IN NOVEMBER, the World Bank report “Fiscal Policy and Redistribution in an Unequal Society” was launched at a big Treasury seminar in Pretoria. This was just after the mid-term budget policy statement that signalled the coming budget will be an austerity budget.
The World Bank report was met by a choir of positive responses from the economic policy establishment.
The bad news was the “declining fiscal space”. The good news was that “South Africa uses its fiscal instruments very effectively”. The researchers claimed that “the fiscal system lifted 3.6 million individuals out of poverty” and reported that the average income gap between the richest and the poorest 10 percent of the population (deciles 10 and 1) decreased from 1 000 times to 66 times, “when taxes and all social spending are combined”.
Finance Minister Nhlanhla Nene will have the full support of the World Bank for a budget that freezes the size of redistribution from rich to poor. “South Africa’s fiscal deficit and debt indicators show that the fiscal space to spend more to achieve even greater redistribution is extremely limited.” Luckily, however: “In sum, fiscal policy already goes a long way toward redistribution.” I shall deal below with the 3.6 million allegedly saved from poverty. Let us first look at the World Bank’s “times more” measure of inequality.
We start from R1 000 to R1 in income disparity and transfer one single rand from the rich to the poor person, so now she has R2. Inequality has been more than halved. The reason is that R999 divided by R2 is R499.5. The tax rate is 0.1 percent. We can exclaim “look, only 66 times more”! when R14 has been taxed from the R1 000. The rich person keeps R986 and the poor person now has about R15. The tax rate is a magical 1.4 percent.
In the report, the metaphor “from 1 000 to 66 times” relates to the top 10 and bottom 10 percent. The income difference was R200 compared to R207 369 in average “market income” a year. After taxes and social grants the difference was R2 131 for the poor and R141 075 for the rich a year. R66 294 has been taxed from the average person in the richest decile. Of this, R1 931 was transferred to the poorest. This R1 931 is 0.9 percent of the market income and 2.9 percent of the whole tax. While R64 363 went to other purposes.
What if the average tax rate on the top 10 percent was not 32 percent (as it appears in the WB report)? Let’s raise the average to 33 percent. The reform takes an extra R2 138 a year from the average top 10 percent individual. Let the very wealthy at the top contribute most to that increase. If we add the whole amount to the average income of the bottom 5 million people, it doubles their average income to R4 269.
Let us now see if there is any pressing need for such a drastic change.
In a new report on poverty, Statistics SA has just set the limit for extreme poverty, the food poverty line, at an energy intake of 2 100 kilocalories (kcal) per day: “A level that meets the minimum daily caloric intake requirement.” This has become a WHO (World Health Organisation) standard for developing countries. The cost for a nationally representative food basket giving this energy was R335 per person a month in 2011, Stats SA says. This is the food poverty line.
But households must buy things other than food, such as clothes. To establish the second poverty line, Stats SA studied households who spent just around R335 per person a month. On the average, they bought necessities like clothes, for R176 per person a month. This leaves only R159 for food. This is less than half of R335.
Now, the national average price for one calorie in the report means that this R159 a month buys 997 kcal of food a day in 2011.
But the starvation limit for a woman is about 1 200 kcal and for a man 1 600. For this reason, and even when overlooking the immense and growing problem of food quality among workers and the poor, the food poverty line is a starvation line, the line of aggravated hunger and damage to health. After the increase by R14 to the higher norm of R335 a month, an estimated 10.9 million South Africans are classified as living in extreme poverty.
The NDP (National Development Plan) poverty goal for 2030 and the Millennium Development Goals’ promise of halving poverty by 2015 is the so-called “Lower Bound Poverty Line” (LBPL).
It is set by adding the average R176 for absolute necessities to the R335 for food. This makes R501 per person a month. Below R501, the assumption is that individuals simply must sacrifice food in order to buy necessities. The LBPL is the food sacrifice line, as it were.
More importantly, there has been a crucial change. Stats SA only mentions this in a footnote and says nothing about its staggering statistical consequences, which also invalidates all comparisons back in time.
In 2007, Stats SA and the Treasury wrote in a joint publication: “The daily energy requirement, as recommended by the South African Medical Research Council is 2 261 kilocalories per day.” The WHO food poverty standard was already ten years old in 2007. It was rejected. One reason could be the extreme spread of HIV and TB in South Africa among the poor. And yet today Stats SA uses the outdated WHO standard, previously rejected in 2007.
What has changed, beside “the fiscal space for redistribution”?
If the statistical policy remains unchanged, all three poverty lines increase by R25. The new food poverty line becomes R360. The political benchmark LBPL changes from R501 to R526. Should the government retract its political directive to Stats SA, 1.3-1.4 million more people become classified as extremely poor and people living below the LBPL rise to 19.6 million from 18.6 million, which is 39 percent of the population (not the reported 37 percent).
And the percentage of people below the upper bound poverty line (which moves from R779 to R804) will no longer be 53.8 percent, but will rise to 55 percent.
In the WB report, on the other hand, the old rand values are of course used in the tables. But when praising fiscal policy in South Africa, the WB researchers use US dollars: $1.25 (R14.50), which is the international food poverty line, and $2.50, which is the international LBPL. They use so-called PPP (purchase power parity) comparisons.
And indeed, these limits can only be used for (not so meaningful) international comparisons, not for praise of local policy. The World Bank’s own tables over so-called PPP exchange rates 2011 show this. Comparing the total cost of the same amounts of a food basket in the US and South Africa gave a “purchasing power price” exchange rate of R6.64 to $1.
Strictly compared with a food basket and using the bank’s own tables for 2011, $1.25 PPP translates to R249 a month. The person on such an income is going to bed hungry. If carefully weighting the PPP exchange rates published by the bank, between food and all non-food items, a LBPL of $2.50 (PPP) yields R425 per person a month. This is even below the old LBPL of R443. The person on this income is by definition sacrificing food in order to buy school uniforms. She is in no way one of 3.6 million individuals “lifted out of poverty”. The World Bank researchers are making scandalously untrue claims in support of budget austerity.
Over the last 15 years, the fiscus has given back billions in tax revenue to the country’s highest income earners.
Previous published in Business Report February 24 2015 at 08:00am
By Dick Forslund
Dick Forslund is senior economist at Alternative Information and Development Centre.