Because many of the factors that influence the final decision on the choice of financing are most pertinent in concession-based contracts, a crucial question is "How long should the concession be?" Concession terms vary widely across the globe and across sectors; some terms are for less than 10 years, while others can run for up to 99 years.
There are three factors that should be considered in setting the concession term:
∙ Is the opportunity monopolistic or competitive in nature?
∙ Is there debt to be repaid during the concession period?
∙ Is the level of investor return an issue?
Monopolistic or competitive infrastructure
Many governments will want to retain some degree of control over monopolistic infrastructure (for many of the reasons highlighted in Chapter 1.2). They will need to consider carefully the balance between the length of the concession and the industry's regulatory regime to ensure that users are not faced with unsustainable price increases and/or deteriorating service. It might be preferable to let a series of shorter-term concessions rather than a series of long-term concessions, as happened with the United Kingdom's rail franchise. One of the major drawbacks of short concessions, however, is that they can limit the appetite and ability of the concessionaire to make significant capital investment for many of the reasons outlined below.
Repayment of debt
If the concession requires significant upfront capital investment funded by wholesale debt, then the concession length will need to strike a balance between the period over which debt is available, the period over which it is amortized, and the level of fee or user charge. Of particular concern is the presence of significant debt to be repaid when the concession period is short. In that case, the annual finance costs may create prohibitively high user charges. For example, the annual debt cost of repaying US$100 at a 6 percent interest rate over 15 years is approximately 25 percent higher than the annual repayment amount over 25 years.2
Case in Point 1: Mexican toll roads program | ||||||||
Success is not about signing the contract and arranging the finance but is about taking a robust and sustainable approach. | ||||||||
Overview | ||||||||
In the period 1989-94, the Mexican government let a series of 53 concessions for toll roads. The program more than doubled the size of the national toll road network and represented a combined total investment of US$13 billion in 1994 dollars. But the viability of the toll roads was greatly undermined as a result of miscalculations of investment costs as well as over-optimistic forecasts of operating revenues. This situation was worsened by the 1994 Mexican currency crisis, which essentially stalled the toll road program: commercial banks were left with non-performing loans estimated at US$4.5 to US$5.5 billion, concessionaires were forced to write off large portions of their investments, and toll road users were burdened with very high tolls. By 1997, the government cancelled 23 of the 53 concessions, recovering the right to operate, maintain, and exploit these roads while absorbing US$7.3 billion in bank loans and short-term borrowings. | ||||||||
Building upon these lessons, the Mexican government launched three new programs in 2003 that have resulted in an increase in private investment in road projects. | ||||||||
The table below provides a high-level summary of some of the issues encountered in the earlier program and how they have been addressed in the current program. | ||||||||
1989-94 program approach | ||||||||
2003 program revised approach | ||||||||
|
Mexico's 1990s toll road program, in which concessions were awarded to the bidder proposing the shortest concession period highlights this problem. The short concessions led to very high tolls, resulting in traffic well below forecast. Ultimately, the majority of the concessions reverted to public ownership (see Case in Point 1: Mexican toll roads program, which provides more detailed commentary, including the lessons learned that were reflected in the more recent 2003 program).
Case in Point 2: Chilean private-public partnership road |
The Chilean public-private partnership (PPP) roads program was established in order to modernize the country's road infrastructure to meet the needs of a growing economy. The program invited the participation of the private sector in the construction, maintenance, operation, and financing of these roads. There were three main aims: 1. to use private-sector expertise to develop and finance public works, 2. to externalize the construction and operation of the facilities, improving the level of service and security, and 3. to free public resources to focus on projects and programs with higher social priorities. Between the early 1990s and early 2000s, Chile awarded, on a competitive basis, 21 real toll road concessions worth an estimated US$5 billion. Bidding started with smaller projects in order to test the market and reduce risk to the private sector. The bidding attracted 27 consortia from more than 40 Chilean and foreign companies from 10 countries, with financing arranged through both the domestic and international bank and bond markets and supported by exchange rate reserves. Prior to launching the program, the government established a dedicated agency to manage the procurement. They also enacted specific and detailed legislation relating to concessions and put in place a transparent procurement process. By starting with a number of pilot projects, the government was able to refine both its bidding process-in particular, its bid evaluation criteria-and key contract terms. Some of the most notable changes to the contract terms tried to address some of the issues relating to predictable and realistic forecasting of traffic. Different approaches included the government putting a cap and floor on the level of toll that could be bid; a variable concession term that adjusts to ensure investors the return they bid; and the government providing minimum revenue guarantees. Overall, the government wanted to award concessions that could deliver long-term financial stability and balance the toll level against the traffic volumes. The PPP program was transparent and competitive, and is generally considered a success story. |
Level of investor return
A concession period that is long enough for investors to achieve their bid return must, at the same time, not be so long the investors can make windfall gains. This can expose the public authority to the criticism that they "gave away the concession too cheaply."
One of the challenges to achieving a target or "acceptable" investor return is that if the forecasts are based on a number of variables, including the level of user demand or financing costs-then it is difficult to know the long-term outcome. This might lead to a shorter concession period over which forecasting might be more certain. A longer concession can mean a significant risk transfer over a long period of time-for example, if there is demand risk for more than 50 years. Here it could be argued that if the private party actually makes higher profits than forecast, that is still acceptable because they were also willing to accept the risk of no profit at all, or possibly even capital loss.
There are a number of contractual ways around this conundrum. For example, the Chilean roads concession has demand risk, but the period of the concession can be flexed so it terminates once the investors have reached their target return (see Case in Point 2: Chilean private-public partnership roads program). To take another example, the United Kingdom's Dartford River Crossing reverted to government ownership once the
capital cost was repaid and investors reached their target return.
Very long term concessions also introduce the issue of how the market values very long-term returns. When thinking about how NPV is calculated (see Appendix A.2), it may be appropriate to consider the timing of different investment cash flows and to adjust the discount rate to reflect the changing risk profile over the current, medium, and long term. The greater difficulty in forecasting revenue/costs should also be considered in this calculation.
"If governments lead and set understandable - Michael Till, Partner and Co-Head, Infrastructure, Actis |