Three federal infrastructure loan programs – WIFIA, TIFIA and CIFIA – collectively offer loan features that are useful and effective for long-term infrastructure financing. Usually, the features are provided through a program loan made directly to a qualifying project. But there’s also an indirect path – WIFIA and TIFIA can lend to water state revolving funds (SRFs) and rural state infrastructure banks (RSIBs), respectively. These entities then can make separate, smaller loans that incorporate or reflect the program loans’ features.
In these cases, the ‘wholesale-to-retail lender’ approach has some sector-specific policy objectives. For the SRFs, the goal is to encourage leverage at smaller, unleveraged funds, thereby increasing the impact of federal grant funding. A more expansive effort was proposed by the SRF-WIN Act in 2018, but it did not pass. For RSIBs, there’s a significant interest rate subsidy (50% of UST base), in effect providing a federal grant to states for their rural sectors.
Mass-Producing Loan Features and Outsourcing Transaction Origination
But now that the federal program lending to state lenders approach is statutorily established and (to some extent) utilized, there’s a more generalized way to look at it. The programs’ loan features are essentially financial, not sectoral. They have been designed and approved to provide a type of financing to US infrastructure that will encourage overall infrastructure policy outcomes – sustainability, resilience, social and economic goals, accelerated renewal, etc. The features can work towards these goals at specific projects regardless of whether they’ve been delivered directly by the program or indirectly by another lender that has received a program loan. In many ways, incorporating these financial features is the easiest part of loan origination and execution – they get written into the documents with more-or-less standard language and operate in much the same way for all loans. Unlike other aspects of federal infrastructure loan program transaction processing, loan financial features can be mass produced.
In this context, federal infrastructure program lending to state lenders is a type of outsourcing that can deploy financial features based on federal capabilities to a wider range of infrastructure projects than the programs can do on their own. Given the huge scale of the US infrastructure renewal challenge, especially relative to the size of federal infrastructure loan programs, shouldn’t such outsourcing be an overall policy objective regardless of specific sectoral applications? If the loan features are useful in encouraging infrastructure outcomes, can be mass produced with existing federal capacity, and are deliverable through other public-sector lenders that are already originating infrastructure investments, then why not?
Infrastructure Pension Obligation Loan Program
There are likely various ways to implement outsourcing federal infrastructure loan features, ranging from simply expanding existing processes (e.g., more funding and features for SWIFIA, WIFIA’s SRF section) to an open program for all qualified lenders, including those in the private sector. For this post, I’d like to sketch out one that would involve state-level public-sector pension funds. I think this will illustrate the potential of outsourcing even within the public sector but also highlight some of the challenges.
State pension funds certainly have scale and investment capabilities. There are about 300 funds with about $4 trillion of assets. Obviously, most of their investment operations aren’t connected to financing long-term infrastructure projects, which would likely fit into their private market fixed-income and alternative buckets – maybe 1% or 2%? Still, that small percentage is about $40-80 billion in portfolio volume, roughly the current aggregate capacity of TIFIA, WIFIA and CIFIA combined.
There’s no question that the investment teams of state pension funds would know how to utilize the financial features of federal infrastructure loans. The harder part is how to attach and ensure compliance with both standard federal crosscutters (e.g., Davis-Bacon, NEPA, tec.) and infrastructure policy objectives (resiliency, sustainability, etc.). And there will be a related question about how the benefits of the loan features should be shared between the pension fund and the ultimate borrowers. The benefits should motivate both the pension fund and the borrower to implement policy outcomes that wouldn’t have happened otherwise – but how to split them? Is rigorous oversight necessary to prevent windfalls to either? Pensions and borrowers will obviously make arms-length decisions based on self-interest, so there’s an element of market discipline on the borrower side. To the extent that potential windfalls accrue on the pension side, it’s worth pointing that these are public-sector funds, so the ultimate beneficiaries are the state’s citizens. That’s not so different from a typical WIFIA loan where it’s unclear whether the benefits are improving the infrastructure project or simply being used to reduce the community’s water rates.
Federal infrastructure loans to state pensions could be implemented through special sections in existing sectoral programs, as WIFIA and TIFIA do for SRFs and RSIBs, respectively. But perhaps the potential scale and specialized aspects of the lending would justify a separate program. To simplify the illustration of the concept, especially in connection with comparison to state pensions’ bond alternatives (discussed below), this post will assume the latter as a ‘Infrastructure Pension Obligation Loan program’ or IPOL.
Not a Bailout
There’s one thing to make very clear right from the start – the IPOL program is not intended as a bailout [1]. US public pension underfunding is a serious issue. It rightly gets a lot of public discussion, including the possibility of a federal bailout at some point, something underscored by a recent $36 billion rescue of private, multi-employer plans. More specifically, a 2020 proposal for federal infrastructure lending to pensions was explicitly intended to alleviate pensions’ underfunding, with the infrastructure focus being essentially a bipartisan side-benefit. Despite the proposed program’s WIFIA-like acronym, the proposal got no traction and apparently little attention. But the political dynamics are there.
