Monday, November 28, 2005

Destitution not dearth

Destitution not dearth
Aug 18th 2005
From The Economist print edition

Niger's harvest last year was not so terrible. Why is the country now so hungry?

“MUCH about poverty is obvious enough,” wrote Amartya Sen, one of the world's best-known and most respected economists, in his 1982 classic, “Poverty and Famines”. “One does not need elaborate criteria, cunning measurement, or probing analysis to recognise raw poverty and to understand its antecedents.” But the thesis Mr Sen propounded in that book was not obvious at all: some of the worst famines, he argued, have taken place without any significant fall in the supply of food.

One of the examples Mr Sen chose to illustrate his thesis was a famine that gathered force from 1968 to 1973 in the Sahel region of Africa. The Sahel, from the Arabic word for “shore”, typically refers to a group of six countries on the western fringes of the Sahara, where the desert sands lap up against the vegetation of Africa's semi-arid zones. The countries worst affected by this disaster 30 years ago were Mauritania, Mali, Upper Volta (now called Burkina Faso)—and Niger.

Niger is once again in the grip of a food crisis, if not a full-blown famine. The distress sales of livestock, the heavy migration and the deprivation the country suffered in the early 1970s have all revisited it again this year. How well does Mr Sen's thesis explain the country's latest encounter with mass hunger?

Much about Niger's current crisis appears obvious enough: the rains last year ended early; the locusts were rampant. Who can be surprised that the country is short of food? But Niger's harvest last November was merely mediocre, not disastrous. Although the rains ended early, the country's cereal production was only about 11% below its five-year average, according to the UN's Food and Agriculture Organisation (FAO). It was 22% greater than the harvest of 2000-01, a year that passed without alarm. The locusts did more damage to the region's fodder than to its food, prompting pastoralists and their herds to begin an early migration to greener pastures in Niger's coastal neighbours.

Purchasing powerlessness
Niger's distress shows up most clearly in prices, not quantities. A pastoralist's terms of trade depend on two prices in particular: the price of what he can sell (his livestock) and the price of what he must buy (food). In Niger this year, the latter has soared; the former has plummeted. According to one report, the price of millet and sorghum rose to 75-80% above its average for the last five years. By June, the sale of one goat bought half as much millet as it had six months earlier. It is precisely this kind of cruel twist in the terms of trade, Mr Sen argued, that can bring a community to its knees. These unfortunates will suffer a lack of power to purchase food, even if there is no lack of food to purchase. Why did prices move against Niger's pastoralists so far and so fast?

The spike in the food price may have reflected high foreign demand as much as low domestic supply. Traditionally, during the lean months before their harvest, Niger's farmers import cereals that are cheaper to grow in wetter, coastal neighbouring countries than in their own country. But according to CILSS, an intergovernmental body responsible for the region's food security, significant amounts of grain have this year been flowing in the opposite direction. Ghana, Benin, Côte d'Ivoire and Nigeria have all been buying up grain in the region.

This is partly because these countries' own harvests were disappointing. But in Nigeria's case, the FAO thinks that government policies were also to blame. Nigeria has imposed controls on imports of rice and wheat products; it has also taken steps to protect and promote its millers and poultry farmers. Both of these policies have raised demand in the country for millet and sorghum, which provide alternative sources of flour as well as chicken-feed. As a result, Nigerian cereals that might have found their way to Niger are instead being consumed at home. Nigeria has twice Niger's income per head and more than ten times its population. Its powerful market pull may have helped to undermine the purchasing power of Niger's pastoralists. “In the fight for market command over food,” Mr Sen noted in his book, “one group can suffer precisely from another group's prosperity, with the Devil taking the hindmost.”

Nigeria, with Burkina Faso and Mali, has also restricted grain exports to Niger this year, violating its trade treaties with the country. Such restrictions have often played an ignoble, supporting role in the history of famine. A ban on cereal exports between India's provinces, for example, condemned Bengal to ruinously high prices in its great famine of 1943.

What of the other term in the terms of trade? Livestock prices have fallen in the past year, partly because northern pastures were damaged and animals were emaciated as a result. But the deterioration in the terms of trade can also generate its own momentum. Higher cereals prices prompt herdsmen to sell more of their livestock. These distress sales drive the price of animals down further, forcing pastoralists to sell still more of their herd. In his book, Mr Sen raised the theoretical possibility that a pastoralist's supply curve might actually bend back on itself: as the relative price of livestock falls, a hungry pastoralist might supply more animals to the market, not fewer as elementary economic principles would imply.

If mass hunger were simply the result of there not being enough to eat, the remedy would be obvious: more food. The emergency rations now being shipped, flown and trucked into the Sahel are indeed necessary and urgent by the time hunger and destitution are acute and widespread. But if mass hunger begins with a collapse in purchasing power, rather than a shortage of food, it does not take an airlift to prevent it. What is needed is a way to restore lost purchasing power by, for example, offering employment, at a suitable wage, on public works. The market respects demand, not need. But give the needy enough pull in the market, and the market will do most of the rest.

