PPP Project Failures: the tragedy of the evitable (III). Failing to structure

In this new article we advance further in the project cycle, still in pre-procurement phase (before project approval), to discuss how the right project (the right public investment), suitable for a PPP and even if properly prepared, can suffer dramatically and be either a complete failure (rescue/early termination) or a significant failure (re-structuring the project through re-negotiations), simply because we were unable to define the correct PPP structure (financial structure and/or risk allocation structure).

Although there are examples in other sectors and with respect to other risks, this is especially true when talking about linear transport infrastructure and mostly relevant in relation to demand and revenue risk.?

The need for traffic/volume risk sharing in transportation?

The most evident case of failing in defining the right risk structure is related to traffic risk (or “traffic and revenue”, T&R) in certain projects that are user-pays projects.

This failure is naturally connected with that one discussed in the previous article, the one related to lack of proper revenue forecasting by means of being overoptimistic.

Do we believe our projections? So bet on them, be accountable, i.e., assume a part of the risk.

In my opinion, government should at least share a relevant portion of T&R risk of any real toll project that is green field (i.e. projects that require significant initial investments and doesn’t look “super-feasible”).

In addition to the uncertainty of T&R estimates, in urban transportation the demand risk -as perceived from the private perspective- is even bigger. Many things beyond private party′s control (but in full control of the government) will influence the demand: traffic light priority, reorganization of existing bus lines, tariff levels, mobility and public transportation plans...

In many projects demand risk is too big to be efficiently managed and/or priced by the private party, and without some risk sharing (MRG or bands), the project will suffer from lack of appetite from investors, or these will act too aggressively assuming that if things go really wrong some restructuring will come in the end.?

If a government is able to finally support financially a project, this should be contemplated from the outset and built into the contact.?

However, in transportation projects based on user charges (basically, real toll projects), a considerable number of failures comes from the inappropriate lack of risk sharing mechanisms. We have seen what may happen when a much worse than expected traffic scenario shows up. Madrid Access toll roads “Radiales” and M1/M15 toll motorways in Hungary are clear examples well known and analyzed in multiple case studies.

The most typical approach is to provide a Minimum Revenue Guarantee (MRG) – like a number of road projects in Turkey (the € 1,2 billion “Eurasian tolled tunnel” under the Bosphorus, Gebze-Izimi motorway with 6,5 bn US$ capex, Third Bosphorus Bridge, etc.) but there may be other approaches than can work, such as the movable term of the concession (guaranteeing a minimum revenue in net present value terms like in Chile), or other subtler schemes to temper down the risk focused on protect bankability (participative loans).

The most typical approach is to provide a Minimum Revenue Guarantee (MRG) – like a number of road projects in Turkey (the € 1,2 billion “Eurasian tolled tunnel” under the Bosphorus, Gebze-Izimi motorway with 6,5 bn US$ capex, Third Bosphorus Bridge, etc.) but there may be other approaches than can work, such as the movable term of the concession (guaranteeing a minimum revenue in net present value terms like in Chile), or other subtler schemes to temper down the risk focused on protect bankability (participative loans).

Even in government pays projects, when it has been decided to pay based on volume or demand, it is necessary to temper down to a significant extent the risk, to make the project bankable. These is the case of some LRT PPPs in Spain where the shadow fare or “payment per user” changes depending on certain bands of demand, which may be seen as well in the early toll-free PPPs (shadow toll road PPPs).?

Some key takeaways about traffic risk

• ?In addition to proper traffic forecasting and the need for a clear rationale for the project, green field toll roads need a MRG or some other form of support for traffic downturn scenarios
• The assumption of part of the traffic risk by the authority would likely influence favourably in the quality of traffic forecast at project appraisal?