Because political temptation to mix and match separate issues will always exist, the IPOL should be designed with anti-bailout characteristics. The focus should be explicitly and solely about federal infrastructure policy objectives and how state pensions (regardless of their unfunded status) can help implement a federal loan outsourcing approach – the infrastructure projects are the point, not the loan delivery mechanism. Pension qualifications, basic subsidies and federal appropriation levels should reflect that focus – a high minimum investment-grade rating (perhaps AA/Aa2?) and US Treasury flat minimum pricing should allow IPOL program credit subsidy leverage to be about 100:1, as it is in WIFIA. Even a $50 billion program would require only about $500 million in discretionary appropriations – that’s enough to encourage policy-oriented infrastructure lending, but it’s not bailout territory.
IPOL Loan Features
As with other federal infrastructure loan programs, the IPOL’s central mechanism to implement policy outcomes is the value of program loan features. For those, the place to start is what’s already established and offered at WIFIA, TIFIA and CIFIA. There are seven that I think are especially relevant.
Four of the features concern interest rate management during transaction origination and portfolio management thereafter:
- UST Rate Lock: As noted above, basic pricing should be the applicable US Treasury rate at loan execution. More importantly, loan drawdowns at this fixed rate would be optional within a period of years (maybe 3-5?) to allow the pension fund to originate and process investments. The post-execution rate lock during drawdown is a common feature at the three existing programs, but it was expanded at WIFIA in construction financing terms for long-term capital improvement programs at utilities, not just specific projects. Portfolio assembly is analogous [2].
- Rate Lock Reset: The IPOL program can reset of the rate lock if Treasury rates fall before drawdown. This is not statutorily required, but it’s an established procedure at WIFIA and TIFIA.
- Limited Buydown: The program can set a limited interest rate cap at the date a loan application is accepted, which may be months or even years before loan execution. This permissive feature is statutory in TIFIA and (with additional details) at CIFIA, though currently absent in WIFIA for some reason.
- No Cancellation or Prepayment Penalties: This is a standard feature in all three existing programs.
Three other features involve non-pricing aspects of the loan:
- 55-Year Term: TIFIA offers loan terms of up to 75 years for projects with long useful lives and a 55-year term is being proposed for WIFIA. Assuming that the pension is building a long-duration portfolio in an actuarial context of 60-70 years, an IPOL term of 55 years (at least?) would seem to be analogously justified.
- Credit Rating Based on Portfolio or Pension Recourse: As mentioned above, IPOL loans should require a high investment grade rating. The rating may be based on the expected (and demonstrable) rating of the portfolio when funded or, more practically, on recourse to the pension. Most states are rated at least AA/Aa2 and generally backstop pension obligations.
- 80% Portfolio Leverage: CIFIA offers loans of up to 80% of project cost and the same leverage was proposed in the SRF-WIN act for SWIFIA loans. If IPOL’s goal is to outsource in scale, a high leverage ratio would make sense when combined with high minimum credit quality described above.
IPOL Loans vs. POBs
The value of WIFIA loans to highly rated public water agencies is assessed by a comparison to their debt market alternatives, tax-exempt revenue bonds. In the same way, federal IPOL loans should be compared to the state pensions’ financing market alternative, pension obligation bonds, or POBs. Despite the pensions’ public-sector status, POBs aren’t tax-exempt. As a result, direct arbitrage is rare, and POBs are used by only a few states to leverage equity returns, a somewhat risky strategy which depends a lot on market timing. IPOL loans cannot be used for that purpose — the benefits of the loan features (most notably the UST interest rate and rate lock) should be directed towards (and perhaps limited to) low-risk, long-term infrastructure investments like highly rated private placements. Some positive return should be expected to provide motivation for policy implementation but as discussed above, not excessive windfalls.
The risks of POB issuance are well-recognized. IPOL loans should have a far lower risk profile — here’s a comparison with GFOA’s current advisory on POBs (click on to enlarge):
Next Steps?
At this point, the concept of outsourcing transaction origination to deploy federal infrastructure loan features is a thought experiment and the IPOL program simply an illustration. This post is really just an initial outline for future discussions if there’s interest — which I think there will be.
_____________________________________________________________________________________________
Notes
[1] More precisely, not more of a ‘bailout’ or subsidy than federal infrastructure loans are already in terms of transfer payments to specific communities. Possibly less, if loan features are explicitly designed and assessed in terms of real-world infrastructure additionality.
[2] Readers of this site might wonder how including this and the other interest rate features will not lead to the FCRA loss ratchet described in previous posts on WIFIA’s economic cost. A good question that, if it’s ever raised by policymakers, perhaps ought to be answered in the context of fairness as opposed to logic. Federal infrastructure loan programs should work with the funding tools at hand, including FCRA subtleties, on an equitable basis as long as the fundamental goal of improving US public infrastructure is furthered. Realpolitik, in other words.