Land and Agricultural and Development Bank of South Africa on Quarterly Reports

Meeting report

The Chairperson welcomed all those present in the Committee.

Mr M Molotho, Chairperson: Land Bank Board, gave preliminary remarks and introduced his team. 

Land and Agricultural Development Bank of South Africa Quarterly Reports Update for Period July 2020 – March 2021
Mr Ayanda Kanana, CEO: Land Bank, began the presentation by the Land and Agricultural Development Bank of South Africa. The Land Bank was unable to secure market funding due to its default status. The Bank’s liquidity status was causing constraints. Since the default and restructuring, it had not been business as usual at the Land Bank. A significant financial loss of R2.8 billion was incurred in 2019/2020. Since May 2020, there were negotiations to get out of default status and allow the bank to resume operations. Because there was ‘no way’ the Land Bank could handle the objectives for success, they required state support. They wanted the Land Bank to be more accessible to new entrants. The Land Bank defaulted in 2021 and was thus decreasing its financial support to various sectors. They needed collaboration for third party funding.

Ms Khensani Mukhari,CFO: Land Bank, said that in 2019/2020, there had been a disclaimed audit opinion, largely due to the default state of the Land Bank and the valuation of loans. Credit controls were another concern—particularly the input data for determining credit losses. There was an approved remediation plan to address the audit finding. The results of this would be declared in the upcoming audit.

Mr Kanana presented that, from the performance scorecard, five of eight targets were achieved. With executive committee processes, eight of eight targets were achieved. With non-performing loans, four of five targets were achieved.

See presentation attached for further details

Discussion
Funding, Intermediaries, Default Status
Mr D Ryder (DA, Gauteng) said they had a situation where the previous time the Committee met with the Land Bank it was decided they would be a hybrid commercial institution. Though, it was evident from the presentation that the best way for the Land Bank to survive was with bailouts or increased funding.  The problem with this was that it was clear government did not intend to put additional funding into the Land Bank over the medium-term. He asked for an indication, in the absence of Treasury, what was National Treasury’s stance on being profitable going forward? Funding going forward needed to be secured to continue operations, especially with an audit disclaimer which would make other funding difficult. What were the arrangements with intermediaries, of which 50% of loans were taken by such intermediaries? This was undesirable by a Department-funded initiative like the Land Bank. If the Land Bank was in default, was there legal action pending? What were the discussions like that were happening?

The audit disclaimer and R2.8 billion loss was concerning, especially since the AG had been warning for two years that the Audit Disclaimer was on its way. Mr Ryder asked Mr Kanana to speak frankly about whether the Land Bank was being used as a cash cow for “State Capture”. Operational expenditure was at approximately 20% but expenses were at approximately over 50%. Could they sort their funding model out? Would there be a moratorium on employee positions? Ultimately, were they dealing with something that was viable going forward?

Mr S du Toit (FF+, North-West) said the Land Bank might be viewed as a vending machine to send money through, as evidenced by what was presented. It was said that some loans which were defaulted on did not have any securities signed. What percentage of loans were honoured? What was the Land Bank’s current total debt? What was the current annual debt service cost? What was the total expected loss as a result of the wrong funding model being used? Where there were non-preforming loans, were there any attempts to recuperate the funds lost? If so, to what extent?

Mr Z Mkiva (ANC, Eastern Cape) said that the Land Bank needed to acknowledge that South Africa was predominantly a rural country, so that they were informed by the reality on the ground. They needed interventions beyond commercial farmers. 

The Chairperson said that the Land Bank was indispensable. Even though having a Land Bank was not turning out to be an asset, a Land Bank was essential going forward.  He questioned whether the reason for so many achieved targets was due to limited or easy targets being set. He asked how the Land Bank related to the Development Bank and what level of co-operation there was between the two. Given the Constitution and the Expropriation Bill before Parliament, what would become the role of the Land Bank? Would their role not become enhanced as land reform was ramped up? What was National Treasury proposing in terms of support, since the Land Bank was so reliant on being rescued by the state? He asked for a sense of their trajectory and the pre-conditions for support going forward?

Responses
Mr Moloto delegated the questions raised to members of his team.

