05/11/2003

The Ashkelon desalination BOT project

A premiere in many respects: the first BOT project in the Israeli water sector, the largest seawater reverse osmosis desalination plant in the world, the first project to beat the ambitious target price of 0.50 US$/m3 for desalinated seawater, the first limited recourse financing package arranged for a water project in Israel. The main innovation of the Project lies, however, in the structuring of a mixed bank/institutional local financing. A remarkable achievement that paves the way for upcoming transactions. By Pierre Coindreau and Christian Rasoamanana, PricewaterhouseCoopers.

The Ashkelon Build, Operate and Transfer Project ("the Project") involves the financing, design, construction, operation, maintenance and transfer of two sea-water desalination plants with a total production capacity of 100 million m3 per year. The Project will use the Reverse Osmosis technology. The plant is due to start partial operations in February 2005 with full operations scheduled for October 2005. The Project was awarded to VID Desalination Company Ltd ("VID"), a special purpose company sponsored by Véolia Water (formerly Vivendi Water) and its two Israeli partners, IDE Technologies Ltd and Elran (D.D) Infrastructures Ltd. (Elran Group – a Dankner Co-operation).

The total Project Costs is approximately NIS 1 billion (USD 250 million). The whole project debt was raised locally and is NIS-denominated. The Ashkelon deal has set a precedent on the Israeli financial markets as the first Israeli deal to be financed by way of bank debt and institutional private placement. The financing was arranged by Bank Leumi Le-Israel B.M. who also provided the bank debt. The institutional bond issue which represents 60% of the total financing was managed by Gmul Sahar Underwriters Ltd. and Leumi & Co. Underwriters Ltd. The institutional financing process is particularly noteworthy since it involved more than 60 institutional investors.

PwC acted as financial advisor to VID from bid preparation through to Financial Close. Goren Capital Group, a Tel-Aviv based investment bank, coordinated the institutional financing.

The Ashkelon desalination project is the first of a series of coastal seawater desalination projects launched or to be launched by the Israeli Government to reach financial close.

A project of strategic importance for Israel

Since 1998, Israel has been experiencing a water crisis, the most severe one since the State of Israel exists. The crisis is first a supply crisis. The combination of climatic factors (low precipitation level, prolonged droughts) and the lack of management of water reserves led to the drying up of the existing resources (mainly Sea of Galilee-Lake Kinneret and coastal and mountain aquifers) which reached their lowest level on record. There is a rising concern that the over-exploitation of these resources and their pollution due to insufficient treatment of waste might lead to an environmentally hazardous situation and more particularly to an increase in the salinity rate.

On the demand side, the rising trend in the water consumption – the household sector consumption almost doubled in 15 years – is expected to continue as a result of the increase in both population (expected population of 7.5 million in 2010 compared to 6.4 million in 2000) and standard of living. In addition, water supplies that Israel undertook to supply to Jordan in the peace agreement with the latter, are to increase gradually in the future. The Water Commission's master plan foresees in 2010 a total demand for water amounting to 2,288 million m3 per year (compared to 1,930 million m3 in 2000), the Ministry of National Infrastructures expecting 2,400 million m3 by 2020.

In order to bridge the widening gap between a growing demand and lacking water resources without continuing excessive pumping, several measures are considered/implemented: (i)implementation of a water preservation program (mainly reduction of water usage in agriculture), (ii) increased use of recycled sewage water, floodwater and saltwater for irrigation, (iii) water imports by ship or pipeline, but subject to political considerations, (iii) desalination of brackish water in coastal aquifers and their subsequent replenishment, (iv) regulation policy of the water consumption by way of price increases so that tariffs reflect the actual production costs, (v) accelerated implementation of the desalination plant construction program (one 50 Mm3 capacity plant needed every 2 years according to the Water Commission). The Government decided in April 2002 to double the initial 200 million m3/year desalination plant construction program that is expected to be full operational in 2005.

The Ashkelon desalination facility is in this context a major step towards a solution to the water crisis in Israel. Located in the South of Israel, the Project is aimed at supplying large southern Israeli cities.

Development history

Diagram 1: project milestones


In July 2000, the Ministry of National Infrastructures and the Ministry of Finance, acting through the Tender Committee, launched the tender for the financing, design, construction, operation and transfer of a seawater desalination facility with guaranteed production capacity of 4,170,000 m3/month and 50 million m3/year. The Pproject is governed by ruled out under the terms of a Build, Operate and Transfer Agreement ("BOT Agreement") to be entered into with a government agency, the Water and Desalination Authority (“WDA”) whose obligations (including payment obligations) under the Agreement are deemed to be obligations of the State of Israel. The BOT Agreement is awarded for a period of 24 years and 11 months. The production of the facilityplant will be sold to the WDA on a "take or pay" basis. As a Government Agency, the payment obligations of the WDA are backed by the State of Israel.

