Ideally, a federal infrastructure loan program should be more than a zero-sum transfer payment of some part of a loan’s cost from federal taxpayers to the borrower. If that’s the goal, it would be better (and more honestly) done as a grant. Instead, if the loan includes specific features that the federal government is more economically efficient at providing than the private debt market, then there’s a chance that overall efficiency is increased. In real world terms: fewer financial resources are used to achieve the same outcome (e.g. new infrastructure) and the cost to federal taxpayers is less than the national benefit. Maybe not much — but the right direction.

Of course, there aren’t many things where the federal government is actually more efficient at providing than the ultra-efficient private debt markets – assessing complex credit risk is definitely not one of them. But here are two that I think qualify:

• Interest rate management – due to gigantic economies of scale the US Treasury can access in managing the national debt. The marginal cost of sharing a little bit of this management expertise in connection with a rate lock on an infrastructure loan during construction is probably very small – certainly much less than the cost of the usual transaction-specific bespoke arrangement.

• Very long-term loans – notwithstanding current political fireworks, the US as a borrower (and as presumed repayor) is a forever thing. Or forever enough – even the Roman Empire took four centuries to fall. So the federal government can also be near-forever lender.

The reality of both of these features is much more nuanced, but it’s accurate to say that interest rate management and very long-term lending are strengths of the federal government. A loan program that can package and deliver these features efficiently to qualified borrowers (i.e. creditworthy and willing/able to go through the policy hoops) will, exactly like a private sector business with lower production costs, help move the invisible hand of an efficient economy in the right direction. And where are these features especially valuable? Financing infrastructure projects with long construction periods and long useful lives. As in basic water infrastructure. Which brings us to today’s poster program.

There is an issue at the WIFIA Loan Program. But not the one that Congress is focused on.

A Bond Buyer article about the latest water funding bill notes in passing that the WIFIA Loan Program “is designed to work with bonds and other funding sources.”

Since WIFIA limits its share of the project’s capital costs to 49% and requires an investment-grade rating, the borrower must find the 51% balance elsewhere and (given the rating) that’ll most likely include debt markets, including bonds.

WIFIA’s policy theory is that a selected project won’t happen soon (or at all) without a WIFIA loan. In that sense, WIFIA and bonds are designed to work together to unlock new infrastructure investment. CBO’s legislative scoring of the WIFIA Program reflects this expected outcome – it’s assumed that the 51% non-WIFIA part of project capitalization is financed with tax-exempt bonds that wouldn’t have been issued otherwise.

The reality over the past three years has been quite different. The vast majority of WIFIA borrowers are highly rated public water agencies. They’re almost universally financing projects that certainly will happen regardless of a Program loan. Without a WIFIA loan, these borrowers would simply issue 100% water revenue bonds in the usual way.

It would be more accurate to say that WIFIA “was designed to work with bonds but in fact mostly just replaces them”.

A short version of the June 30 white paper just published in Water Finance & Management, a leading water sector magazine.

This article was scheduled before the recent moves by the House Appropriations Committee to effectively defund the program in 2021. I don’t know (at all) any of the politics behind this move, but on the surface it certainly is surprising. The WIFIA Program works well, doesn’t cost much, is totally low-risk and accomplishes completely non-partisan stuff (voting against water infrastructure? Really?). And defunding it in the middle of a economic crisis that’s expected to hit state & local infrastructure agencies especially hard? I just don’t get it.

Here’s a quick picture about where all this focus on the WIFIA Loan Program is going:

Why? Because delayed investment in US basic infrastructure is actually a gigantic state & local liability, on top of all the others. It’s a type of expensive and dangerous off-balance sheet debt that needs to be refinanced on-balance sheet as quickly, cheaply and efficiently as possible. Covid-19 just makes everything worse — the kicked can is hitting the wall.

Solutions? Not P3s or private equity. Not project financing. Not magic unicorn technology. Probably not giant federal grant transfers either, but if that happens it will come with its own set of serious economic and jurisdictional problems — there’s no free lunch.

Instead, the Fed’s MLF shows a realistic path. Washington recognizes the problem and although big-scale infrastructure funding is mostly talk, they are in fact willing to deliver big-scale financing. The WIFIA program is effectively a prototype that shows how this is efficiently done for long-term infrastructure loans, basically as an extender to muni bonds. Putting parts of the two together in scale and for a broad range of municipal infrastructure is the basis of something that’s big, simple and cheap enough to actually work. That’s the point.

The WIFIA Loan Program is not like TIFIA. But could it end up like US ExIm?

Federal infrastructure loan programs are meant to improve US infrastructure.  Obvious, right?  That’s probably the outcome when the program makes project finance loans.  But the story is more complicated when a program actually makes highly rated balance sheet loans.

Does this techy financial distinction matter?  Yes – a lot.  If you want federal infrastructure loan programs to accomplish very specific goals (e.g. build more new toll roads), then subsidized and customized financing for a specific project is the way to go.  But if the goal is to more broadly improve US infrastructure in a particular sector (e.g. better water systems), and there are plenty of established and credit-worthy local public agencies in that sector already responsible for this, it’s more straightforward to make balance sheet loans to these agencies on terms that encourage (but don’t dictate) better infrastructure outcomes.

This difference is playing out now in two federal infrastructure programs that are very similar in legislative design but headed in radically different directions – TIFIA and WIFIA.