This year banks have been asked to look at financings for gold deposits in several African countries. The hurdles, however, are numerous. Whilst recent political instability has affected previously ?solid' states such as Zimbabwe and Kenya, there are still a number of governments that sponsors describe as showing gratifying enthusiasm.

The ability to operate to the comfort of lenders is still dependent on the goodwill of the relevant governments. One sponsor outlines the shared experience: ?The South American situation is similar to that of Africa. You need to entrench your ability to use offshore accounting and to repatriate profits, and you'd like to get government assurances. The political environment is still the prime determinant ? no reputable commercial bank would lend to the more unstable countries in the region?.

The solution, over the past two years, has been to extend and expand the structures used in Latin American and African deals, and even draw in private insurers when possible.

Mali, for instance, has become one of the more desirable places for mining companies since the discovery of large-scale deposits during the last decade. The first mining concern to move in the country was South Africa-based Randgold Resources. Randgold's first venture was the Syama project, financed with $35 million in loans, in exchange for an equity stake, from the International Finance Corporation (IFC). The mine, partly a brownfield concern, was not as profitable as had been anticipated, and had led to profit warnings on the part of the sponsor at the end of last year.

For its second excursion into Mali, however, Randgold has been cannier. The Morilla project has avoided multilateral participation altogether, thus avoiding diluting Randgold's share. The mine is a good prospect, with reserves estimated at 104 tonnes gold, and costs are kept low through the use of open-cast mining techniques. It also benefits from a long term power purchase agreement (PPA) with a Rolls-Royce Ventures project, which will be financed as a separate project. The $330 million project has been financed by a $90 million loan from Standard Bank and NM Rothschild. The loan has a tenor of 5 years and is therefore relatively attractive to the limited number of banks interested in regional mining credits.

However, Randgold has lined up cover from AIG, a clear sign that with an established presence, augmented by backing from other major groups, the private insurance market will back African mining projects. Gold mining behemoth Anglogold, for instance, has also developed the Sadiola mine in Mali, a joint venture with the IFC and IAMGOLD, now in its second year of operation, and is also developing the Yatela project there. With AIG cover the process of syndicating the debt was relatively simple.

Nevertheless, the tenors offered by the insurers tend to be far more suitable for the operation of smaller gold mines, coming in with a deposit life well below those of base metals, at around five to seven years.

Some sort of political risk insurance (PRI) is still essential everywhere in Africa (apart, perhaps, from Ghana), although tenors are still creeping up slowly. Gerard Holden, global head of mining and metals at Barclays Capital says: ?Comprehensive PRI cover is available from both multilaterals and private insurers and we are seeing increased appetite for African risk across the board. Spreads for countries such as Tanzania have definitely tightened over the last 12 months. The private market is now offering tenor and pricing previously only available from the multilaterals.?

The insurers are also bullish. William Mulligan, who advises AIG's underwriters on global mining projects, is enthusiastic about both Africa's potential and what AIG can offer ? up to 15 years and $150 million per project, albeit with the caveat that coinsurance with an ECA is desirable. He adds that project fundamentals are the insurers greatest area of concern. Even ?stable? governments are likely to interfere with a weak project.

Much depends on the (often chequered) ownership history of the mines in question. Randgold bought out Syama from BHP as a going concern, with many of the high contracted costs already in place. It has now sold a 50% stake in Morilla to Anglogold largely, according to finance director David Ashwood, to beef up its slightly thin balance sheet. ?Morilla is attractive to the big players ? you've got 6 million ounces of gold so there's a huge amount of potential. And we're a relatively small company trying to get on its feet?, he says.

The big change in ownership this year, however, has been sparked by Ghanaian producer Ashanti's difficulties. Since its flotation in 1994 it has been pursuing an aggressive strategy of on-balance sheet financings, largely using its hedge book as a generator. Whilst project level hedging arrangements are usually uncomplicated (in the case of Morilla the loan arrangers act as counterparties), the more sophisticated instruments are prevalent at a corporate level.

Gold prices, however, are notoriously manipulable, whether from within the industry or without. Recent central bank gold sales were the stimulus for a run on gold before a framework for sales between government banks was agreed. More recently, amid much fanfare, a number of gold companies, including Placer Dome, announced a winding down of their hedge books. The move was met with some cynicism, one banker describing it as, ?smoke and mirrors. They still had positions open and were trying to get the gold price down. It worked ? at least in the short term?.

Ashanti was caught in the earlier price spike, becoming liable for margin calls from a number of its hedge counterparties. The subsequent drain on cash, exacerbated by the strike at Obuasi earlier in the year, resulted in Ashanti being unable to fund the ongoing development of its Geita project from its own resources. A US$100 million bridge facility to inject liquidity was arranged by Barclays Capital and provided by a syndicate sourced from Ashanti's existing lenders.

