Swap shop
By November 1998 the bidders had been reduced to two consortia who were invited to submit BAFO bids:
? The Bakwena Platinum Corridor Consortium (BPCC), led by South African contractors, Murray & Roberts, with Concor, an associate of the Hochtief group, and WBHO Construction and SG as lead financial advisor; and
? Tselo Ya Batho Concessionaire, which was led by Cintra/Ferrovial, of Spain, and the South African contractors, Stocks & Stocks, and Kagiso Financial Services as lead financial advisor .
Certainty of deliverability
The bidding process for the project spanned some of the most difficult times experienced by South African financial markets, including the emerging market crisis of 1998 when short term interest rates peaked at 25% and long dated government bonds were trading at yields of 22%. Interest rates at these levels seriously jeopardised the financial viability of the project and although by the time the BAFO was being prepared, interest rates had returned to their previous levels of around 16%, it was clear that SANRA needed to be convinced that the proposed funding could be delivered.
Prior to BAFO submission the sheer size of the project was becoming a concern to the parties involved and in December 1998 Group 5 ? a South African construction company which had been eliminated in the previous round of bidding ? joined the Tselo Ya Batho bid team. Co-extensively, Dragados joined BPCC in a leading role and the debt underwriting capacity of BPCC was also increased with the addition of two further local banks.
Debt
Inflation linked debt
The project's revenue profile was typical of toll road concessions, commencing at a relatively low level and then steadily increasing over time as traffic volumes grow and toll rates escalate with inflation. Accordingly, securing underwriting for a significant tranche of inflation linked debt was necessary to match debt service to the revenue profile and was also beneficial for hedging purposes. This requirement was exacerbated by the insistence of the South African banks providing term loans that high minimum annual debt service cover ratios be achieved without any principal grace periods beyond the initial construction period and that principal be repaid on a straight amortisation basis, without any sculpting of the debt repayment profile. The graph in Diagram 1 illustrates indicative debt service profiles of equal amounts of term loan debt and index linked debt:
Another benefit of inflation-linked debt is the lower interest carry during the construction period, as the running yield commences almost at the level of the real coupon, which reduces the overall project costs and consequently also reduces the funding requirement.
Inflation linked debt is not a typical banking product and is really only attractive to long term insurance companies and pension funds which have liabilities linked to inflation. The main difficulty in securing underwriting for a tranche of inflation linked funding was the relatively undeveloped state of the market in South Africa at that time and the lack of comparable benchmarks during the bidding process. In addition, the South African banks that had underwritten a tranche of inflation linked debt for the Maputo Corridor project, had been unable to place it and consequently were not prepared to underwrite any further inflation linked debt and take the risk of placing it with third party investors.
To overcome these difficulties a new and innovative swap mechanism was developed. BPCC took the placement risk on the inflation linked debt, but was able to benefit from the reduced service profile, equivalent to the debt service of an inflation linked bond, during the crucial early years of the concession. In terms of the swap mechanism (see Diagram 2) any portion of the inflation-linked debt which the underwriters were unable to place would swap into ordinary variable rate senior bank debt with retrospective effect after a five year period. The effect of the swap would be as if the banks had held conventional variable rate debt from the original issue date, although, a portion of the term loan interest due would have been capitalised during the five year swap period.
By financial close the market for inflation linked debt had developed significantly, with two sovereign issues and one project issue, and the inflation linked debt for this project was oversubscribed. Nevertheless, the swap mechanism gave BPCC an important competitive advantage by securing the benefit of inflation linked debt in the bidding process.
The European Investment Bank (EIB)
The inclusion of a tranche of funding from the EIB was considered attractive as it offered:
? lower interest rates;
? an increased grace period;
? a sculpted repayment profile; and
? a longer tenor.
However, although the EIB was prepared to take South African country risk, it was not prepared to take project risk and required a project guarantee. Initial discussions with the South African banks were not promising as they preferred to grant 20 year term loan funding directly to the project, rather than provide a risk guarantee to a third party and were not prepared to accept any increase in exposure compared to their direct lending with a straight amortising profile. Discussions were also held with a number of major European banks, however, none of them were able to get comfortable with the EIB definition of ?political risk?, which covered all major political risks such as expropriation, nationalisation etc, but did not extend to risks unacceptable to international banks, such as a gradual deterioration of law and order in the country.
