The following article by Wandile Sihlobo and Johann Kirsten was published in Business Day on 30 July 2018. Sihlobo is head of agribusiness research at the Agricultural Business Chamber and Kirsten director of the Bureau for Economic Research at Stellenbosch University.
The expropriation of commercial land and farms could negatively affect food security and agricultural growth, write Wandile Sihlobo and Johann Kirsten.
In its 54th national conference report and resolutions, the governing ANC highlighted that the interventions regarding expropriation of land without compensation would largely focus on government-owned land, prioritising the "redistribution of vacant, unused and underutilised state land, as well as land held for speculation and hopelessly indebted land".
It remains to be seen if commercial land or farms will be part of the expropriation process. Tampering with such areas would potentially have a negative effect on food security and agricultural growth — an outcome that the ANC is trying to avoid.
Meanwhile, the EFF opposition party argues that all land should be nationalised, or "wholesale expropriation without compensation" should be applied. In such a scenario the effect would not only be limited to land for food and agricultural production, but would also — according to its proposal — include land for housing and for industrial and retail use. We argue in this article that this strategy will by default destroy the asset value of a large portion of SA's land and could by definition also have a large negative impact on financial institutions and the property market.
A bit of background: in March 2018, outstanding bank credit to the private sector (businesses and households) totalled R3.5-trillion, according to the South African Reserve Bank June 2018 Quarterly Bulletin. Of this, mortgages accounted for 39% (R1.4-trillion), with households accounting for 68% (R929bn). Put into context, the amount of mortgage exposure (households and corporates) that the banks have is equivalent to 29% of South African annual GDP (at March 2018).
This includes, predominantly, credit extended to buy houses and vacant land for building a residential structure. In SA, the general practice is that banks fund up to 40% of the acquisition of vacant land — which happens to be in line with the areas identified for expropriation within the ANC documents. In the case of free-standing houses, the value of land is built into the selling price, while flat or apartment owners generally have an undivided share in the land on which the structure is built, which is owned jointly through a body corporate.
Furthermore, it is estimated that about 70% of all residential property transactions in SA involve freehold property. This implies that a large majority of housing transactions include, directly or otherwise, private land acquisition. The state assuming ownership of all land without compensation therefore means, from a consumer perspective, a loss of the land component of the acquisition, while retaining the ownership of the building structure. As the law reads, the land portion cannot be disconnected from the immobile asset such as a building or a house in SA's application of property law.
For an average household in this country, the property represents its largest investment from which its members derive wealth.
At an aggregate level Reserve Bank data show that net wealth (the value of residential buildings minus mortgage advances) derived from residential buildings as at December 2017 was about 16% of households' net wealth. This, however, conceals the true contribution of residential property to households' balance sheet. This is because pension funds, which is households' biggest component of financial assets, also invest in property via the equity market. This excludes households themselves investing directly into the equity market and also indirectly via other investment vehicles such as unit trusts.
An aggregation of all this shows that the destruction of all property value would have serious implications for SA's national asset base and the foundation of the economy.
Hence, we have continued to argue that a wholesale or blanket expropriation of land without compensation policy or approach could be viewed as a destruction of land value, some of which is financed by debt.
The often cited figure is that of agricultural debt, which was estimated at R158bn in 2017, according to data from the Department of Agriculture, Forestry and Fisheries. But, if we consider the aforementioned components of the economy, the effect could be much wider. The exact value of this component remains unclear, but as demonstrated by the numbers above, it is likely to be significant.
In terms of the agricultural debt, the effect would not only be felt by the commercial banks. The government also has "skin in the game" through the Land and Agricultural Development Bank of SA (Land Bank), which accounts for nearly a third of agricultural debt.
The balance is accounted for by commercial banks, agricultural co-operatives, private persons and other institutions.
The Land Bank presents an interesting picture in terms of its exposure in the agricultural land and long-term credit market. At the end of 2017 the long-term loans and loans secured by mortgages owed to the Land Bank was worth R16.2bn (of which R8.5bn was for individual mortgages). The Bank has also provided cash advances to agribusiness and co-operatives equal to R25.5bn. With some other advances and loans, the total assets are equal to R46.56bn, which is then financed via the bank's liabilities (mainly promissory notes and Land Bank bills).
Many of the cash advances to agribusiness are also on-lend to farmers to acquire farm land, which implies that the farmers' land exposure of the Land Bank and the agribusiness could easily be at least 50% of the Bank's total asset base. A land policy scenario described above will therefore also risk the Land Bank's financial stability and one could foresee that roughly R20bn-R30bn of the state budget will have to be used to save the bank.
In terms of section 8 of the Expropriation Act, an expropriation will extinguish a mortgage bond, but not the debt.
“Considering these components, the question then becomes, should a property owner continue servicing their loan when they no longer have ownership rights to that property or a component thereof, and the bank has no security to fall back on?"
Simply put, if the land is expropriated, the owner still owes the bank, but it becomes an unsecured loan (which would typically be associated with a higher interest rate).
Considering these components, the question then becomes, should a property owner continue servicing their loan when they no longer have ownership rights to that property or a component thereof, and the bank has no security to fall back on?
Furthermore, should financial institutions — and real-estate corporations — simply write off their assets on their balance sheet?
Wholesale expropriation of land without compensation therefore could be likely to trigger a major devaluation of financial institutions' assets, and ultimately their balance sheets. (The blanket expropriation approach is somewhat different from the ANC's official views, as outlined in its documents, but more in line with the EFF's position.)
Corporations are valued on the strength of their balance sheets, which affects their ability to raise capital and fund expansionary projects. Ultimately, this could trigger a disruptive stock market repricing. At the same time, the nationalisation scenario could trigger a liquidity risk in the Land Bank and other commercial banks. Given the size of the outstanding debt illustrated above, it is unlikely that the state will be in a position to financially rescue these institutions.
In our previous article published on July 9 2018, we demonstrated the value of property rights, using land as an asset, and its role in a market-based economy such as SA's that is an integral part of the global economy. The cases presented show that in the event of the nationalisation of land, the potential beneficiaries will not be able to build wealth without assets anyway. We further argued that nationalisation will not enrich anyone, but will rather be a nightmare for the state and its citizens.
The aforementioned implications of the linkages of land to the overall economy suggests that the negative consequences could be rather far-reaching and would not yield any value to potential beneficiaries of the process anyway.
These arguments do not imply that we should not urgently deal with the inequality in land ownership. It is necessary that we implement a decentralised and effective land-reform programme to restore land rights to the majority of our people.
There is, however, a much more responsible solution that will not destroy our financial sectors, our pension funds and our economy. We will share views and ideas regarding a possible workable concept in our final article of this series in Business Day in two weeks' time.
