Download PDF with 020726 InRecap comments
EPA WIFIA and/or OMB seem to have recognized at some point after the first resets in October 2020 that restrictions were required. Unrestricted resets would have meant borrowers constantly lowering the commitment rate until they hit the absolute low in an interest rate cycle during a 5- to 10-year construction phase, all while using short-term finance for actual construction draws. That pattern would have resulted in the maximum possible upward re-estimate and funding loss when the loan was finally drawn (i.e., the financing ‘option’ was exercised).
The first announcements of resets didn’t mention restrictions. But the July 2021 ‘fact sheet’ spelled out two — only one reset was available per loan, and only up to the point where 50% of project eligible costs were expended. Since the reset has no statutory basis, these were self-imposed, either by the program itself or OMB.
Okay — a good idea. The restrictions would lower the probability and extent of future funding losses. But a downward reset of any kind increases that probability. They must have known this, yet AFAIK credit subsidy estimates at re-execution, like the estimates at original commitment, did not include any apportionment for future funding losses.
In effect, EPA WIFIA and OMB seemed to recognize the probability of funding losses in WIFIA loan commitments based on predictable borrower behavior, but other than minor restrictions on the resets, they did nothing else about it.
Why? What could have been the mindset? I don’t want to be cynical, but I can only think something like this: “Yes, we know there’ll be funding losses, but the FCRA re-estimate process will cover them with permanent indefinite authority for mandatory appropriations, and it’ll all be buried in off-budget accounts. However, we don’t want to be too obvious about what’s going on, and unrestricted resets might prompt some awkward questions. So, we’ll scale them back a bit — the borrowers and their lobbyists will understand.” [1]
There may well be more innocent explanations, IDK. But ‘awkward questions’ should be asked.
________________________________________________________________________________________
Notes
[PDF] A short note on the $699m SFPUC loan highlighted in the fact sheet. As usual for ‘narrative’ purposes, the $432m of ‘savings’ was on an undiscounted basis — the real NPV difference by lowering a 3.09% commitment to 1.45% was probably around $180m. Big enough, but SFPUC probably could have got about the same lower rate in the tax-exempt bond market at the time, though without the WIFIA loan’s optionality.
More importantly, the reset SFPUC loan probably wasn’t drawn until their short-term rates were higher than 1.45% — maybe 2023? By then, 20Y UST rates for funding the loan were well above 3.09% — meaning that the entire $180m of ‘savings’ to SFPUC ratepayers was a transfer payment via mandatory appropriations from federal taxpayers. Probably more, as most drawdowns may have occurred 2024-2025.
Note that if SFPUC had drawn the entire loan on the day of reset to 1.45%, taxpayers would not have borne any cost — UST buyers were offering that rate then. But not later — the borrower’s optionality for delayed draw (very handy when ST rates were low) was the cause of the funding losses. And what did that optionality accomplish, exactly, re US infrastructure renewal?
[1] Maybe this is the reason that water lobbyists never sought to make the reset statutory, something which is not otherwise easily explicable? It’s one thing for EPA and certain OMB officials to be in on the scheme, but quite another to get it through wider Congress, where even innocent questions might end up leading in the wrong direction. Better to keep a low profile and trust ‘our people’ to deliver, no? [Updated further thought: CBO would have to score such an amendment — they would definitely ask some awkward questions.]