Mr Faride Stiglingh, Executive: Portfolio Management Services, Land Bank, contextualised the Auditor-General’s findings and disclaimer on the basis of a lack of a controlled environment. This was related to service level partners. The Land Bank had a direct and indirect portfolio. Six to seven Service Level Agreement (SLA) partners performed intermediary services in the indirect portfolio. For these services, they were given a remuneration structure with a fixed fee and margin share based on the margins that were made on the disintermediation of the funds. They also shared appropriately for the risk to the portfolio from which it originated on behalf of the Bank.

In 2010, this was structured in such a way that intermediaries were contracted to perform the services under the auspices of their own infrastructure. This meant that intermediaries needed to invest in the infrastructure to originate, manage and monitor loans. Intermediaries also had to manage the portfolio of non-performing loans to the extent where if the loan deteriorated, the intermediary was responsible for the entire process of the end-to-end valuation from origination to recovery the processes.  This was very important context to the AG’s findings. This particular contract enabled the bank to hold the intermediary reliable for the management of those functions. The intermediary in return needed to make sure they had the requisite infrastructure from both the people process and systems, as well as governance processes in place to effect service provisions (support and managing the loans).

For the last five to seven years, those particular functions were under signed agreements between the Land Bank and the intermediary with the understanding that the governance controls and protocols were to be included. When the AG stood in, they were expected to audit those processes in accordance with the signed agreements and prevailing policies adopted by the Land Bank and intermediaries. These procedures were found to be adequate and compliant with the policies and agreements in place.

In the last three years, when the AG stepped into the Bank to review the process protocols, the AG found that, in 2018/2019, protocols were again performed in accordance with the signed agreements. However, the AG found that the Bank needed to standardise their processes and protocols and align them with that of the Land Bank. At the time, the bank operated on two fronts: direct and indirect (SLA) portfolios. The SLA portfolio operated on their own protocols as opposed to the embedded protocols in the direct portfolio. This is how that audit finding had come about. The AG felt it was important to align the policies, operating environment and audit frameworks to ensure that the SLA’s accorded with the latest accounting provisions in as far as what the Bank had adopted. There were substantial differences in the way the Bank managed its portfolios versus that of the SLA’s. This had led to the 2020 AG finding, where the AG believed the differences had become elevated and material (pertaining to risk)—thus issuing a disclaimer to the Bank indicating the it needed to remediate particular differences quite extensively to ensure full alignment with operating protocols of the bank.

Based on the AG's findings, the Land Bank embarked on the SLA Improvement Plan. The Executive and Board had identified these particular concerns as early as January 2019 (before they were raised by the AG). The SLA Improvement Plan would remediate the challenges. The Land Bank could not make unilateral challenges without reaching consensus with the SLA counter-parties. As such they were in a re-negotiation process. Where this was near impossible (due to cost and lack of willingness) the Lank Bank adopted a harsh stance to terminate SLA’s. In December 2020, close to 80% of SLA portfolios were terminated due to lack of compliance.

Non-Performing Loans
Regarding non-performing loans (NPLs), and whether they could be related to the misappropriation of funds in the bank or looting, both direct and indirect portfolios had compliance processes in place where new clients or entrants taking up facilities with the bank were subject to protocol for the clear delineation of responsibilities. This was referred to as the three lines of defence – where duties were separated with clearly defined policies. Credit applications, with supporting documentation needed to be compiled with the farmer’s ability to pay debts (first line of defence).

Front-end bankers, with the support of business-level analysts, were used to support any potential farmer wishing to enter the market in compiling their particular credit application (first line of defence). This information (credit application with supporting documentation) would be compiled with a motivation of the entity’s ability to afford the debt it was applying for.

Once this application was consummated from the origination, it was dispatched to a centralised credit origination environment within the Bank (second line of defence). Once the transaction was packaged in accordance with all the government protocols, this was passed onto the credit environment where all all the assumptions would be tested, as well as whether the transactions originated in a sound manner. After testing of the client and transaction, the application was presented at to the credit adjudication committee of the Bank. The Bank was organised in a manner that stratified (organised) based on the size of transactions and this was referred to a specific credit committee.

Sizeable transactions were evaluated by the Bank’s credit risk management committee, a group of specialised analysts, representing different disciplines in the Bank, as well as specialists representing different commodities and managing risk. The entire executive served on this committee as well to ensure proper adjudication. It was important for the delegation of authority levels to be defined along with the separation of duties and different lines of defence.

