Private participation in infrastructure Bratislava June 2016 Dejan

Private participation in infrastructure Bratislava, June 2016 Dejan Makovšek, Economist/OECD-ITF

There are plenty of reasons to consider PSPI The main reasons come from the mitigation of the failures of traditional (public) model: • prevention of time inconsistent behaviour of governments (budgeting/hidden expropriation) Þ There is no national balance sheets, no current cost accounting => “saving” on infrastructure maintenance is never immediately visible Þ To counter the above infrastructure managers pursue cash-flow protection strategies (excess employment, union power…), which hurts operational efficiency • Limited incentives for efficient delivery of infrastructure Þ Public project managers need to follow multiple objectives (union pressures; tendency to cut as many ribbons as possible, also leads to ex-post scope changes and cost overruns) Þ difficult to ensure accountability (multi-year construction - in a couple years a project will be somebody else's problem) In summary – these are public governance challenges!

… and there are many forms of it Characteristics Service contract (outsourcing) Forms of PSP Managemen Lease/ BOT and t contract Affermage variants Scope (discrete piece or network) Discrete Normally Discrete new Existing network existing assets discrete existing assets or networks and point and network existing assets (e. g. port refurbishment normally infrastructure (e. g. terminal) and existing point sea/ airports) networks (e. g. infrastructure water) (e. g. sea/ airports) Contract duration 1 -3 years Money at risk ex ante Provider of service or management Divestitures (privatisation) What PPPs encompass Commercial risk None for the private party Concession 2 -5 years None No Private 10 -20 years 25 -30 years Yes Both options (yes or no) No Yes Private Perpetual/ subject to licence Yes Private

4 The two main vehicles of PSPI Bidder 2 Bidder 1 Monitoring Economic regulator Bidder 3 Carrot Infrastructure manager [regulatory contract] PPP contract Duty to finance Budgeting, Time inconsistency… Stick

The PPP (Project Finance) Model The ultimate cost of service gets determined through competition! Owner (developer) SPV C L I E N T Availabil. payment /user charge Equity investor Equity Construction risk Debt Service (Potential) incentives of common ownership Construction contractor Lender-comfort measures Lenders

The RAB Model (I) • The “RAB” by itself is a set of principles, which require that the assets in the RAB are maintained in line with financial capital maintenance. • Normally used for existing assets, but can be extended to deliver new assets as well. • It’s transparency and approach prevent hidden expropriation. For the investor the regulated revenue must recover: – depreciation (investments to preserve the value of the RAB), – new investments into new infrastructure (expansion of the RAB), – operating costs, – financing cost (the cost of equity and debt, which involve an appropriate reward or return). • The recovery of operating cost (and thus also the returns) is put at risk – subject to efficiency targets set by the regulator.

The RAB Model (II) The cost of service gets determined by the regulator! Dedicated Source (if necessary) Economic regulator (independent agency) Budget revenue Strategic policy development Duty to fund infra manager (by tariff setting) License fee (Fund the regulator) Payment State Infrastructure Manager (of a network) Users (and/or state) Services Budgetary expenditure Regulatory framework • Incentives for efficient operation • Monitoring • Legal duty of financing activities • . . .

8 Value for Money Dimensions Theoretical expectations Vf. M dimensions Competition = 1 st best Regulation = 2 st best Traditional PPP RAB Infrastructure delivery x Infrastructure operation x Low High Medium (relative) Cost of financing

What is the comparative performance of PPPs and the RAB model There is no comprehensive comparative performance research of PPPs vs other forms of infrastructure provision and management available!

10 Value for Money Dimensions in Infrastructure Empirical evidence Productive efficiency evidence exists, but it is not clear, whether the price is fair for a given service level Vf. M dimensions Traditional PPP RAB Infrastructure delivery x ? ? Infrastructure operation x ? (relative) Cost of financing Low High Medium Vf. M to society n/a ? ?

11 PPP challenges Three prerequisites to achieve Vf. M in competition for the contract • Information on risk/returns Inefficient risk pricing at multiple levels • Competition • Credible commitment to the contract. (massive renegotiation problems in many geographies)

What are the challenges of RAB model in developed countries The operational efficiency is well established in the literature, returns/cost of financing are low/subject to regulation. The key challenge appears to be capital expenditure efficiency: • For the regulator the capex efficiency is much more difficult to determine and takes much more time to unfold (whether it was efficient or not + narrative/benchmarking issues)… • The biggest challenge to the model (assuming government commitment is credible) are the perceived incentives of private operators to inflate their regulated asset base, on which their return is calculated. • The manifestations of these incentives are known as “capex bias” (preference of capex solutions vs opex ones) and the AJW effect (e. g. “goldplatting). • But 30 years of empirical research yielded little trace of the AJW effect (Law 2014). It’s not clear why (these incentives are offset by other ones or there are not that many opportunities to prefer capex solutions over opex; measurement issues).

To summarize … The emerging picture is: PPPs Regulated privatization Returns Assets

14 Value for Money – in what circumstances Current empirical evidence Competition = 1 st best Regulation = 2 st best Sector dependant? Vf. M Traditional PPP RAB Network infrastructure (natural monopoly) x ? ? ? (? ) Node infrastructure (competitive market) x ? n/a

Caveats • The findings in this study are mainly based on research studies, conducted in developed economies. There are obvious complexities across sectors and countries at different stages of development that warrant care in their interpretation. • More empirical evidence is needed to help understand PPP outcomes better. • The capex bias does not appear to be a serious issue in the RAB model, but more research is needed, as it is not clear why. • In cases, where public governance is challenged, PPPs can bring an improvement to the traditional approach (e. g. breaking union capture, potentially reduce pillaging of tolls). • When public sentiment does not favour privatisation or a concession of the network, it may be politically more feasible to execute a project finance PPP. • A PPP can be deployed comparably fast, whereas the setup of an economic regulator and the build-up of its capacity takes more time. Indeed it is preferred that the RAB is first applied on a public entity and then later privatized/concessioned. • Both PPPs and RABs cannot in themselves resolve the issue of governments’ ability to commit or corruption issues – a regulator, that is not independent and well-resourced or a poorly executed PPP will both deliver poor Value for Money. • In networks a change in infrastructure policy will be easier accommodated by the RAB, whereas many affected PPPs will require a lot of renegotiations. • …

Thank you! dejan. makovsek@itf-oecd. org

Can private finance fill the gap? (I) • Infrastructure is funded either by all the tax payers (direct government provision, shadow tolls) or by the users (which is a select group of taxpayers) • Both options theoretically exist in traditional public provision or a PPP (in a PPP the taxpayers pay, when an availability based model is in place) • So if there is somebody who is willing to pay for the infrastructure, why would it matter, where we borrow the money? Why would changing the source of financing solve a funding problem?

Can private finance fill the gap? (II) A transport analogy to infrastructure financing models (THE STEERING/Driver) The National Transport Plan (THE FUEL) The financing mix (THE DRIVER’S WALLET) Affordability (public finance) (private finance) (THE ENGINE) The infrastructure governance model