Recent documents and proposals on the credit amnesty have caused a predictable and not unjustified flurry of concern from lenders and investors.
The analysis underlying the amnesty proposals highlights the importance of access to credit. But only in passing does the documentation mention the most critical category of consumer credit: mortgages.
While there are problems of access to credit across other classes of consumer credit, they would be better characterised as a problem of too much access, rather than too little. In contrast, the mortgage numbers speak volumes. In 2008, mortgages accounted for roughly half of all new consumer credit granted in South Africa. In 2012, this had declined to a quarter.
Over the same period, unsecured credit increased from 10 percent of consumer credit granted to 24 percent. Growth in unsecured credit extension has been particularly fast in higher-income segments, which in the past would have had access to mortgages but no longer do. That this has occurred at the same time as the growth in the black middle class is tragic.
Developers, estate agents and mortgage originators, particularly those who operate in the so-called affordable housing market, are nothing short of exasperated by banks’ strict lending criteria and limited granting of mortgages. Aside from the challenges associated with affordability, potential buyers must have perfect records to get a mortgage. And there are very few of those buyers around. Arguably, it is in the mortgage market where credit bureau data has the most profound impact on access.
But the story is not so simple. Even if that data were to be removed or cleaned, access problems would still remain because the problems with mortgages do not reflect demand side issues alone. There are serious constraints on supply. Indeed, credit-granting criteria are used to ration credit to ensure that the best risks are selected for the little funding that is available. The key problem is that the availability of funding is severely restricted in the first place. Why this is the case should be the primary focus of investigation.
That said, there are other issues with bureau data; this is acknowledged by the Credit Providers Association and is the focus of much of its efforts.
But perhaps more problematic than the data retained by bureaus are the underlying processes that generate the data. I refer in particular to the data on judgments, comprised in the main by what are known as garnishee orders.
In 2012 the forensic audit division of law firm ENS Africa investigated the garnishee orders at a mining company employing 45 000 workers. Of the 13 000 workers who had garnishee orders, a staggering 59 percent of these orders had not been authenticated; 39 percent were cancelled as soon as they were investigated.
According to Peter Allwright, who led the investigation when he was still at ENS Africa, it was found that those orders had in fact been paid previously and the poor employees had been repaying that debt several times over.
This is nothing short of disgraceful. Given the extent of irregularities uncovered by this investigation, the burden of proof should be on debt collectors to show that this is an exception rather than on the regulator to prove that it is not.
The investigation clearly raises serious concerns about the data on judgments, which on the face of it would justify their complete removal – not so much on the grounds that they limit access to credit but more so because they are unjust. Indeed, it makes you wonder who the amnesty should be for.
We probably need to do things differently in credit markets. Debates about things that might or might not happen, such as the debate about the amnesty, often tend to be a little unproductive.
There is too much focus on analysing the past and too little on experimenting to predict the future. Indeed, it is odd that lenders do not advocate for “champion-challenge” methodologies in assessing proposed regulations; if these are so useful in assessing lending criteria, surely they would help to resolve debates about the impact of removing data.
* Illana Melzer is a co-founder of Eighty20 Consulting, an independent consulting company that focuses on customer value analysis and database analysis.