Today’s Citizens Budget Commission panel on “How Public-Private Partnerships Can Help New York Address Its Infrastructure Needs” (the link is to the full report, with good case studies) offered an excellently reasoned and evidenced summary of what the private sector can and cannot do when it comes to public-infrastructure investment.

First, a rough definition of private-public partnerships, as generally agreed upon by the panel. It’s a structure of running an infrastructure project under which some private-sector party takes at least some of the significant risks on the whole lifecycle of the project, from design to construction to long-term operation and maintenance to (sometimes) financing. (This author doesn’t like the term “partnership,” because a realization of a tension of interests is always a healthy thing in a business agreement, but, CBC didn’t invent it.)

The most interesting statistic to come from report authors Maria Doulis and Charles Brecher was this one: although labor-friendly northeasterners tend to look upon PPPs with suspicion, of the $16 billion worth of top ten PPP projects done in the US in the past decade or so, nearly half — $7.5 billion — are in New York or New Jersey. These projects are the Hudson-Bergen Light Rail project ($1.9 billion), the JFK Airtrain ($1.8b), JFK Terminal Four ($1.4b), Cemusa’s NYC “Street Furniture” deal (the shiny new newssstand kiosks you see around, $1.4b), and the Camden-Trenton River Line ($1b). The Port Authority, in particular, has been a good experimenter here.

Other important points from the discussion:

PPPs are not a font of free money. CBC chief Carol Kellermann and panelists, including people from city and state government, were very clear about the fact that whether you’re getting money for infrastructure directly from users (i.e. tolls), or from a guaranteed revenue stream (i.e. annual availability or capacity payments from the government to the private operator for running a free road or something like a dam), the revenue sources are the same as they would be on a project backed by traditional muni bonds and run by the public sector.

The panel acknowledged that any government entity simply looking to get a few billion bucks upfront by monetizing a multi-decade stream of toll revenues, a la the Chicago Skyway deal a few years back, isn’t thinking about these things the right way. First, because it’s easy to waste the money, meaning it’s just a form of bonding out future revenues, and second, because, with the uncertainty of discount rates over a long period of time, there’s no real assurance that the government is getting the better end of the deal.

Another way of looking at it is: when i-bankers run around saying that there is a $100-200 billion pot of money for this stuff, it’s only because those bankers believe that they can structure the flow of most of that money, over time, so that it can get from the public sector to the privately run infrastructure back to the public sector in benefits. The money still comes from you, the taxpayer or user.

PPPs, done right, are a way of paying for asset depreciation. Governments are good (sort of) at paying for construction costs, but tend to let regular maintenance slide. This neglect often requires more government borrowing to maintain an asset whose maintenance should have paid for not from the capital budget but out of operating expenditures. A good PPP will monetize the cost of operation up front, as the private-sector builder-operator will include it in the permanent stream of revenue it will demand either from the government or the public (or both). Of course, estimates of required ongoing maintenance could be off, and, there’s no reason why the government couldn’t start to be more responible for this task itself. Still, anything to add discipline to this part of the infrastructure problem is a good idea.

PPPs, done right, can achieve construction and operational savings, but. But, first, the government’s assessment of a private-sector company or partnership’s capacity to handle this job over the long term is important; choosing a private-sector “partner” is like assessing the quality of a company for a long-term investment in a penson fund. But, second, operational efficiency could be diluted by the labor requirements that the public-sector brings with it. For example, New York State’s Wicks law, which requires inefficient multiple contracts for many public projects, could be a problem here. “Expansion of PPPs would require significant changes in State law relating to public procurement procedures,” notes the report.

My Monday WSJ op-ed on the same topic (and in general agreement with the CBC’s findings) is here. After publication, my suggestion in the piece that we demand public-sector competence on infrastructure projects was derided in some quarters as unrealistic. The reality is that we’re more likely to achieve such competence, or at least get close to it, if conservatives as well as liberals accept that there’s always going to be a government role in infrastructure, and focus on holding government accountable for its role, however it chooses to execute it, rather than thinking that we can financially engineer it away.