[Preamble: every time something new happens in the land expropriation without compensation debate, I end up re-sharing this post. The public debate tends to focus on the bad impacts of land reform on the economy – but often misses the internal logic problem in the land reform debate. If you are simply going to redistribute land, then history suggests that the land will eventually re-redistribute back to where it started (ie. into the hands of the elite). Which is quite self-defeating – short term economic crisis in exchange for a return to the long term status quo.]

On Tuesday, the SA Parliament passed a motion to move toward land expropriation without compensation. And so we’re now moving in the direction of accelerated land redistribution.

Here is the problem: in even semi-capitalist societies, land (like all capital) has an annoying habit of re-redistributing itself.

After the World Wars, when the old bastions of wealth were destroyed, wealth was more equally distributed. Since then, it has concentrated again. By all accounts, this is the nature of capital – it gravitates toward itself. So if you want to have ‘land distribution’, you have to become a Marxist state.

So let’s talk about the practical problem with land reform in a business-friendly economy.

Land expropriation and redistribution: what do you do with the title deeds?

Zimbabwe is now a decade and a half into its land redistribution program.

Here’s a question: has land really been redistributed if the new ‘landowners’ don’t hold a title deed? Because Zimbabwe’s new farmers don’t have title deeds – they just have long term leases.

So here’s another question: why don’t they have title deeds?!

The trouble is that the standard narrative on land reform is half-baked. It goes:

• There is economic inequality!
• *takes back the land from commercial farmers and big corporates*
• *gives it to the rural voters*
• *wins next election*

Here is the longer narrative:

• There is economic inequality!
• *takes back the land from commercial farmers and big corporates*
• *gives it to the rural voters*
• *wins next election*
• *rural voters take their title deeds to the bank and apply for loans to finance their new farming operations using their title deeds as collateral*
• *unfortunately, rural voters tend to be new to larger scale farming, and their new farms suffer from the general pitfalls of being an undercapitalised, financially-stretched startup*
• *rural voters default*
• *bank takes ownership of land*
• *bank sells the land title back to commercial farmers and corporates*

Much like some of the concerns around BEE share schemes, land reform is easy to implement but difficult to maintain.

Private Property Rights can’t really be selective, unfortunately

It doesn’t help to override the private property rights of the commercial farmer and the corporate if you then allow the recipients of the land to have private property rights.

Because if you do that, then the normal process of success and failure in markets mean that the land will gradually return to the hands of the wealthy (even if they’re not the same people that owned the land originally).

And this market process is tilted in favour of the corporates and commercial farmers:

• Corporates and commercial farmers hunt out economies of scale and efficiencies in production;
• As they do so, they find ways to lower their customer prices (as well as reduce their risk exposure to weather elements and such);
• Small-scale farmers are then forced to match the price of the market – even though they have higher costs, less efficient production methods, and higher risk.

Sure, you may get exceptions.

But in general, if the State embarks down this road and wants to do so permanently, it means that the State has to adopt marxist-esque economic policies for both land ownership, price-setting, and support programs (subsidies) for the new farmers. That is: the recipients of land become tenants, not owners.

As the EFF is pressing for. And as Zimbabwe has discovered.

Zimbabwe’s solution now is to encourage the banks to lend against long-term leases, while the State directly owns all the land.

I’m not sure how that will work in the case of default – I guess the banks would then takeover the lease occupations and sub-let the farms?

Either way, we’re back to trying to solve the re-redistribution problem.

Where Land Reform Has Worked

Most people will hold up South Korea as a model of land reform. Some points on that (taken from this article):

• When the American Administration redistributed land in South Korea at the end of World War II (over fears of a communist take-over), it was not a case of taking land away from established farmers and giving it to new and inexperienced farmers.
• The established farmers were already farming the land – they just didn’t own it. And they were paying rents to their non-farming landlords.
• On top of that, after the redistribution, those established farmers were heavily subsidised with fertiliser, inputs and other credits.

The main conclusion: giving land to experienced farmers, together with effective financing, is a land reform policy that can make sense. For two reasons:

1. Agricultural production is not interrupted; and
2. The land reform won’t start to reverse itself through the credit-default process.

A second related conclusion: generally speaking, land redistribution policies have failed when

1. The new farmers are inexperienced; and
2. The land redistribution is under-supported in terms of financing and subsidies.

I guess the final question is: “Is South Africa another South Korea?”

If it isn’t, then land reform is not a long-term solution to South Africa’s inequality problem.

It would be, at best, a short-term (and destructive) once-off tax on the agricultural sector.

And why should South Africa settle for short-term solutions?

Rolling Alpha posts opinions on finance, economics, and sometimes things that are only loosely related. Follow me on Twitter @RollingAlpha, and on Facebook at www.facebook.com/rollingalpha. Also, check out the RA podcast on iTunes: The Story of Money.