As it related to NPL’s, when a client failed to meet their annual or bi-annual payment, the customer would become a delinquent. Though they might have been up-to-date before, they would accrue their first “mis-payment” based on their contractual arrangement. Those arrears would be reached by the post-investment portfolio managing services division to ensure the separation of duties. A first payment default was when a client was underperforming. After 90 days of not paying, the client was moved to the NPLs – which were managed on an accelerated basis. Through soft interventions, there was an attempt to get the client to pay. They would move to hard interventions of debt collection, for instance, threatening with legal action, restructuring, advancing capital moratoriums. Summons and attachments were the subsequent actions taken if still not successful in the recovery of underlying assets. They made sure they recovered assets, and that properties were taken to auction. In this process, they worked with an internal and outsourced (external) legal team.

In response to the question of misappropriation that, there was no misappropriation of funds through the NPL: “certainly not”. Funds were not being mismanaged, either. Rather, they were sitting with rising NPL rates either because of poor prevailing agricultural conditions such as drought and famine disease, or economic difficulties such as COVID-19, rising interest and inflation.

NPL Management Strategy, Repricing of Loans, DBSA
Mr Kanana said they had spent time in one of the previous quarters trying to understand the root causes of entrants being so high. The intention was to ensure they could come up with a response to keep this number low. They assessed, in particular, how farmers were managing their farms. Most of these were family-run farms, with the expenses from the farm itself. The division between business and personal expenses were sometimes blurred. Organisations were needed to put management and accountability structures in place to prevent this from happening. 

As part of the Bank’s SLA improvement process especially, there were loans issued as a ‘loan on a loan’ which should not have happened. When disposals were signed-off after auctions, for example, it was only then discovered that the debt was fairly high, yet the ‘for sale’ value implemented was very low. They had spent effort working out how the Bank would protect against having sales ‘just to get rid of a particular farm’. Their conscience was not clear when they signed off ‘just to get rid of a client’. To make sure they turned the NPL process around, they created the packaged programme, where they went in and assessed a client based on what was available to them, that is, whether it was due to drought or mismanagement of received funds. They also owed the board a business course around property management.

The Land Bank needed to be able to take some of these loans from farms and put into their balance sheet and look after them: either to sell them at a later date or to make them available freely later. To do this, however, money needed to be spent on setting up these processes so that they could secure farms. A key challenge was that some NPLs stayed within the legal system for a long time. Often by the time they needed to claim against a particular farmer, a lot had deteriorated, and loose tools were no longer there. There had also been issues of ‘back to basics’, where Land Bank staff simply had to perform their tasks, such as calling the farms, pressure them to pay and demand what was due. This was an aspect of recovery where they did not need money, but rather to simply do their work, preventing NPLs arriving or staying longer in the system. In the previous quarter, the Land Bank had seen many farms roll into default. There had also been cures, where systems were able to recuperate and prevent farms from going into NPL status.

Land Usage
The use of the land to buy a farm and pay for production inputs would cripple any new entrant. The farmer was immediately rendered unsustainable and likely to get into a default situation. The land needed to be used without necessarily having to buy it. If communal land was available for such a purpose, they should use it productively and profitably to ensure agriculture sustained the particular community. This should be rolled out country-wide. Though the Department of Agriculture had a programme of leasing land to communities, these very communities came back to the Land Bank for investment to work the ground. There was a need to be able to respond to particular programmes.

In response to the question about strategy, they could provide a framework and would continue working towards ensuring that resources were supported for rural development.

Collaboration with the Development Bank of South Africa
Mr Kanana was not aware of any collaboration with DBSA, at least in his time at the Land Bank. Though, there might be collaborations signed-off. For example, when the Land Bank’s liquidity situation was particularly problematic, they contacted the Industrial Development Corporation (IDC), who had been helpful in ensuring the sector had support.

