The World Bank published new estimates of the number of people in poverty yesterday. They are very important and they’ve been universally misreported.

The estimates show:

• the developing world is poorer than we thought; there are 1.4 billion people living in poverty (about one quarter of the developing world), not 985 million as we previously thought
• nonetheless, progress in reducing poverty has been about as fast as previously believed – poverty has been declining at the rate of about one percentage point a year, from 52 percent of the developing world’s population in 1981 to 26 percent in 2005. This is a reduction in the number of poor of about 500 million people.

As the full paper explains, the new poverty line is $1.25 a day in 2005 prices, compared to the old poverty line of $1.08 a day in 1993 prices.   This is actually a downward revision of the poverty line in real terms: if it had been kept the same in real terms (ie adjusted only for inflation) it would be $1.45 a day in 2005 prices.  (There are currently 1.7 billion people living on less than $1.45 a day in 2005 prices, the equivalent today of the old poverty line – which is nearly twice as many as we previously thought lived in poverty.)

The meaning of the poverty line is often misunderstood.  Some people assume that the poverty line measures the number of people who have an income of $1.25 a day; and they reassure themselves by thinking “a dollar will go a long way in some countries”.   But the poverty line is measured as $1.25 a day at purchasing power parity – that is, people below this line are able to buy each day what $1.25 would buy them in the United States.  This really is an absolute measure of poverty.

Of course, the newspapers got this all wrong:

James Politi in the FT reported

The new figure was estimated after researchers at the bank raised the threshold for extreme poverty from earnings of $1 a day to $1.25.

(Not true; the threshold has been reduced in real terms). The BBC reported

The new estimates suggest that poverty is both more persistent, and has fallen less sharply, than previously thought.

(Not true: it has fallen at the same rate as previously thought; just at a much higher level.)

Finally – a big untold part of this story is the big changes in the purchasing power parity estimates that underpin these poverty figures.  These show massive changes in the estimates of GDP at PPP. For example, here in Ethiopia, GDP per capita is now estimated to be $591 per year, compared to $1084 under the previous estimate.  India is down 40%, now below Pakistan in income-per-head; and China’s income per head is also 40% lower than the previously estimated.