Two weeks ago, when looking at the recent surge in short-term funding rates in general, and Libor in particular, we said that this is the result of a scramble by various funds to change their asset ahead of an October 14 deadline for money market reform. Recall that "On October 14, 2a-7 money fund reforms will require some prime money market mutual funds (those that invest in non-government issued assets) to float their net asset value (NAV) or, under certain circumstances, to impose redemption gates and liquidity fees on redemptions. Rather than face these regulatory constraints, many investors have started pulling assets from prime funds, and a number of prime funds have converted to government-only funds (which are exempt from these regulations). Since late-2015 alone, prime fund assets have declined by nearly $450 billion, reducing the supply of dollars that funded private sector short-term liabilities."
Adding to the impact of the decline in prime fund assets, the behavior of the remaining prime funds has also pressured shorter-tenor funding rates. As part of the 2a-7 reforms, prime funds are required to hold at least 30% of their assets in securities that are convertible to cash within 5 business days, or otherwise face either liquidity fees and/or redemption gates. In anticipation of these changes, many prime funds have lowered the weighted average maturities (WAM) of their assets in recent months, reducing the supply of dollars available at longer tenors.
The result has been a spike in various market indicators that at least historically have been an indicator of broader market funding stress and liquidity shortfalls.
As we further explains, there was one notable consequence as a result of this regulatory intervention which is set to pressure what have traditionally been reliable metrics indicative of funding stress and systemic risk, among them swap spreads, the TED-Spread and the FRA-OIS spread, namely that "the market is about to lose perhaps the last metric indicative of underlying funding and liquidity tensions. After all, with central bank intervention having broken all conventional signalling pathways, including equities, corporate bonds and Treasuries, there will no longer be any reliable sources hinting at fundamental risk in the market, certainly for the short-term and perhaps over an indefinite amount of time."
However, while a blowing out Ted-Spread may no longer be a leading indicator and a warning of imminent liquidity crisis, the reality is that the as of this moment, Libor continues to blow out and is not only at the highest it has been in 6 years, but is now wider than the 3 Year Treasury. In fact, as of last night, 3-month Libor had not posted a decline since July 14.
We documented the real world implications of this in "Forget The Fed’s 0.25%, Short-Term Rates Have Already Risen By 1% For The Real World"
But while blowing out unsecured funding rates may no longer be a flashing red flag, a question has emerged as a lot of debt references Libor, debt ranging from household debt to non-financial business debt: some $28 trillion of it, to be specific, and just in the US. The question is just how concerned will the borrowers of said debt be once they get their next due balance.
Here is Goldman with the calculation:
Complete data on loan terms for all borrowers are not available, but we can make some rough approximations of the share indexed to LIBOR by combining a few datasets. For households, fixed-rate home mortgages account for the bulk of outstanding debt. Adjustable rate mortgages (ARMs)—which typically reference LIBOR—account for 20% of the value of loans outstanding and 15% of the value of new issuance, according to our mortgage strategy team. Additionally, ARMs are skewed toward higher dollar value loans, usually made to better credit quality borrowers (e.g. ARMs account for only 12% of the number of mortgage loans outstanding). Home equity lines of credit (HELOCs)—which account for 5-6% of household mortgage debt—typically reference the prime rate.
The remainder of consumer debt references a mix of benchmark rates. Government-provided student loans reference the 10-year Treasury yield (with infrequent resets), while private student loans frequently reference LIBOR. Interest rates on virtually all credit card debt are based off the prime rate, and auto loans are usually structured as fixed-rate term loans (with an average maturity of just over five years). Combined, the share of household liabilities referencing LIBOR likely ranges from 15% to as much as 20%, depending on how we classify loans in the “other” categories (Exhibit 1, left panel). In many of the loan categories there will be exceptions to norms, so this estimate should be considered only a rough approximation.
We find slightly higher numbers for the nonfinancial business sector (Exhibit 1, right panel). As of Q1 of this year, capital market borrowing accounted for 44% of business sector liabilities, with loans the remaining 56% (we exclude trade payables, taxes payable, FDI, and “miscellaneous liabilities” from the Flow of Funds Accounts’ definition of total liabilities). Although the Flow of Funds Accounts do not provide a detailed breakdown of these liabilities, we can make some rough inferences using the Fed’s Survey of Terms of Business Lending (STBL). According to this report, only about 13% of bank loans are now based off the prime rate—a fraction which has declined significantly over the last 20 years. Based on academic research which uses more detailed loan-level data, we assume the remaining loans typically use LIBOR as a benchmark. Construction and land development loans—a subset of commercial mortgages—are usually floating rate, and may be tied to LIBOR as well. Altogether, we estimate that roughly 25-30% of business lending is LIBOR-based.
The breakdown of nonfinancial debt referencing Libor is as follows: it amounts to just over $28 trillion, with trillions more added if one adds the financial Libor-referenced debt.
What is the real world implication of this? Simple: financial conditions are getting dramatically tighter, and just the recent spike in Libor rates across the curve, which amounts to roughly 50 bps for the 3M tenor, indicates that both households and businesses will have to pay up some $140 billion more, and substantially more if their contracts reference longer Libor maturities.
And while it means higher profits for the issuers of variable rate debt, it means less cash will be spent on discretionary purchases, something the Fed has been desperate to avoid by keeping rates near zero. In fact, the recent split between the Fed Funds rate and Libor suggests that the Fed's policy is now only partially operational. But what it certainly suggests is that, as noted earlier this week, "the US consumer that is acknowledged to be the last string the expansionary economy hangs by, has been dealt a de facto 1% tightening." That this is happening just as the consumer may be rolling over (according to BofA internal credit and debt card data), is a warning sign that the US economy, which in Q1 avoided a contraction only thanks to consumer spending, may be about to suffer an even greater slowdown once those who have Libor-tracking debt get their next payment invoice.