The season is open, but it is already clear that the game is changing and some of the old rules no longer apply. On the demand side, the sheer weight of potential borrowing is forcing significant new thinking about how available pools of liquidity should be best accessed, about the need for strong structures and the importance of timing a deal properly.

As the market grows, the margin for error shrinks. As the table on page 6 shows, there are dozens of projects in the pipeline that are chasing up to $24,000 million worth of private-sector finance. While it is unlikely that all these deals will be completed before the end of next year, only 50 per cent will have to be signed for the bumper year of 2002, in which just over $12,000 million worth of finance was provided, to be exceeded.

Where is all this finance going to come from? In the past, the international banking community would have carried the main load. But recent MEED surveys have been charting the rapid evolution of the market. Back in 2000, international houses provided 64 per cent of the underwriting on GCC deals. This figure fell to 58 per cent in 2001 and then dropped to 39 per cent last year. And the trend has continued on the deals done so far this year.

There are two forces at play. First, the appetite of some regional international banks has declined. 'The mergers in Japan and Europe have forced some natural exits from Mid East deals,' says one international banker. 'And - particularly among US banks - others have very flexible mentalities and fast-changing views on country risk: they can move in and out very quickly. Despite all the talk you hear about relationship building, for credit committees the money does the talking.' And over the past 12 months or so, there have been more international banks leaving the region than arriving.

To date, however, the slack has been more than taken up by the growing number of regional and local banks pushing their way into transactions. It has been a three-step process. Local heavyweights such as National Bank of Kuwait (NBK), National Bank of Abu Dhabi (NBAD) and Qatar National Bank (QNB) have moved from being lowly participants in domestic deals to essential lead arrangers and, increasingly, to being aspiring lead arrangers on cross-border transactions. The act of raising themselves up the value chain has coincided with a number of new arrivals further down. Institutions such as Ahli United Bank, National Bank of Dubai, First Gulf Bank and BankMuscat are finding places in the lead arranging groups of domestic deals.

The surge in regional liquidity since the last oil price crash in 1997-98 has provided the fuel for this supply-side transformation of the regional project finance market. It is worth noting that money supply indicators have grown every year in every GCC market since 1997 (for detailed analysis of regional liquidity see the Economic Briefing in next week's MEED). But high regional liquidity is deceptive, and it will not indefinitely support the indigenous provision of project finance. Growing concern over asset:liability tenor mismatches will provide a brake on the flow of finance.

The problem is simple. The structure of most regional banks' balance sheets hampers the extension of long-tenor loans as the liabilities side is predominantly built with short-term deposits. With regional bond markets in their infancy, access to long-tenor finance is limited. Of course, the timeless nature of equity provides a natural cushion for some long-tenor lending but it is not limitless.

The statistics make interesting reading. The top 15 regional and local lead arranging banks in the GCC in 2002 had aggregate equity of almost exactly $20,000 million at the end of last year. Given the fact that sizeable proportions of short-term deposits are, in reality, rolled over almost indefinitely, there is no direct correlation between equity bases and capacity for long-tenor lending, but it does provide an indicator. And the signs are that some banks are bumping up close to their ceilings. For example, Arab Banking Corporation (ABC), long an innovator and market leader in regional deals, has been noticeably absent over the past year from 10-year-plus transactions. The regional banks may have been hungry for seats at the top table, but given their comparatively small size they do not have limitless appetites.

One of the main impacts has been a shift in the focus of risk mitigation. Project sponsors and their financial advisers have drifted towards the construction of increasingly large lead arranging groups partly to make room for the aggressive regional and local players and partly to reduce syndication risk (see chart opposite). 'What we have been seeing is a concentration of bank interest at the top end,' says a regional banker. 'We don't see many international 'stuffees' anymore: the internationals either come in at the MLA [mandated lead arranger] level or don't do the deal at all. This is fine, but some of the stronger regionals and locals are adopting the same strategy. The result is top-heavy deals that are effectively expanded clubs with almost all the banks interested in the deal already aboard at the MLA level. And, paradoxically, unless the MLAs are careful syndication risk grows the more you move to mitigate it.'

While there have been no out-and-out failed syndications in recent years, there is a growing feeling that deals are coming closer to the edge as regional balance sheets fill up with long-tenor assets and the portions international banks are prepared to book decline.