Conditions of the facility included that the company must raise a minimum of US$150 million in equity or auction approximately 50% of Geita to reduce leverage. The closed auction was won by AngloGold with a bid of US$335 million comprising US$205 million in cash and a US$130 million project financing of Geita. Proceeds from the sale will be used to repay the US$100 million bridge facility and retire a significant proportion of Ashanti's corporate revolving credit facility.

Tanzania, in particular the Lake Victoria gold belt, has been the second, more significant, prospect on the continent. To date there have been three project financings, including the Golden Pride, sponsored by Resolute and Samax and bought out by Ashanti. It has also been the scene of the first major deal to close this year, the Bulyanhulu project.

Bulyanhulu is being sponsored by Barrick Gold, the producer that has risen from humble origins as a Nevada producer to become one of the lowest cost producers in the world. The Tanzanian project is the first project financing that Barrick has carried out since Goldstrike, 15 years earlier. Opinions are divided over the reasons for this move ? explanations range from a desire to protect its corporate rating, to the mitigation of country risk to the development of a project structure than can be applied to its other properties in Africa.

The $200 million facility has been arranged by Citibank, CIBC, Barclays Capital, Deutsche Bank, Dresdner Kleinwort Benson and SG. Barrick bought out the mine from Sutton Resources for $281 million, elbowing out of the way a tentative facility agreed with Standard Bank London. Whilst a straightforward term loan, it follows many of the intricate structures common to emerging markets deals.

The financing is fully covered by PRI from the Export Development Corporation and the Multilateral Investment Guarantee Agency. Ultimately, a nine-year term for Tanzania was too much for the private market to handle, even with expanded appetite and a rapidly improving risk profile for the country. The two agencies are splitting the cover equally, although given the size of the deal the more complicated umbrella insurance structures have not been required.

Bulyanhulu has benefited from the benign attitude of the Tanzanian government, described by those dealing with it as ?relatively supportive?. It has gone out of the way to attract foreign investors to the country, and recently put on an EU-sponsored appraisal visit to around 15 properties in the country. For a state where large-scale privately financed mining only began in 1998, Tanzania has been extremely successful in attracting big-name sponsors.

The most unusual feature of the Bulyanhulu deal is that hedging is not taking place at a project level, but through gold price support from Barrick to project company Kahama. This measured solution, whilst exposing the parent to price volatility, is a hybrid of the increasingly sophisticated corporate instruments (which many players are reassessing in the wake of the Ashanti crisis) and the project arrangements that can restrict cashflow upstream. The facility is described by sources close to the deal as ?fairly highly leveraged and aggressively priced?.

Further details of the structure include a pre-completion guarantee, coming directly from Barrick, as well as a comprehensive security package to bolster bank comfort. The accounting structure, as with Morilla, is held offshore to mitigate currency risk as far as possible. The project also benefits from a long term PPA with Tenesco, the state-owned electricity provider. Tanzania's infrastructure is robust enough to deal not only with the power needs of the mine (low in any case compared with base metal concerns) but also the trucking of the semi-processed ore to the coast.

The indications are that Barrick will be using the structure in the future, in particular for its properties in South America, which might explain bank willingness to do a deal with Barrick on these terms. Market observers agree that the deal is likely to set a trend in both African and Latin American deals. Bulyanhulu is a property low down on the cost curve with production costs estimated at around $120/oz.

Don Newport, from Standard Bank London's Mining Group, adds, ?these deals are usually financed using reasonably conservative price assumptions. Hedging is typically going to happen at some level?. He believes that sponsors have taken the Ashanti experience on board: ?The lessons have been learned, even though the Ashanti difficulties were unique and unusual. It wasn't so much that they were overhedged but it was the type of programmes that they used?.

Mulligan at AIG says that since the crisis he has had to impress upon underwriters the importance of going through a developer's hedge book with a fine toothed comb. For those extending cover the simpler the arrangements the better, especially if the developer comes from outside the charmed circle of three or so mining giants with which financiers are familiar.

Miners and banks are now hoping for a period of price stability or a modest increase in gold prices, if only to unlock some of the less solid projects that are still stalled. The increased sense of confidence, however, is palpable. The Democratic Republic of Congo is still off-limits, at least for the time being, but there are few countries that the players won't examine. But the lessons learnt from Latin deals during the lean years are now paying off. ?Africa used to be a land of juniors, venturers and Tiny Rowlands. Now you've got all of the big players like Phelps Dodge, RTZ and Anglo coming in and spending big money on exploration,? says one seasoned industry observer.

It appears that a happy medium of mitigation of many external risks coupled with strong sponsor support can now be served up to the banks. Gerard Holden at Barclays Capital reflects the mood: ?Providing the structures are robust and the risks effectively allocated between the sponsors, other project parties and providers of finance, there is good appetite for mining projects in Africa. The increased activity of the majors in Africa means the edge of the envelope is being pushed back all the time. Two years ago, five years was the maximum term that could be syndicated for Tanzania. In the last month the Bulyanhulu financing has shown that today the bank market has appetite for nine year risk.?