Anatomy of a High Yield Credit Freeze
Anatomy of a High Yield Credit Freeze
An Auto Loan Case Study
Sizing Up High Yield Markets
Economies can be simplified into bundles of transactions.
Transactions require money as a medium – there are 2 ways to get money to spend in a transaction.
Money from income
Money from debt
Money from income is easy to derive — it’s a function of wages and employment.
Income sources can be reasonably anticipated with demographic modeling and some productivity assumptions.
Debt is more complex.
The range of leveraging/deleveraging in debt derived spending is what creates cyclical pressures in an economy.
Recently I’ve been focused on the dislocation in high yield corporate credit markets.
On it’s own – HY credit is not a meaningful market as measured by market size — just 14% of the size of the US treasury market.
But high yield companies tend to have capital withdrawn and prices reset the quickest.
When high yield companies retrench spending, it has a daisy chain influence on other corporates & markets.
Here’s a flow
High yield food company faces higher borrowing costs
Cuts advertising budget
Impacts Google topline
Google guides topline lower which moves markets
Markets lower imply higher risk premia in high yield
Cycle repeats
If multi-billion dollar high yield corporate borrowers experience stress — there is significant read through into SME businesses (think strip mall businesses) that are more difficult to guage with public data.
Just how sensitive are companies to a change in borrowing costs?
In March-2022 I calculated the sensitivity to a 2% change in borrowing costs to guage how a rising floor of the price of money ripples into profitability.
Using median S&P 500 metrics for leverage (which is overwhelmingly investment grade quality) a 2% change in debt costs leads to a 9% change in after-tax EPS.
This is exaggerated given most debt is fixed-rate in the S&P.
But my analysis also didn’t capture higher loan-to-value (LTV) ratios that result when equities sell off due to revalued P/E ratios from rate hikes.
Credit investors heavily rely upon LTV ratios to price credit in a relative value framework.
For instance if a company had $25mm of debt and $75mm marketcap it would have $100mm of enterprise value. This the market’s assessment of asset value.
Suppose you’re the sole lender for the $25mm and you made a 8% fixed rate loan — you’d tell your investment / risk committee you are getting paid 8% to be first 25% LTV in the business. This reflects how much risk value is beneath you in % terms.
LTVs can change rapidly depending on the business. If your equity gets cut from $75mm to $30mm - you become 45% LTV. Credit is a short volatility position on enterprise value.
Lenders look to get paid more as LTVs rise given closer to the money strikes on enterprise value.
Borrowing costs for US corporates have magnified since the beginning of the year due to treasuries / credit spread widening.
It is not sensational to say HY spreads have blown out in the last 2 months as capital has become scarce.
High Yield 5-Year CDX Spreads Last 10 Years
Issuance has effectively gone to 0 as companies have little faith investors will clear their deals and banks retrench from getting hung on deals.
I want to discuss how high yield stress can influence other markets and quickly sew the seeds of a reflexive recession.
This means enjoining a macro premise with a corporate bottoms-up framework — something few investors are capable of given silo-ized backgrounds (few macro investors know how to read an income statement).
Autos
Buying a car is the second most important spending decision the average person makes.
In the US – having a car enables commerce in a way that isn’t necessary in Europe or Asia given public transport networks and city density. You need a car if you want to have a job.
Because autos are a disproportionally large spending item vs income — it necessarily involve the use of debt.
80% of new car buyers use financing which is arranged by the auto dealer.
Auto loans are lower quality vs mortgages loans as measured by median FICO. As a result - we will see stress in the consumer via auto delinquencies before most other markets.
Let me give you a mental model for how auto ‘risk’ is cleared in the US:
Ford makes a car for $45k
It sells that car to a franchised auto dealer for $50k — franchises have exclusivity arrangements with OEMs like Ford
The auto dealer uses a line of credit to buy the car — that line of credit is subsidized by Ford and is called ‘floor-plan’ financing
The dealer takes ownership of the car on the lot and waits ~40 days to find a customer
The dealer sells the car for $52k and nets a $2k ‘metal’ profit
That customer can’t pay in cash so it needs to borrow moeny
The dealer arranges financing for the customer which ultimately comes from investor securitization vehicles
The dealer gets paid a financing commission and sells the customer warranty that gives them another $2k
The dealer’s profit pool can be simplified as
$2k metal profit + $2k financing/warranty = $4k
$50k inventory financing (x) interest rate (x) 40 days held / 365 days
Here is what gross profit per unit looks like with a range of interest rates and inventory days
If you read closely - this series of transactions involved 2 large sources of financing:
The dealer acquired a subsidized line of credit from the OEM to buy the car
The dealer arranged debt to finance the customers purchase via ABS markets
For a dealer to have inventory to sell it must have access to a line of credit called ‘floor-plan’ financing — dealers use this to buy cars from the OEM.