The market has been given another layer of complexity by the quiet but significant development of the secondary market. 'There has been increased activity in secondary trading over the past 12 months,' says an international banker. 'It has been mainly supply driven. Japanese groups such as SMBC [Sumitomo-Mitsui Banking Corporation] and MFG [Mizuho Financial Group] have been selling down after their mergers to reduce concentration risk - which was to be expected. Some US banks have been dumping portfolios, some wholesale and some have just been slicing and dicing straight after the assets have been booked. The German landesbanks have also been selling down for their own domestic reasons. The result has been a fair amount of potential appetite for primary deals being satisfied elsewhere.'

The dynamic of the secondary market has been straightforward: those institutions interested in sector allocations and yield rather than relationships have been building good books at a discount. 'The main beneficiaries have been the mid-tier regionals which have no real expectation of picking up relationship-driven ancillary business,' says a regional banker. 'They have also enjoyed the opportunity of picking up exposure at more manageable tenors. Some of the 12-year deals done four or five years ago have now got seven or eight years to run: credit committees can get a lot more comfortable with this.'

The direct impact on new project financings is difficult to measure, not least because of the opaque nature of secondary market activity. On the one hand, some highly liquid regional banks have preferred building portfolios in this secondary market to participations in primary syndications. On the other hand, it has helped international banks to free up balance sheets for new deals. 'We operate with sector and country ceilings,' says another international banker. 'The maths can be good on selling down one transaction in the secondary market to make room for a new deal. It can also help financiers get primary deals past credit committees: 'Yes, it says the tenor is 18 years, but the reality is that we are not looking to book an 18-year asset: the secondary market is liquid, we can exit when we want.''

In a similar fashion, the growing number of refinancings is also altering the primary market. And they are coming thick and fast. Over the last three years a total of four major refinancings have been completed - for Oman LNG, Equate, Qatar Fuel Additives Company (Qafac) and Saudi Petrochemical Company. The likelihood is that another three - for Qatar Vinyl Company, Taweelah A-2 and Saudi Chevron Phillips Company - will be closed over the next six months alone.

Here too, the impact is divided. Refinancings may offer international banks a route out, but they also offer a route into established deals for regional lenders that are more comfortable with corporate-focused structures than pure project credit. 'On the one hand refinancings are sucking liquidity out of the greenfield market, but they are also bringing confidence to it,' says an international banker. 'Like the secondary market, they offer an exit route which makes primary deals easier to swallow.'

They have also been easier to swallow when cut up into more manageable mouthfuls. Perhaps one of the most visible trends in recent years has been the development of the multi-tranche transaction. Rather than loading up on commercial debt alone, financial advisers and sponsors have increasingly worked on tapping a variety of liquidity pools in the same deal. The use of Islamic tranches has moved from being exotic to mainstream over the past couple of years, with the Hidd, Shuweihat and Umm al-Nar financings including deal-making Islamic components.

However, the Aluminium Bahrain (Alba) financing explored the outer limits of multi-tranche structuring: a commercial component was combined with a metals trading facility, a local bond and an Islamic tranche and a $450 million export credits portion is still to be signed. The tactic is simple: diversification reduces risk and can ease a deal through the market.

Over the next few years, new liquidity pools will need to be tapped if the full potential financing requirements are to be met. Probably the most important will be attempts to develop regional bond markets. Few expect there to be much of an immediate future for direct project bonds - institutional investors are invariably too inflexible and difficult to handle for the tastes of regional project sponsors. However, bonds might increasingly be used as components in refinancings: they lend themselves to the less rigorous structuring and more corporate-style focus that have become the norms on the refinancing of mature and established projects. But active regional bond markets will also be of profound indirect importance to regional project finance transactions. They will offer financial institutions access to long-tenor finance that will itself be made available for project lending.

Perhaps the most radical solution to managing the problems thrown up by the pending deal flow is the most simple, if sensitive: rethink pricing. 'We have a situation in which pricing patterns are beginning to look wrong and one of the reasons why some of the international banks are leaving the market is because the risk/reward ratio is not particularly attractive,' says a regional banker. 'We are going to have to see some upwards pressure on pricing over the next year - even though the structures on deals are getting better.'

It is not unusual - or surprising - to hear bankers arguing the case for wider margins and larger fees, but with a growing volume of projects competing for cash in what is no longer a virgin market, they might just have a point. What this could do to project viability remains to be seen.

Additional reporting by Clare Dunkley