The proposed price of desalinated water was the key selection criteria (70% of the score vs 15% for the technical offer, 10% for the financial proposal and 5% for general impression). The price comprised fixed and variable elements. The energy costs made up a large component of the variable costs. In this respect, the bidders were offered two alternatives: buy electricity from IEC, the national power utility or build their own IPP. The choice of the technology also heavily influenced the cost of water. The bidders had the option to formulate their bid on distillation technology, reverse osmosis or hybrid design process. It is interesting to note that all bidders have elected the RO technology.

The bid submission took place on 1May 2001, with a 2-month delay compared to the initial schedule. Following review of the Technical and Financial Proposals contained in the Bids and the clarification process, the Tender Committee declared all Bids non compliant and decided to permit all Bidders to amend their bids (including to revise their price proposals) for 15 August 2001. This possibility to submit downward revisions to price proposals at this stage of the tender process was viewed as a first round of price negotiations as the tender documents provided for such a revision only later on during the Best and Final Offer. On 3 September 2001, the Tender Committee declared all bidders qualified for the BAFO stage during which they were allowed to change only their price proposal. Shortly after the BAFO submission, VID Desalination Company Ltd. ("VID"), the special purpose company established by Vivendi Water, IDE Technologies Ltd and Dankner Ellern Infrastructures Ltd was designated Successful Bidder, over the American Ionics and the Spanish Cadagua. With a price of 52.7 UScents/m3, VID has set a precedent on the desalination market (the previous lowest price submitted for desalinated water was 65 UScents/m3). The signing of the BOT Agreement took place in November 2001. Financial close was expected to take place within a maximum of 6 months from the award of the 50 million m3 p.a. project, i.e. by end of February 2002.

The BOT Agreement stipulated that the Consortium could be requested to double the plant production capacity from 50 to 100 Mm3 a year at the request of WDA. The decision of the State to exercise this option so rapidly - negotiation on the additional 50 Mm3 started in the course of December 2001, i.e. one month only after the execution of the first BOT Agreement, and the capacity extension was awarded on 28 January 2002 – came as a surprise even if the doubling of the capacity was likely given the severe water shortage prevailing in Israel. The very attractive water price offered by VID was certainly a key factor in the go-ahead decision while casting doubts on the economical viability of other solutions to the water crisis considered by the State such as the import of water from Turkey. The capacity extension enabled VID to lower the desalinated water price below the 0.50 US$/m3 mark, a first in the desalination history. The additional agreements were executed in April 2002.

If the doubling of the capacity has made the Ashkelon project the largest seawater reverse osmosis desalination plant in the world, it also strongly impacted on the financing of the project. The nearly doubling of the financing resulted in a capacity issue as Bank Leumi was no more prepared to underwrite the full amount of the facility and requested to syndicate as much as 50% of the total financing. Two alternatives were then available to VID: syndicating the facility to banks and/or to Pension Funds. The situation of the banking market in Israel at that time (and still today) - most Israeli Banks were constrained in their lending capacity due to problems with capital adequacy ratios - combined with the reluctance of foreign banks to participate in long term financing because of the security situation, made extremely difficult a traditional syndication to banks. The only eligible bank was Bank Hapoalim, one of the two Israeli largest banks with Bank Leumi, but its "Related Party"[1] status with one of the member of the Consortium prevented it from participating in the VID's financing. The only possible alternative was therefore the "Pension Funds" route. The participation of the institutional investors necessitated a rating of AA- minimum from Maalot, the local rating agency. The discussions with Maalot started from end of April 2002. The rating process took some 3 months and in August 2002, Maalot announced its decision to grant a AA- rating for the project debt subject to the meeting of some requirements including the arranging of additional - but limited - guarantees and commitments (in particular, Maalot did not require completion guarantees).

In parallel to the discussions with Maalot, a road show was organised to present the project to most key large and mid-size institutions in Israel, including insurance companies, pension funds and provident funds. In order to help to raise institutional financing, VID appointed a Tel Aviv investment boutique, Goren Capital Group with a good experience in dealing in the Israeli institutions market. A tender for the institutions was launched at the end of October 2002. The institutions were offered two disbursement alternatives: drawdown or one-time. In the drawdown option, the money raised ("the Drawdown Institutional Tranche") is disbursed after Financial Close along with the bank debt. The money raised under the one-time option ("the Upfront Institutional Tranche") is fully disbursed at the time of the institutional tender, deposited in an interest bearing trustee account and released upon Financial Close. The interest rates for both the Drawdown and Upfront tranches were fixed at the tender date. The strategy to favour as much as possible the drawdown option was a success as the Drawdown Tranche made up more than 92% of the total institutional financing. It was important to complete the tender rapidly as some other important transactions were expected in the market. However, raising funds before certainty of Financial Close - successful syndication of the financing was a condition precedent to financial close but not the only one - created several risks. One of the risks was the legal risk of material changes to key documents by either Bank Leumi or WDA who had to approve the final financing documents before financial close. Indeed, the institutions gave their commitment based on draft documents and a fundamental worsening in the loan documentation between the institutional tender date and financial close could have caused some institutions to pull out of the deal. The greater risks lied, however, in delaying the Financial Close beyond 90 days from the date of the institutional tender as after this date, the institutions were released from their commitment. The Facility Agreement was eventually signed on 23 January 2003.