Debt, Debt Services Cost, Initiatives to manage
Ms Mukhari said that of Land Bank costs, 60% was staff salaries. Essentially, there was a fine balance between keeping the Bank operational and reducing costs. The Land Bank, for example, had extremely manual processes. Normally this meant that manual processes were compensated by people (labour). There was work underway. From the containment perspective, as an example, there had been a moratorium on filling vacancies, with  a strict process followed that vacancies filled needed to be signed off by the CEO. They were also intentional about looking internally first, as well as evaluating where some employees could absorb tasks required. Other operating expenses were watched and managed very closely to ensure that the Land Bank only spent what it absolutely had to. The Land Bank was in the process of insourcing and managing its IT system. The Bank had taken steps to manage costs in the immediate future, medium (5 years) and long-term. As at end March 2021, the funding liabilities were at R9 billion (this was after the Land Bank was able to pay R5 billion). The cost of servicing debt was approximately R3 billion.

Operating Model: Organisational Structure and Cost Containment
Ms Dudu Hlatshwayo, Deputy Chairperson: Land Bank, indicated that the business operating model had to be informed by the strategy and vision of the organisation. The Land Bank had drafted Vision 2025, where they aimed to develop and support transformation in earnest. The Land Bank also had a broader interest to serve the sector as a whole so that food security was ensured. As much as the board was intent on containing costs, two important issues needed to be resolved: the liability serving needed to be agreed upon and in-sourcing of the SLA books needed to be addressed. Four of six SLA books were in-sourced. In view of the in-sourcing model embarked upon, more resources within the Land Bank were required in order to handle the in-sourced books of the SLA’s. The Land Bank believed it inappropriate to start re-organising processes while these activities were taking place. In October 2020, the Land Bank Vision 2025 had been discussed as the overarching strategy, underpinned by the liability solution which was yet to be finalised. The Land Bank believed this liability solution was one of the biggest dependencies they needed to take into consideration before re-organising themselves. The Bank had also ensured it contained costs by not filling positions that had not been categorised as critical and/or essential. They had also not paid short-term bonuses, nor annual salary increases in the past few years. She confirmed that the board was keen to ensure they retained costs and that they had a fit for purpose organisational structure going forward. They were still busy with insourcing with the books of the SLAs. They would need large investments in future. As such, they were ensuring they did not lose critical skills so they did not need to be sourced from the market. They were doing their best to work with what they had.

Goals of Land Bank, Expropriation Bill, Land Reform Process, Approach
Mr Sydney Soundy, Executive Manager: Strategy and Communication, Land Bank, said that the financial performance presented (per slide 24) indicated the context of deriving no viable solution to get themselves out of default and into normal business operations. Critical to the board and the shareholder was that during this time, as they tried to get the solution in place, they needed to ensure the Bank’s other operations were appropriately stabilised so that when the opportunity to turn around came, the fundamentals were still in place. The scorecard as presented was thus largely developed out of the desire for stabilisation. They wanted to be ready for the potential turnaround. This included looking for partnerships in future that would ensure funding and that private entities were brought to the Bank to support the sector. Essentially, the scorecard elements dealt with these issues. Financial performance was on a significantly negative trajectory and was not supporting the sector. This had an impact on shrinkage and declining margins. This needed to be managed. There had been a loss of R2.8 billion in the previous financial year. They projected an unaudited R1 billion loss. This could be juxtaposed against the scorecard in relation to financial performance and localisation.

A good land reform process would have benefits for the economy in general. This included the Land Bank, who had been part of negotiations and advice relating to reform. At a broad strategic level, a reform programme which provided more accessibility to productive land would be welcome. The process was being handled very responsibility: there was an appreciation of the rights of creditors (collateral health especially, should this be challenged) become accepted. The land reform process—through ownership or tenure—required support from financial institutions. If lease agreements were for a longer period of time, investors looked at what it would take in terms of time of tenure to generate income and profitability. Unless there was a proper support system to ensure that beneficiaries were supported to be successful in their operations, the land reform process would not yield the objectives it needed to achieve. It was important that whoever was a beneficiary to the land reform process had the appropriate state machinery and support infrastructure, from the point of project preparation, to help new beneficiaries with their business plans, technical support and other services. There was leveraging of blended institutions, along with bringing in market linkages, so that the produce of beneficiaries went into the system so they could generate the income they required. The land reform question was one that needed to be assessed holistically.