Floor-plan financin has refinancing ‘roll’ risk given they are 9-12 month duration and are subsidized by the auto OEMs.
When dealers sell the car to the customer, they pay down the floor plan financing which allows them capacity to reborrow and buy more cars from the OEM.
Auto financing in the US is set by the securitization conduit.
Securitization is a source of funds from investors who pool capital. Risk tranches are divided and priced to suit investor risk/reward tolerance. This creates a ‘tower’ with different flavors. The weighted average rate of the tower is what the customer pays.
Securitization towers are useful because sometimes 1 + 1 = 2.25. There is more capital efficiency when we chop risk up to reflect investor preference — but a natural result of towers is codependence on investors.
Here is an ABS comp sheet from a OneMain Oct-2021 investor presentation
When credit markets freeze up — investors are in a Mexican standoff as to what the right price of credit is. This is because credit prices are often derived from relative value within the tower. This is similar to how credit reprices based on LTVs.
When high yield spreads widen — everyone else wants to get paid more. That means customers end up paying more to clear the risk in the tower. At some point affordability measured as monthly payments vs income becomes a gating item.
To review — auto dealers have thin margins that asymmetrically rely on functioning credit markets.
Dealers borrow from Ford to buy Ford cars until a customer shows up.
But that customer needs financing from a securtizaiton tower that’s risk premia is moving around with market volatility.
Sentiment around new car sales or broader consumer issues can rapidly change the capital availability and price in credit.
Implications
Credit markets reprice in 2 distinct regimes:
When recession odds go from remote to possibility
When recession goes to possibility to probability
Credit markets follow equities lead in terms of direction but the magnitude of an equity sell-off is reflexively dependent on credit prices / availability.
That’s why in a bear market — investors like Stan Drunkenmiller focus on IG / HY credit spreads to frame a bottoming. Credit prices reverberate across corporate borrowers and consumer financing towers.
Simply put - credit must bottom for equities to truly bottom. Bear market rallies in equties that do not drag credit along are better to fade.
The recent repricing in credit may not look substantial on a 20 year look back — indeed 2016 and 2012 cycles show much wider spreads.
But HY spreads in 2016 was skewed by a true 0-to-1 default cycle in energy which was the largest industry issuer in the index. Additionally - rates were effectively 0% for most the last 10 years and credit spreads typically grind tighter in a rising treasury rate environment.
Looking at HY Yield-to-Worst ex Energy seems like a better way to compare where we are today vs history.
High Yield Yield to Worst ex Energy Since 2015
This will be the first ‘reset’ in the economy that will not have a drop in interest rates along side it. Previously, the Fed would pivot when it began to see the credit market freeze — like in 2018/19 — but now its inflation mandate prevents pre-emptive action in the credit market.
My premise is that high-yield credit costs reverberate throughout the economy as everyone reprices risk to calibrate to public markers. Equity sell-offs create higher LTVs in public high yield markets which then create a daisy chain to price effects.
The auto ecosystem is one example where the profit pool requires issuance of debt across multiple stages. I expect to see auto delinquencies as a front-end consumer indicator of credit issues.
Auto credit books are lower quality than mortgage orginations and residual value / negative equity risk is more pronounced in autos vs housing. High yield market spillovers will show up here in a way the market will notice given the size and inter-relatedness of the auto loan market.
To be clear — auto loans are not the reason the economy is going to get stuck but instead its weakness will be the result of lower real wages. The best argument for a “soft landing” is the $4trn above trend consumer cash deposits and the $23trn of wealth creation since 2019 on household balance sheets (detailed here).
But here the challenge has always been the distributional impact of this wealth creation. The marginal propensity to spend for wealthy consumers is much lower lower than middle/low income households. The consumer resiliency thesis becomes stale if we see material consumer credit issues in a core area like autos.