In light of this complex credit story, the fact that the project took almost two and half years to complete is hardly surprising.

Contractual structure

The diagram 2 shows a typical BOT project structure with VID passing risks, through the contractual structure, down to the subcontracting parties.

Diagram 2: contractual structure

From a legal point of view, the extension has been dealt with by a new contract executed between VID and WDA ("the Second BOT Agreement"), while the initial agreement ("First Agreement ") remained valid. A “Supplement Agreement” made the bridge between the two Agreements. This Supplement Agreement provided, inter alia, for the unification of the project schedule (e.g. one Financial Close for the extended Project), the water quantities requested, the water price, the procedures for force majeure, termination and buyout.

One major change in the contractual structure of the extended project compared to the initial 50 million m3/year project was the power supply structure. The power island moved from a smaller open cycle gas turbine (entirely dedicated to the desalination plant) to a larger combined cycle unit with excess power to be sold to third parties allowing a lower energy price to the plant.

The financial part of the contractual structure is relatively straightforward as there is only one financing agreement executed with Bank Leumi and the pension funds - the expectation was that VID would not avoid signing two separate financing agreements with one inter-creditor agreement between Bank Leumi and the pension funds -. This reflects the fact that eventually, the addition of the pension funds to the financial scheme brought less complexity than expected, the institutional tranches being in many respects similar to the bank tranche.

Raising finance: the challenges

Syndication issue: the successful syndication of at least 50% of the total financing became the biggest challenge for VID and its advisers.

The syndication issue has been exacerbated by:

  • the difficult situation of the local banking market that faced increased liquidity constraint: most Israeli banks were closed to the minimum capital adequacy ratio required by the Supervisor of Banks and hence were limited in the amount of funding they could make available;
  • the economic and political environment: the global economic slowdown and particularly the crisis in the high tech industry combined with the security situation harmed the Israeli economy, which contributed to the low performance of banks and further restricted their potential for providing new credit;
  • the reluctance of international lenders (including those already operating in Israel) to participate in the financing of long-term projects in the political context prevailing in Israel. In addition, participation of foreign lenders would have considerably complicated the financing and was never considered as a realistic option;
  • the limited number of banks: only two major banks (Bank Hapoalim and Bank Leumi) can play a significant role but they often tend to avoid to work jointly;
  • the banking regulatory constraints which further limit access to bank financing: two regulations raised problems: (i) the "Group of Borrowers"[2] regulation that sets the credit exposure limits of a bank towards a borrower and a group of borrowers and (ii) the "Related Party" regulation that restricts the ability of a party related to a bank to raise debt from this bank (see above). In the case of VID, the Israeli members of the Consortium were affected by both rules.

Competitiveness of the financing terms: in addition to the issue of availability of sufficient lending volume, which actually limited the range of alternatives offered to VID for syndicating the financing, it was important that the potential finance providers match, to the extent possible, Bank Leumi's financing terms that have been used de facto by VID in its offer to the State of Israel, mainly regarding:

  • debt maturity: 23 years at minimum. Clearly, foreign banks were unable to match such a tenor.
  • availability and grace period
  • flexibility for the drawdown: miminum amount for each drawdown, maximum number of drawdowns
  • currency: the strategy was to favour NIS-denominated funding in order to limit foreign exchange exposure as the fixed price is linked only to the local CPI.
  • repayment profile: the acceptance of tailored-made repayment was a key criteria for the selection.
  • average life
  • interest rate: Bank Leumi offered the flexibility of fixing the interest rate at each drawdown or upfront.
  • security package: one of the biggest challenge has been to convince the lenders and particularly the pension funds and Maalot, the local rating agency, to accept the construction risks without any completion guarantee, which could not be taken for granted given that the plant is the largest of its kind in the world and that there are few precedents of limited recourse financing of desalination projects in the world.

Key factors of success

  • a strong project rationale: the project is of the highest national importance, particularly in light of the continuing and deteriorating crisis in Israel’s water market
  • the ultimate government nature of the risk: the project is based on a take-or-pay type agreement with the State of Israel as counterparty
  • the quality of the sponsors, with a right mix of local and international experience
  • a well structured project with well structured contracts and appropriate allocation of risks
  • the historical credit appetite of the local debt market for infrastructure projects in Israel offering very long term NIS denominated CPI linked loans (up to 28 years)
  • the large volume of institutionals credit: Israel benefits from a remarkably high level of institutional financing reinforced by recent amendments to regulations permitting an increased participation of insurance companies in AA rated project bonds
  • and last but not least, the high level of determination of all parties involved “to make it happen”: although the project took a very long time to achieve financial close, its closing demonstrates that the arranging of a limited recourse finance package, even in a difficult environment and in a market with few precedents, is possible if there is sufficient will from the different parties (sponsors, lenders and government).

Conclusion