This presented BPCC with a major dilemma as the banks supporting the competing bid had guaranteed a 25 year tranche of funding from the EIB on the N3 Toll Road Concession and it was suspected that they would take full advantage of a similar tranche of EIB funding in their funding proposal for this deal. As BPCC were unable to match this potentially competitive 25 year EIB funding, a decision was taken to exclude EIB funding from the BAFO, on the basis that it could not be assured with the requisite level of certainty. In addition, the following difficulties with raising a tranche of EIB funding were emphasised, in an attempt to undermine any EIB funding included in the competing bid:
? The EIB Second Framework Agreement from the European Union to finance South African infrastructure and development projects was fully committed at that time and further lending would be subject to the EIB securing a new mandate;
? The EIB was then only able to guarantee the availability of Euro or US dollar based funding, which was not attractive to BPCC, due to the costs of hedging the currency risk. At that stage the EIB did not operate a rand treasury and their ability to offer rand funding would be dependent on their ability to raise back-to-back funding in the EuroRand market or the swap market and the availability of this source of rand funding could not be guaranteed; and
? Although the EIB had previously accepted guarantees from South African institutions, it was not clear that they would again accept such guarantees.
Nevertheless, EIB funding continued to be pursued and the option to utilise EIB funding was included in BPCC's funding package, under a multi-option facility negotiated with the South African banks whereby the Senior Bank Debt facility could be drawndown, either as term loans or as guarantees for the EIB.
Underwriting
BPCC and its lenders made a considerable effort to provide as much support as possible to the financing of the project:
? The lenders legal advisor undertook a comprehensive review of the Concession Contract and the BAFO Contract was substantially in a form that sponsors and the lenders were prepared to sign;
? The debt term sheet was negotiated at length with the lenders and their legal advisors, such that it incorporated all the principal terms and conditions required for the final loan agreements; and
? The lenders traffic advisor undertook a detailed audit of the traffic study, which formed the basis of the traffic and revenue projections for the Banking Base Case financial model of the BAFO submission.
The Underwriting Banks obtained full credit approval, from each of their respective credit committees, to underwrite the debt funding required for the project and this enabled BPCC to submit a BAFO bid that had 80% of its debt funding underwritten.
Equity
One of the main concerns regarding the financing of the project had been the availability of the required levels of equity, which was exacerbated by the fact that the previous toll road projects had been smaller in size and were also funded with higher debt: equity ratios
Due to the perceived higher risk of this project, the lenders were insisting on a debt: equity ratio of 75:25, with the sponsors required to hold 40% of the total equity for a minimum period of seven years.
The substantial Sponsor equity requirement was one of the main reasons for the inclusion of Dragados in the bidding group at the outset, and the financial support, credibility and extensive experience of Dragados enabled BPCC to submit a bid with over 70% of the equity requirement underwritten. It was relatively easy to demonstrate more than adequate appetite for the remaining 30% of the equity with letters of support from various funds and institutions.
A robust and resilient project
The requirement for a robust project and funding structure, able to withstand extreme downside scenarios, was a critical factor for all parties involved in the project:
? SANRA needs to establish a track record of successful projects in the market to enable it to continue to attract strong bid for future projects;
? Sponsors and other equity investors substantial investment would be forfeited if the project were to default; and
? Lenders were no longer fully guaranteed on default, as had been the case on the Maputo Corridor project.
Sensitivity Analysis
A consideration of the Project's sensitivities to various risk factors was an integral part of the detailed risk analysis and risk allocation process, which was carried out prior to BAFO submission. This was supported with detailed sensitivity analysis to illustrate the critical sensitivities and to demonstrate that the Project was bankable and robust.
This analysis identified that the greatest threats to the project were:
? Decreases in revenue as a result of lower toll rates (i.e. increased levels of discounts and concessions) as this did not entail any reduction in subsequent expansion and heavy maintenance works, which risk was addressed via the toll strategy and the sponsors providing standby equity; and
? Decreases in inflation rates without a corresponding decrease in interest rates, which was addressed by the hedging strategy discussed below.
Overall, however, the sensitivity analysis showed that the funding structure was extremely robust and able to repay all its debt within the concession period in all but the most extreme downside scenarios tested.
