Wednesday, April 21, 2010

South Africa's very own Fannie and Freddie

by Galen Sher

Improving rates of land ownership has extensive socioeconomic benefits for a country. In South Africa, where many working citizens do not own their own property, finding ways of improving land ownership is critical for our development. A popular approach is the US model of Fannie and Freddie - government institutions that extended credit to low-income earners and guarantee loans to low-income earners issued by private banks.

They did so successfully and profitably for a number of years, greatly improving the proportion of Americans who owned their own homes. Unfortunately the incentive structure around them allowed them to fuel a bubble in the residential housing market and moreover to pose an enormous systemic risk to the US financial system.

South Africa is poised to launch our own institution in a similar vein: a large fund that guarantees loans made to low-income (i.e. riskier) borrowers, with the aim of increasing the number of SA home owners.

Over at Moneyweb, Chris Blaine published this piece on 19 April. I have commented in italics.

JOHANNESBURG - Government has put together a billion rand fund to put a "floor" under banks' "gap" market loans. Is this the US subprime market reinvented for SA?

Minister of Human Settlements Tokyo Sexwale announced that he has briefed banking executives on government's plans. These will be announced in Sexwale's budget speech on April 21. This is so as to "not surprise them".

The discussions revolved mostly around the "gap" market, defined as those earning R3 500 to R9 000 per month. This is the group that finds it most difficult, if not impossible, to access bank credit explained Sexwale.

The broad idea is that the fund will insure banks against losses from loans made to the "gap" market. Sexwale says this will not only enable banks to lend, but will raise the cap of what banks will commit to the sector.

Government loan insurance (partial insurance, I hope!) allows banks to price their loans lower than they should be priced, given the cost of capital and the true riskiness of the people receiving the loans. This insurance makes loans cheaper and hence more affordable to low-income borrowers. The overall cost of providing a home loan to a risky borrower does not change, but the cost is shared between the bank and the taxpayer.

The fund is essentially to give banks confidence to make loans they wouldn't have before. A bank executive heading for the exit just after the meeting commented that if the banks could've made money from a market they would be in there already.

This insight was mirrored by Cas Coovadia of the Banking Association. Somehow though, Sexwale believes that this government "insurance fund" will allow banks to make responsible loans compliant with the National Credit Act.

A major difference between the (arguably failed) US system and the proposed SA system is that SA has the National Credit Act, which imposes stricter conditions on SA banks' lending than US banks faced. These stricter lending practices should mitigate, but will not eliminate, the risk of 'liar' loans and 'ninja' loans, etc that took place in the US.

Sexwale acknowledged that the US, and global, economy was brought to its knees as a result of banks making loans to people they knew could never repay them. He spoke of local bank Unifer as an example to bear in mind. It lost billions through dodgy loans leading up to the small-bank crisis.

Interestingly, when asked if the fund will insure loans for specific groups in the "gap" or everyone, Sexwale responded "everyone". It will be interesting to see if this really is a colour-blind scheme in implementation.

The scheme must be colour-blind for it to be constitutional in SA.

So what's the difference between this SA government initiative and the US subprime implosion?

Sexwale said we have proper regulations in place, banks that value their reputations... and "[we] trust the banks aren't run by crooks".


I'm afraid this explanation is simply not good enough. The crookedness or otherwise of SA bank managers has little to do with the sustainability of such a housing scheme. Minister Sexwale needs to have a clear policy in place for dealing with the house-price inflation that will accompany such a policy.

This intervention will divert taxpayer resources from productive economic activity towards a low-return housing investment. This tax-and-spend system creates capital inefficiency that is frowned upon by hardline libertarians. Personally, I see the potential social benefit of such inefficient capital investment. At the very least therefore I don't think this is a clear-cut debate to be argued purely on the basis of philosophical principle. Rather, I think this debate poses a cost-benefit problem that could be investigated through a thorough economic analysis.


Importantly Sexwale and the bank executives said they are working as a team to make this scheme workable. Hopefully this will result in a system that benefits those intended. Details were scarce as the scheme seems to be still in its early phase.

SA banks could benefit handsomely from such a policy, depending on the extent of the government guarantee. Can I hear anyone shout "Rent-seeking"?

The fund is envisaged to start at R1bn, but Sexwale said that as the economic cycle improves he could approach Treasury to provide more guarantees, further underpinning bank loan books. This sounds a bit like chasing the cycle though, something US banks did as the housing bubble grew.

The question lingers, if the banks could really make money in the "gap" market then surely they would be there already? Does this scheme not privatise profits and socialise the losses just like America did?

On the first question above: largely, yes, if there were money to be made here then SA banks would probably already be investing here. There are various reasons why SA banks may have avoided this market even if it were profitable, but let's ignore these for brevity. Let's assume that there is no money to be made in this market segment - but surely profitability is not a necessary condition for government investment? Just because this would be a loss-making exercise does not make this bad policy. Government projects are often unprofitable.

Another issue is that government has been criticised over its implementation of plans. In fact, a similar discussion such as Monday's one happened in 2005. From the sounds of Monday's briefing, nothing has happened in the five subsequent years.

No comment!