Saturday, November 26, 2005

People power

People power
Nov 24th 2005
From The Economist print edition

Two academics use game theory to explain why democracy is so hard to achieve


IN 1381, a mob of angry Essex peasants revolted against the poll tax, and marched on London, destroying tax registers and records as they went. The Essex men wanted an end to their serfdom and the right to rent land at fourpence an acre. King Richard II, just 14 years old, bowed to their demands and the mob dispersed, although not before invading the Tower of London, trespassing on the royal bedchambers, and killing the Archbishop of Canterbury.

The conflict between mutinous masses and self-preserving elites is the theme of this ambitious, even audacious, book by Daron Acemoglu, of the Massachusetts Institute of Technology, and James Robinson, of Harvard University. Their aim is to figure out when such struggles result in democracy, and when that democracy endures. The peasants' revolt of 1381 was not such a case. Once the uprising had ebbed, the young monarch reneged on his promises to the Essex men, rounded up the surviving ringleaders and had them executed.

As this sorry episode shows, people power is fleeting; sullen discontent flares only rarely and briefly into forceful dissent. This is obviously a problem for rebels, who find it hard to sustain a revolt. But, as this book points out, it also poses a conundrum for the elites who might want to appease them.

It is a king's or dictator's prerogative to change his mind. As the sovereign, he can overrule everyone, including himself. Thus after an insurrection peters out, the sovereign cannot hold himself to any concessions he might have made when the rebels were at his bedroom door. Dissidents should anticipate this. If they do not want to meet the same fate as the Essex men of 1381, they should not settle for the ruler's sops, but push instead for outright revolution while they still can.

This dilemma sets up the book's main thesis: democracy is a solution to the elite's commitment problem. By agreeing to a peaceful extension of the franchise, the elites institutionalise their concessions, and the masses lock in their power before it drifts away. Institutions, such as universal suffrage, are harder to overturn than policies, such as rent of fourpence an acre. It is easy, all too easy, for the elite to renege on a promise to the people; rather harder for it to mount a coup against them.

Uncircumscribed authority can be a handicap. It makes it impossible for the sovereign to make a lasting concession even if he wants to. Such paradoxes are familiar to students of game theory, a subdiscipline of economics on which the authors draw heavily. Game theory sheds light on strategic encounters, in which each player's move must take account of the others'; it is a good way to explore the machinations and manoeuvres of politics.

The authors hope to convert readers to their method as well as their argument. The pace of the book suffers from this laudable pedagogical purpose. They tell the reader what they are going to say, say it, then tell the reader they've said it. Their technical apparatus, like scaffolding, is no doubt a great help in building their theory. But an awful lot of scaffolding is left out on show, which is a great help to anyone teaching this book, but a distraction for those who just want to admire the edifice.

And there is much to admire. True to their title, Messrs Acemoglu and Robinson offer a unified theory of both democracy and its opposite. Some Latin American countries have swung between the two with metronomic regularity. Argentines won universal male suffrage as early as 1912, but lost it to a coup in 1930. Democracy was restored in 1946, overthrown in 1955, re-introduced in 1973, subverted in 1976 and cemented, one hopes, in 1983.

In the authors' eyes, the demise of democracy is a near mirror image of the fall of dictatorship. Anxious incumbents try to buy off the military, much as Richard II sought to placate the Essex men. Chile's Salvador Allende, for example, raised army pay and benefits. But in the long run, a democracy cannot commit itself to serve the interests of anyone but the swing voters. The military elite know this, and take their chances when they can. Allende was duly deposed in 1973.

The book takes its title from “The Social Origins of Dictatorship and Democracy” by Barrington Moore, an American sociologist who died last month. The conclusion of that treatise has been summed up as: “No bourgeoisie, no democracy.” Messrs Acemoglu and Robinson agree that the middle class can be a midwife for democratic rule, not least because the elite is more willing to cede power to merchants than to the mob. They also agree with Moore that agrarian societies tend towards authoritarianism. Land is easier to expropriate than capital; and peasants are easier to repress than factory workers. Thus feudal lords fear democracy more than capitalists do; and have an easier time suppressing it.

Such bold generalisations and pithy dictums have fallen out of fashion since Moore wrote his classic in 1966. One more recent scholar counted no fewer than 27 different factors that are said to promote democracy. This book is entirely free of such intellectual indecision. The authors are brutal wielders of Occam's razor, and the 27 factors have been chopped down to a coherent handful. This may leave a lot out, but what historians bemoan as simplistic, economists tend to celebrate as parsimonious. According to two scholars cited in this book, even to look for a general theory of democratic reform requires great temerity. Happily, Messrs Acemoglu and Robinson have temerity in spades.