Food prices and income level

According to the latest report of the Food and Agriculture Organization, UN's organ, food prices jumped to the all-time high in December last year, beating the previous high in June 2008 that "sparked deadly riots from Haiti to Egypt." As the current rise of food prices, coupled with high unemployment, has already provoked riots among Algerian youths, the likelihood is increasing that riots over high living costs spread to other countries. In fact, neighboring Tunisia has seen unrest over unemployment these days, claiming three deaths.

The Food Price Index compiled by the FAO started at the initial value of 110.3 in 1990. From then on, the index has gained more than 100% and hit 214.7 last month, surpassing the previous all-time high of 213.5 in June 2008. But the recession after the Lehman crisis plunged the index to a 22-month low of 139.0 in February 2009. It means that the latest figure has climbed more than 50% since a 22-month low in February 2009 and reached the new high in the same 22 months.

Setting the value in February 2009 at 100, two groups can be observed in the composition of the index. First, sugar has risen the most among five categories since February 2009, and oils and dairy follow. Sugar and oils are gaining more than 100% in nearly two years. On the other hand, cereals and meat have increased only around 30%, though the former rose starkly after losing nearly 15% in June last year.

What's interesting is the relation between consumer price index by country income and food prices. First, if looking at a level of consumer price index and food prices, all countries, irrespective of income levels, have recently strengthened the relationship between them. In fact, every income-level country, high, middle, and low, has more than 0.9 of correlation coefficient in 2000 to 2009, while a negative coefficient is observed in all three in 1991 to 1999.

However, if looking at the correlation between the growth rate of consumer price index and food prices, a different picture arises. The most striking is that the CPI growth rate of high income countries has increased its correlation with that of food prices in the recent decade. In fact, the correlation coefficient of high income countries has jumped from 0.11 in 1991 to 1999 to 0.72 in 2000 to 2009.

It's not clear why the correlation is going up in high income countries. It would be against the intuition that food has a larger part in low-income countries, which is expected to lead to a higher correlation in low income countries. It would, at least, mean that commodities including food have recently pushed overall prices higher in high income countries, although correlation has no clue on which leads which.