Traffic and Tolling
The Platinum Toll Highway project is totally reliant on revenues generated from real tolls, and does not have any direct or indirect subsidy payments from the public sector. In addition, this was the first project in South Africa where all termination compensation, for the various termination events, was based on different variations of the net present value of the project. Accordingly, one of the key elements in formulating a winning bid was the formulation of a toll strategy that took account of public sector policy issues, was sensitive to the demands of the toll paying public, yet sufficiently robust to convince lenders that their debt was secure.
To achieve these objectives, the toll strategy was based on the following underlying principles:
? Minimising the impact on local trips in rural areas;
? Maximising the revenue from long distance trips;
? Minimising the impact on previously disadvantaged communities;
? Tolling the traffic in urban areas which would benefit most from the improved roads; and
? Positioning toll plazas on new sections of road wherever possible.
An additional complication was the profile of traffic users, which combined urban commuters in the Pretoria region with more rural traffic on the extremes of the N1 North and the N4 West. The proportion of revenue anticipated from the urban commuters, given the prevalence of alternative routes in the Pretoria Metropolitan area, was of particular concern to the lenders, and was addressed by appointing the lenders traffic advisor prior to the submission of the BAFO and the development of a detailed traffic model for the entire road network.
The final traffic model comprised of 602 traffic zones (of which 570 cover the Pretoria area), 9,165 road links and 4,120 network nodes, and is of such size and complexity that BPCC had to commission the software suppliers to produce an upgraded version to handle all the links and data. The traffic model proved to be a robust tool for assessing traffic diversion resulting from the imposition of tolls on the highway network and an effective tool to optimise and test alternative tolling strategies during the concession period. It also proved useful in allaying the concerns of the lenders with regard to the impact of competing routes and the sensitivity to underlying key variables.
Hedging Strategy
It was difficult to predict future inflation rates as historically inflation had been between 10% and 15% per annum. But in more recent times the government has been successful in reducing inflation and was targeting inflation of between 3% and 6% per annum.
Packaging this inflation risk was particularly important to mitigating the concerns of the lenders and investors, but at the same time, ensuring the competitiveness of the bid as pricing of the future inflation risk was prohibitive, particularly as a result of the historically high inflation rates. The inflation risk was managed by structuring the project on the following basis:
? A fixed real price contract for the initial construction works was concluded, with escalations linked to inflation;
? Payments to the O&M Contractor were also linked to inflation; and
? About a third of the required debt financing was underwritten as inflation-linked debt, subject to the swap mechanism agreed with the Underwriters.
However, a substantial portion of the project's expenditure was still not linked to inflation and if inflation was lower than projected, revenue levels would be reduced. Accordingly, the conventional wisdom of locking into fixed interest rates did not apply, as fixed interest rates combined with the government achieving its targeted low inflation, would result in BPCC being squeezed between the high fixed interest rates and lower revenues as a result of lower inflation. The longer the period of any fixed interest rate hedge, in this scenario, the greater the negative impact on the project.
As a result a hedging strategy was formulated to achieve the following objectives:
? Protect the project from going into default in the event of a sudden increase in real interest rates in the short term, as has been experienced in crisis situations;
? Enable the project to continue, without going into default, in a low inflation scenario; and
? Ensure that BPCC would have sufficient funds available during the initial construction period, under any reasonable interest and inflation scenario.
Empowerment
The empowerment of the previously disadvantaged is an important objective of the South African government, and for high profile projects such as the Platinum Toll Highway, it is important to provide for empowerment involvement in all aspects of the project.
One of the most difficult aspects of empowerment for concession projects is the raising of empowerment equity, because of the back-ended nature of the returns and the complex upfront construction and traffic risks.
Even though toll road projects are not generally attractive to empowerment investors, BPCC succeeded in including a number of well established and credible empowerment investors.
Some of these investors were particularly concerned about the potential downside risk in the early years of the concession, and because of the importance of empowerment participation to the bid, the structure set out in Diagram 3 below was devised to enable empowerment investors to mitigate these risks and to insist with raising finance, if necessary. The features of this structure are as follows:
? The put options provide protection for the empowerment investors downside risk;
? The call options allow the Sponsors to call a portion of the empowerment investor's equity, in upside scenarios; and
? If necessary, the Empowerment investor could raise funding on the strength of the shares and the put options .
Aggressive bidding
All the consortia pre-qualified for the tender were strong capable groupings and it was clear that a winning bid would have to be lean and aggressively structured in all respects.