Toll roads PPP based on availability payments: it is not an oxymoron

In real toll projects, MRG may be a good approach to make the project bankable and commercially feasible. But MRG should be considered with caution. When there may be a high probability (or simple some material probability) that the MRG will be called, a better structure may be removing completely the traffic & toll revenue risk, by retaining the property and related risk of the fare box in public hands and paying on the basis of availability. ?Some successful relevant cases that deserves to me mentioned are the POMT project and i-595 in Florida (US), ?the A2 highway in Poland, or more recently the project “Rutas 2 y 7” in Paraguay.

To learn more about these type of structures see see a case study about i-595 in Florida available online at www.i595express.com

Government pays PPPs: Availability payments is superior to volume payments

When it has been decided that a road will be toll free, so to be funded with service payments then there should be no doubt: the payment mechanism should be based in availability. This provides higher bankability as cashflows are more predictable (and only depends on a fully manageable risk, operational risk), with the inherent additional benefit of lower margins in interest rate and/or higher leverage clearing clear efficiencies in terms of weighed average cost of capital.

A good example to illustrate this superiority is the R1 Motorway (Slovakia), a 984€ DBFOM too-free road under availability, successfully financed in 2009 (at the height of the financial crisis), and later on refinanced through a 1,2 billion bond in 2013 (26 years BBB+) when it was regarded the “Bond Deal of the Year” by PFI).

In addition, it provides flexibility to move the decision in the future to apply tolls and permits a natural and powerful focus on performance. If the SCUT program in Portugal (the toll free road PPP program) would have been based on availability payments from the outset, it would have not been necessary such long and costly renegotiations that occurred when Portugal had to impose a toll on the use of such highways under the directions of the EU rescue package.

Also, when a project requires public complementary payments because of the insufficiency of user- payments (typically the case of urban rail projects), there is a clear trend of using a payment mechanism based on the availability concept (i.e. availability payments), with notable cases as Zaragoza LRT, Belgium LRT PPPs program (5 LRT DBFOM PPP projects mounting 1.1 billion under availability payment scheme, procured and contracted between 2009 and 2015), or Milan Metro Line 5 amongst others.

It could be argued that being a clear public objective, in this case, maximizing the use of public transport, payments should be made on the basis of volume of use. But the patronage risk can be very difficult to accept by the investors and lenders (due to the many sources of risk that are fully out of the control of the private party), and the most effective manner in which a private partner may help to increase demand is by providing a reliable and high quality service.

In heavy interurban rail, the trend to use availability payments is also clear. As an example, after a first project (out of 4) being structured with traffic risk, the 3 other projects of the French PPP HSR program embraced successfully an availability payment approach which apparently facilitated more the smooth financial close and improvement of conditions.

Selecting inappropriate payment mechanisms in other sectors

What we have said about transport is also very applicable to sectors like water infrastructure (the revenue of a WWTP or a desal project should not be dependant on volume of water as the inlet is not controllable at all) or environmental (mostly waste management, as the tons of waste are not controllable either) but in the capacity of the plant and the quality of the output.

Some key takeaways about defining the appropriate revenue regime and selecting the right payment mechanism

• When road traffic is not a policy objective (as should not be, generally) and in order to focus the PPP on incentivising performance, availability payment is the best approach (even if there is a toll)
• The success ratio in availability is much higher than demand linked revenue schemes
• Another clear case for which availability payments is more sensible is when traffic / demand is not controlled at all by the private partner (HSR in France)
• Even when maximizing use is a strategic objective (e.g. Urban Metro and LRT), availability payments are more indicated (including a limited bonus for higher demand) as the best manner to maximize demand (as in the control of a private operator) is quality of the service (especially when tariff levels are settled by the Authority, as it is the common case)?

If you want to read about risk structuring in PPPs?5. Risk Allocation and Structuring[32] | The APMG Public-Private Partnerships Certification Program (ppp-certification.com)

If you want to read more about payment mechanisms and availability payments design?4. Financial Structuring (from the Public Perspective): Defining the Financial Structure and Payment Mechanism | The APMG Public-Private Partnerships Certification Program (ppp-certification.com)