Here's some good news for Montek Singh Ahluwalia, who has been struggling to push through PPPs in infrastructure in the teeth of Left and other opposition. This paper* says that "from a public finance perspective, PPPs remain public projects, and should be treated as such." It also gives a basic rule for devising the best PPP contracts from all points of view.
The authors, Eduardo Engel, Ronald Fischer and Alexander Galetovic were struck by the sudden and huge popularity of PPPs. They say "articles in the Financial Times mentioning this concept increased twenty-fold, from 50 in 1995 to 1,153 in 2004."
So they decided to find out why this had happened, and also what an optimal risk-sharing contract between the government and the concessionaire, in the face of large risk - it is called Prakash Karat in India - should contain.
Their answer for the popularity of PPPs is that it allows governments to pretend that things have improved at virtually no cost to the exchequer. But if you ask me, it is based on a very old and very simple idea, that of a management contract.
All empires, when they grow too big and unwieldy, need managers for the estates. Feudalism was based on this principle. The mansabdari system was, in the end, a PPP.
It is worth reminding Indian readers here that the idea of management contracts (albeit to employees, not concessionaries - public-public partnerships, if you will) was first revived by Arjun Sengupta way back in 1982.
He had proposed those famous MoUs between the public sector and the government because during the 1970s the public sector empire had grown so large and so suddenly that it wasn't delivering the goods.
Anyhow, where the optimum contract is concerned, this is what the authors say, " the contract that optimally trades (off) demand risk, user-fee distortions, and the opportunity cost of public funds, is characterised by a minimum revenue guarantee and a cap on the firm's revenues."
So the deal is this: we will ensure that you don't lose money, if you promise not to use your temporary monopoly to rip-off the users. In feudalism this was mansabdar autonomy in return for the Emperor's protection.
The difference between this and wholly owned government infrastructure is obvious. The contract contains both positive and negative incentives. In contrast, government-owned-and-operated infrastructure contains nothing, zilch, zero, as it operates only on a vague hope that things will work.
This has happened in all countries where PPPs have come into being. The authors say, "PPPs cannot be justified because they free public funds." That is plain silly.
However, if you get the contract just right, "PPPs are closer to public provision than to privatisation." This is what the Planning Commission needs to tell the comrades.
The argument is this: the transfer of ownership is temporary and reversible. Of course, those who oppose PPPs may argue that this really doesn't matter because even the best contracts hand over all user-fees to the concessionaire for the duration of the contract. As a result, the state gets nothing from its assets.
Nevertheless, say the authors, "what matters is the intertemporal risk profile of cash flows, not the year-to-year risk profile." In other words, risk-forecasting is critical and the trade-off that the government has to determine is the "extent to which subsidies are a more costly source of financing than user-fees."
If they are very costly, the eventual contract will resemble full privatisation because the subsidy is zero or almost zero. But if a subsidy is not all that expensive, a minimum income guarantee along with a cap on user fee revenues, works very well.
Intuitively, I would say that demand-risk in infrastructure in high growth developing economies is front-loaded (like in China) and, therefore, contracts should ensure that subsidies start off at a high level and taper off - non-negotiably - to zero towards the end of the concession.
But that means ensuring zero ministerial corruption. Now who will mitigate that risk?
You and me, of course.
*The Basic Public Finance of Public-Private Partnerships, NBER Working Paper Number 13284, July 2007
