CityPress - - Business -

value against the dol­lar than it would have any­way, said a source with in­ti­mate knowl­edge of the mat­ter.

“The client does not get harmed in the way peo­ple think. He would not have paid a dif­fer­ent ex­change rate; he would sim­ply have got less of the dis­count that large clients usu­ally get,” said the source.

“The vol­umes in­volved are not enough to push the rand. The harm is not on the ex­change rate we all see. “Bonuses are the driver.” The­o­ret­i­cally, col­lu­sion could have seen banks’ clients ei­ther over- or un­der­pay­ing for the rand.

The re­fer­ral by the com­mis­sion ac­cuses the traders of mov­ing the rand in both di­rec­tions.

More im­por­tantly, the ac­tual scale of their fraud would likely have been tiny, com­pared with the mar­ket for rand trad­ing.

Ex­changes of rands for dol­lars amount to R51 bil­lion a day, ac­cord­ing to the af­fi­davit sup­plied this week by the com­mis­sion’s Mfundo Ngob­ese, who heads its car­tel di­vi­sion.

“It [the case] will take years and the peo­ple in­volved will be gone by the time it con­cludes, said Ian Cruick­shanks, chief econ­o­mist at the SA In­sti­tute for Race Re­la­tions and a for­mer vet­eran of Ned­bank Cap­i­tal. “You get these traders who get bonuses based on their own trad­ing profit, so maybe they take a chance. I don’t know, but wher­ever you have a mar­ket, you will have some­one try­ing to beat it.”

Cruick­shanks also ex­pressed scep­ti­cism re­gard­ing how much in­flu­ence col­lud­ing traders could have on the rand.

“The pri­mary fac­tor gov­ern­ing the value of the rand is still con­fi­dence,” he said. Stu­art Theobald, chair of eco­nomic re­search agency In­tel­lidex, also noted that “it [traders’ in­flu­ence] is only ever go­ing to be a tem­po­rary move”.

“One thing the Com­pe­ti­tion Com­mis­sion is not say­ing is that the ab­so­lute level of the rand was ma­nip­u­lated,” he wrote on the web­site of his com­pany, which has a rep­utable track record in fi­nan­cial and com­pe­ti­tion re­search.

The in­sider source em­pha­sised that only a small num­ber of in­di­vid­u­als were cul­pa­ble.

“Traders get paid their bonuses in March and you see a lot of staff turnover im­me­di­ately af­ter­wards. Maybe some­one is un­happy with the R2.5 mil­lion bonus they got, and they sign up some­where else for an­other R2.5 mil­lion sign-up bonus,” said the source.

“The bonuses are premised on vol­umes. Vol­umes are what mat­ter. If I have high vol­umes and lots of volatil­ity, that is where col­lu­sion comes in.”

The com­mis­sion al­leges five dif­fer­ent types of col­lu­sion, amount­ing to dif­fer­ent forms of price-fix­ing and mar­ket di­vi­sion.

The pur­pose of price-fix­ing would ul­ti­mately be to widen the gap be­tween the price at which the traders buy a cur­rency and the price at which they sell it.

Apart from rais­ing profit this way, col­lud­ers would also want to en­sure cer­tain vol­umes go to each of them, which de­scribes mar­ket di­vi­sion.

Traders would di­rect big trades among them­selves by pulling out of the trad­ing plat­form they used at the right time, or by agree­ing that only one of them would of­fer a slightly bet­ter rate.

“The client feels happy be­cause they got a dis­count, but ac­tu­ally no one was com­pet­ing in the first place,” ex­plained the source.

Of the 17 banks be­ing probed, only eight are so-called au­tho­rised deal­ers in South Africa – reg­u­lated by lo­cal au­thor­i­ties.

The al­to­gether 25 au­tho­rised deal­ers in South Africa had al­ready been sub­ject to a re­view by a For­eign Ex­change Re­view Com­mis­sion (FXRC), set up by the SA Re­serve Bank and the Fi­nan­cial Ser­vices Board in 2014.

The FXRC is­sued a re­port in Oc­to­ber 2015, say­ing it found “some ev­i­dence of in­ap­pro­pri­ate shar­ing of con­fi­den­tial client in­for­ma­tion”, but also found “no ev­i­dence of mal­prac­tice or se­ri­ous mis­con­duct”.

It con­cluded that the lo­cal for­eign ex­change mar­ket was com­pet­i­tive and that banks were kept hon­est by “multi­banked cor­po­rate clients”.

About 60% of the to­tal trade in the rand takes place out­side South Africa.

The terms of ref­er­ence for the FXRC noted in 2014 that the rand was es­pe­cially vul­ner­a­ble to ma­nip­u­la­tion as it traded glob­ally, but with­out hav­ing the depth and liq­uid­ity of the ma­jor in­ter­na­tional cur­ren­cies.

The rand was “able to be moved by large play­ers in any par­tic­u­lar di­rec­tion”, it said.

The clients of cur­rency traders po­ten­tially af­fected by col­lu­sion are over­whelm­ingly non-res­i­dent.


Some­where along the line, the com­mis­sion and the banks are go­ing to ar­gue about what harm the col­lu­sion ac­tu­ally caused. This will, how­ever, only de­ter­mine one part of the penalty, if there is one.

The ex­tent to which the ac­cused co­op­er­ate also weighs on that de­ci­sion.

“Quan­tum [the to­tal fig­ure] is usu­ally an es­ti­mate. Of­ten it is a ball­park fig­ure. What will hap­pen is that they [the banks and the com­mis­sion] will have ri­val econ­o­mists make cases,” said a source fa­mil­iar with the work­ings of the com­pe­ti­tion au­thor­i­ties. It comes down to the­ory as much as hard mea­sure­ment. If the com­mis­sion is ar­gu­ing that rand/dol­lar ex­changes were ma­nip­u­lated ex­ten­sively enough to have un­der­val­ued the cur­rency in a long-term and mean­ing­ful way, a mas­sive claim could amount – tied to the sup­posed over­pay­ment for im­ports into South Africa.

The 10% fine, which the com­mis­sion is pur­su­ing, is highly un­likely to be im­posed, not least be­cause not all the ac­cused par­tic­i­pated in the al­leged col­lu­sion to the same ex­tent.

Even if the max­i­mum fine is im­posed, it could be rel­a­tively small. This ex­plains the non-re­ac­tion by banks’ share prices to the com­mis­sion’s re­fer­ral this week.

In the the­o­ret­i­cal case against Absa, the trad­ing di­vi­sion im­pli­cated is re­spon­si­ble for 15% of the bank’s non-in­ter­est rev­enue – which to­talled about R3.5 bil­lion in 2012.

The bank’s trad­ing pool con­sists of about 60 peo­ple, of which five, at most, are im­pli­cated.

If Absa was not al­ready in­dem­ni­fied by its co­op­er­a­tion with the com­mis­sion, it could very eas­ily ar­gue that the fine should be on the rev­enue of the im­pli­cated traders, mean­ing 10% of a frac­tion of the R3.5 bil­lion.

Even the most strin­gent fine would then amount to only R29 mil­lion.

Stan­dard Bank had a trad­ing rev­enue of R8.8 bil­lion in 2012, and only one ac­cused em­ployee, so a sim­i­larly small fine could be ar­gued for, even at 10%.

Even in rel­a­tively sim­ple in­dus­tries such as ce­ment and con­struc­tion, ri­val econ­o­mists have come up with wildly dif­fer­ent es­ti­mates of the dam­age done by car­tels.

Newspapers in English

Newspapers from South Africa

© PressReader. All rights reserved.