Control Environment
Ms T Makgatho, Chairperson: Board Audit Committee, said that given the magnitude of the issues raised by the AG, it was their goal for the year to improve audit outcomes. If they were to follow the normal audit committee programme, it would be three years to fully resolve all issues being raised. They did not have three years, or even a year. The Land Bank gave themselves six years to get to a point where they had improved audit outcomes.  To ensure that the Bank’s internal controls were approved, the Land Bank was closely monitoring what the team was doing. They were also providing guidance on a regular basis. Out of 55 issues raised by the AG, 33 were closed and 12 partially resolved. Those partially resolved related mostly to standardisation and restructuring of debt. The recalibration of models and the re-evaluation of 1 000 properties was also taking place. They were doing their best to ensure that when the FY2021 report was finalised, they had better outcomes. There was absolutely nothing that had come to the audit and finance committee about state capture or corruption at the Land Bank or within the board. The challenges they were faced with were more mundane. They were confident that the report which would be tabled next would be much better.

Planning and Process
Ms Tshepiso Moahloli, Deputy Director-General: Asset and Liability Management, National Treasury, said that from the budget process, Treasury did not unilaterally decide; there were various bodies for approval. National Treasury had the responsibility not only of facilitating the budget, but was also responsible to shareholders. There was a resource scarcity at the government level, which fomented low confidence because of the default. The question became about the scale and pace at which the government supported the Land Bank. Going forward, there would be other requirements in the medium- to long-term which would have to support the Bank. This was why the Minister of Finance was not working alone as a shareholder but had brought in the Minister of Agriculture to have policy aligned with the kind of Land Bank they needed, given the importance of agriculture in job creation and food security. There had been other requests on guarantees. Effectively, a guarantee was an appropriate tool for an entity that was able to repay. If an entity was not able to pay, the guarantee was not the right tool as it would undermine the integrity of the budget. The Land Bank would have to be supported through this process.

Mr Ryder asked via the online chat: “As an entity in this Department, you must have a strategic direction in mind.” He acknowledged that this was a political question. He also appreciated how comprehensive and well-put together the report was.

Mr du Toit asked via the chat: What is the total expected loss as a result of the “wrong pricing” model? He also asked in the meeting how would land belonging to government be handled if the individuals defaulted on their payments.

Mr Moloto said that their approach had been that land belonging to the community could not be seized and sold. They needed to have statement right, which empowered them to evict the tenant and demand that the tenant be replaced by a viable business. Communal land could not be seized. He assumed the same would apply to land that was being held and the Land Bank was approached for assistance. The leasing was critical here.

Total expected losses, pricing model
Mr Stiglingh explained that they used to price loans while procuring funds through debt and capital markets. Those particular funds were advanced to the Bank at a particular cost. When these funds were dis-intermediated, they made sure they added a margin to the cost of funds so they could ensure they generated profit. Historically, they had always managed to dis-intermediate funds above the cost of funds with a healthy return for the Bank (both direct and indirect portfolios). In the previous two years, their cost to funds had radically increased. This meant that funds which were locked in agreements where they could not advance pricing, were running those facilities at a loss. However, in their repricing strategy, they adopted for an aggressive approach. Portfolios due for repricing were adjusted in accordance with the latest cost to funds to ensure profits and minimal revenue leakage. Their repricing strategy for any funds to be dis-intermediated to the sector was a model which accounted for the latest changes in the short-, medium- and long-term nature of the cost of funds. They had historical portfolios which were incorrectly priced, but they were remediating this to mitigate losses to the Bank.

Total expected losses (cost)
Ms Mukhari said that, with the wrong pricing structure as described by Mr du Toit, she would take a step back. The challenge that the Land Bank faced was not so much about lending at a lower rate, but that there were major contracting issues. The way they contracted did not allow the Land Bank to pass on the full increase in the cost of funding to customers. Even though the pricing was in a model, even the slightest movement meant the cost of funding was impacted, and compressed margins. With their default, the unpaid funding liabilities incurred interest. The Land Bank had reviewed their pricing model. The process was signed off and implemented. Approximately R11 billion (particularly in the SLA space) was identified for re-pricing. This was one of the factors influencing in-sourcing books for intermediaries. This was thus more an issue with contracting than the pricing structure.

The Chairperson said they had covered much ground, thanks to the colleagues. He suggested that, in the second half of the year they summoned the Land Bank back (this depended on available time with local elections). Ultimately, they wanted the meeting to be reviewed before they re-invited the Land Bank staff so that they did not cover the same ground in the next meeting but, rather, that they moved forward with a framework by early August.

The meeting was adjourned.