Optimising the construction programme
The initial construction cost of the Platinum Toll Highway was not evenly spread over the length of road, as most of the initial cost was involved in upgrading the N1 east of Pretoria and constructing a new N4 alignment north of Pretoria. Similarly traffic levels ranged from 60,000 vehicles per day around Pretoria, to 2,700 vehicles per day in the more remote rural areas.
The map overleaf provides an outline of the route and new road works.
As the opening of toll plazas on each sector was permitted only once the necessary upgrade and rehabilitation works on the relevant stretch of road had been completed, construction scheduling was an important factor in optimising the financing. The road was divided into seven separate toll sectors and, using the financial model, a detailed analysis of a number of scheduling options was conducted, essentially weighing up the revenue potential of each sector against the carrying cost of the construction works together with various practical considerations, such as:
? Contractor mobilisation and the physical location of the various road sectors; and
? The use of ?cut? from certain sectors as ?fill? in other sectors and the avoidance of the related quarrying costs.
Optimisation of the scheduling of these various construction works had a material impact on the net funding required for the project and accordingly the overall competitiveness of the project.
Highway Usage Fees (HUF)
The Concession included provision for a HUF to allow SANRA to participate in the upside of the project. All previous projects had been structured with a HUF linked to levels of equity returns, which meant that any benefit accruing to SANRA would only be towards the end of the concession period, after the target IRR had been achieved.
BPCC proposed a new aggressive HUF linked to ongoing cumulative revenue receipts which would, in an upside scenario, result in relatively early payments accruing to SANRA, regardless of the equity IRR being achieved.
The mechanism was specifically designed to offer a real opportunity for SANRA to benefit from increases in traffic levels and to act as a ?knock-out? blow if the competing bids were more or less equal in all other respects.
Risk transfer
In each successive road project, SANRA has been trying to improve its position by getting the private sector to accept an increasing level of risk, more in line with international practice.
Compensation on termination
This had become a difficult issue with the banks, as the Maputo Corridor set the precedent for the lenders to receive full repayment of their debt, even if the default was caused by the Concession Company.
In this project, SANRA was seeking to implement a regime more in line with the UK Treasury Taskforce Guidelines, with lenders being exposed to a significant level of traffic risk. To enable lenders to get more comfortable with this risk, BPCC expanded on the detailed calculations and methodology involved, with a view to making these as objective as possible.
Competing Roads
The detailed traffic model also enabled BPCC to test the impact of potential new roads in the vicinity of the concession. This process enabled BPCC and its lenders to identify and focus on the real threats to its revenue stream and propose an aggressive competing roads regime.
Upgrade works
The fragmented structure of the traffic flows over various stretches of the road meant that BPCC could be required to effect an expensive upgrade because of localised increased traffic flows, which may not necessarily generate sufficient additional revenue to pay for the upgrade.
Again the traffic and financial model enabled this risk to be fully evaluated, so that the sponsors and lenders were able to accept a regime whereby BPCC would be required to undertake all required upgrade works to maintain the required level of service on the road.
Conclusion
Ultimately we will probably never know how important individual elements of the funding strategy were in compiling the winning bid, but we believe that SANRA had a difficult task evaluating two very attractive bids and that BPCC's success was a result of extracting maximum value from all aspects of the tender.
Financial Close was achieved in August 2001, an important milestone in the development of PPP projects in Africa and sets a new benchmark for limited recourse project finance lending in South Africa:
? Total project costs of nearly ZAR3.5 billion (c. US$ 450 million) makes it the largest project financing to date in South Africa;
? Ground breaking 20 year debt tenors secured for all debt funding;
? Innovative underwriting mechanism developed to facilitate the raising of inflation linked debt;
? First project in which the European Investment Bank (EIB) is lending ZAR denominated debt following their establishment of a ZAR treasury;
? Substantial empowerment equity participation secured though put/call option arrangements;
? Almost ZAR 400 million of third party equity secured for the project, representing 60% of the total equity requirement;
? 50% of the equity for the project raised internationally including equity funding by a Spanish Development Agency; and
? Introducing tolling in an urban commuter environment for the first time in South Africa and pioneering Electronic Toll Collection (ETC) to handle the high traffic volumes during peak hours, which is a first in Africa. n
Request a Demo
Interested in IJGlobal? Request a demo to discuss a trial with a member of our team. Talk to the team to explore the value of our asset and transaction databases, our market-leading news, league tables and much more.