Treasury must come clean on how much taxpayers spend on BEE premiums - Daily Maverick

While some BEE premiums are recorded and others are not, all BEE premiums are unreported.

Gabriel Crouse 
While some BEE premiums are recorded and others are not, all BEE premiums are unreported.

This year should go down as the last when a South African finance minister delivers the Budget without accounting for BEE premiums. In time, people will wonder how Treasury managed to get away with anything less for so long.

Public procurement comes in at about R1.2-trillion per annum, making it proverbially the biggest deal in town. Senior Treasury official Willie Mathebula explained to Parliament earlier in February that taxpayers pay “premiums” under a 90/10, 80/20 “point” system.

“The maximum premium paid in that 90/10 preference point system is 11.1%. So that’s a premium that the state is prepared to pay to advance transformation. In terms of 80/20, of course, it is 25%,” said Mathebula.

The 11.1% premium cap is on tenders above R50-million, while the 25% premium cap is on tenders below R50-million.

The “preference points” Mathebula mentioned are BEE points. In other words, BEE premiums are capped at 25% per contract, unless a contract is above R50-million, when the cap comes down to 11.1%.

The state is “prepared to pay” BEE premiums, but has not been prepared to say how much is paid – or has been paid, every year, since at least 2008 – ultimately by you and me and the poorest, who pay disproportionately high Value Added Tax.

How much taxpayer money is paid out in BEE premiums? No one knows exactly.

A paper we discovered by the late University of KwaZulu-Natal professor Stephen Oseko Migiro showed that in a small sample from 2010, the average “premium” paid under BEE policy was roughly 20%. This sample came from the procurement office in Mmabatho, North West, and should not be assumed to represent the national average today.

But we do know that taxpayers are funding BEE premiums today.

The City of Cape Town’s Procurement Transparency Report showed several BEE premiums being paid. For example, traffic signal controllers were bought for R80,273 even though they were available in the tender process at R74,750, due to BEE, which is a BEE premium of 7.4%.

If the national average was 7.4%, then annual BEE premiums would total more than R80-billion. By comparison, according to official statistics in 2022, there were 13.2 million recipients of child support grants who received a cumulative R72.7-billion.

Recorded BEE premiums like the traffic tech case above are probably an order of magnitude less prevalent than that. By contrast, unrecorded BEE premiums are paid whenever the best bid is not even on the tender record, because of BEE.

While some BEE premiums are recorded and others are not, all BEE premiums are unreported. The latter point cannot continue.

Section 216 of the Constitution requires “both transparency and expenditure control” at Treasury. This must be vindicated, as I have argued both at the National Assembly Finance Standing Committee and to the NCOP Finance Committee, and to provincial legislatures.

Value vs numbers
Why does this matter? 

The 2010 Migiro study points to a core problem with accounts confusion around BEE.

In that survey, around 50 procurement officials were separately asked whether they agreed or disagreed with the claim that preferential procurement caused an “extra financial burden” to the state. Forty-nine percent agreed, or strongly agreed; while 34% disagreed, or strongly disagreed.

That is amazing. Such disagreement would be understandable if the claim was, “paying BEE premiums is a good idea”. 

That values-based prompt would obviously get different responses from different officials. However, for an accountants’ team to split like that about whether BEE premiums averaging around 20% on the accounting books even exist is not the way a prompt about numbers is supposed to be answered.

Some would call denial of any extra costs’ existence dishonesty; others would call it incompetence. It is probably simpler to think of it as a product of fealty.

Whatever its cause, what happens when trillions get spent under a procurement system that provides for BEE premiums under a leader, ultimately the president, who cannot even admit their existence?

What happens when the Budget comes and goes every year and no one bothers to ask how much was spent on the primary fiscal incentive for BEE?

The Zondo Report found that confusion in the procurement system arising from the “inevitable tension” between a “maximum value-for-money” imperative and a “transformative” imperative systematically facilitated State Capture, and advised maximising value for money going forward.

But it is not just the procurement office. Public debate itself has been “captured” in the confusion.

South Africa is heading into a national election. There has already been a lot of talk. But the quantum of BEE premiums is not under debate, because it is not out in the open.

Were BEE premiums too high, or too low, last year and what should come next? No one talks about that. 

Unlike every other major Budget item, no one knows how much more, or less, Treasury expects to spend on BEE premiums in the coming years.

The good news is that Treasury does what it is told by Parliament. 

On 6 February, NCOP finance committee chairperson Yunus Carrim (ANC) requested that Treasury provide statistics within 10 days on how BEE functions in procurement. Treasury gave some useful statistics, which we acquired.

For example, out of the last R1.22-trillion in spending covered by the central supplier database (CSD) since 2017, about R290-billion was paid to SOEs, and another R47-billion was not racially classified.

R140-billion was paid to “not black-owned” companies while R128-billion went to “partially black-owned” companies. “Black-owned” companies were paid R587-billion.

This means R883-billion was paid into private, racially indexed hands, of which R587-billion, or 66%, went to majority black-owned businesses. Non-black businesses were paid 16%.

Treasury previously described the CSD as “representative” of procurement more broadly. If so, it would follow that majority black-owned businesses received two-thirds of the R6.686-trillion spent over the past seven years, which is R4.445-trillion.

These numbers matter, but the BEE premiums matter even more since they fall directly under cost control.

The “premiums” tell us what extra the “state has been prepared to pay”.

BEE proponents and BEE critics should join the call for Treasury to report these BEE premiums promptly.

Let’s face facts, as the Constitution intended.

Gabriel Crouse is a Fellow at the Institute of Race Relations.