seems to be just what the space-based analysis captures. But many of the villagers in this deforested valley look less than joyous. Poverty levels have been falling, but it is unclear to what extent foreign contract-farming has benefited the locals. Agriculture here tends to be subsistence-oriented, and many of the farmers still practise shifting cultivation.
Another scientist, who has spent about ten years living in Laos, says that in the past decade the government has granted land concessions to Chinese, Vietnamese and Thai investors that exceed the size of all the rice-growing areas in Laos. Concessions for agriculture, along with others for mining and the development of hydropower projects, have helped sustain annual GDP growth of 8%—a rate, he says, that the policymakers regard as “untouchable”. In defending that headline figure, he fears (though he does not yet have the data to validate it), they risk courting a new kind of poverty. Ultimately, this scientist fears, their policies could change the country in a way that is not visible from space: by creating a population of landless poor, deprived of their livelihood and full of resentments.
(Picture credit: NASA / Earth at night)
Source: S. Keola, M. Andersson and O. Hall: “Monitoring development from space: Using night-time light and land cover data as proxies of economic growth”. Forthcoming