The Structuring Process: How Tranching and Subordination Actually Work
Book: Structured Finance and Insurance: The ART of Managing Capital and Risk Author: Christopher L. Culp Publisher: Wiley Finance, 2006 ISBN: 978-0-471-70631-1
We’ve now entered Part 3 of the book, and this is where things get real. Part 3 is about structured finance itself. The actual mechanics of how deals get built. Chapter 13 is the foundation for everything that follows.
What Is Structured Finance, Really?
Culp defines it pretty simply. Structured finance is when firms raise money in a nontraditional way that also changes their risk profile. Or even better: firms raise money at a cost that’s independent of their overall creditworthiness.
That second definition is key. A company with a terrible credit rating can still get cheap funding if it structures things right. That’s the whole point.
Structured products are bespoke. They’re designed for specific borrowers or specific investors. They package cash flows in very precise ways to match a supply of funds with a demand for specific risk/return combinations. Some structures have become standardized over time, but they all started as custom jobs.
The Menu of Structured Solutions
Culp walks through the main types of structured deals using economic balance sheets. Here’s the rundown:
Secured debt. The simplest form. You set aside specific assets as collateral for specific liabilities. Investors get a perfected security interest. If the company goes bankrupt, those assets aren’t available to unsecured creditors. They belong to the secured claimants.
Asset securitization. You actually sell assets to another entity. That entity issues new securities backed by those assets. The original firm gets cash and transfers the risk. This is different from just pledging collateral because the assets actually leave the balance sheet.
Ring-fencing. You separate certain assets (or a whole business) into a separate legal entity. Maybe a subsidiary. The goal is to isolate those assets from the rest of the firm. This shows up in project finance all the time. A below-investment-grade oil company might ring-fence a promising oil field so it can borrow against that specific asset rather than against its lousy overall credit.
Venture capital formation. Set up a new company around a single asset or project. Bring in investors to share the risk and reward.
Funded risk transfer. Structure your liabilities so that they help manage your risks. Instead of issuing normal debt and buying call options on copper separately, a cookware company could issue structured debt where coupons decline when copper prices rise. Same economic effect, but way easier for investors to monitor.
The Step-by-Step Process
Here’s how a structured deal actually gets built:
1. Figure out why you’re doing this. This sounds obvious, but Culp emphasizes it hard. You need to answer: why structure instead of using normal financing or normal risk transfer? If you can’t answer that clearly, don’t do it. The answer also determines which type of structure to use. And structuring should always be motivated by commercial fundamentals, not just to game accounting or tax rules.
2. Build a cash flow model. Model the revenues from the target assets and how that income flows to expenses and liabilities. This is the “waterfall.” Senior expenses (trustee fees, custodian fees) are fixed and don’t depend on deal performance. Subordinated expenses (success fees) come from what’s left over.
3. Hire a structuring agent. This is the expert who designs the deal. Could be a bank, an investment bank, an advisory firm, an insurance company. The earlier they’re involved, the better.
4. Find your investors. You need to know who wants to buy your securities before you design them. Most structured products are privately placed, so knowing your buyers matters.
5. Design the institutional structure. This is where special purpose entities (SPEs) come in. Not every deal needs one, but many do. SPEs can be corporations, trusts, or various hybrid forms. Sometimes you need multiple SPEs because no single structure can satisfy all the tax, accounting, regulatory, and disclosure requirements.
6. Design the securities. This is the hard part. You model cash flow waterfalls extensively. You figure out maturities, tranching, subordination levels, target ratings, and what credit or liquidity enhancements you need.
And here Culp drops a great warning about Enron. Overstructuring is a real danger. If your structure has too many SPEs with too many weird securities and derivatives, people will get suspicious. Sometimes complexity is genuinely required. Fine. But you better be prepared to explain it clearly.
Tranching and Subordination: The Big Idea
This is the most important section of the chapter. Culp makes a connection that’s honestly pretty elegant: tranching a capital structure is exactly the same thing as designing a reinsurance program.
Here’s how it works with a simple example. A firm has $1 billion in assets and issues:
- $300 million in senior debt
- $500 million in junior debt
- $200 million in equity
Now think about what happens when asset values decline. Equity absorbs the first $200 million in losses. Junior debt takes the next $500 million. Senior debt is only at risk for the final $300 million.
Written in reinsurance language, that’s:
- Equity provides “$200M excess of $0” (the deductible)
- Junior debt provides “$500M excess of $200M” (the first reinsurance layer)
- Senior debt provides “$300M excess of $700M” (the top layer)
Senior debt effectively has $700 million in credit protection from junior debt and equity. That’s a 70% attachment point. The capital structure IS a credit insurance program.
Loss Distributions and Ratings
Now Culp connects this to real ratings. Take a portfolio of high-yield debt. You can generate a loss distribution showing the probability of various levels of cumulative defaults.
Rating agencies publish guidance about what default probabilities correspond to what ratings. So you can overlay ratings onto the loss distribution. The breakpoints between ratings become the attachment points for your tranches.
For a typical high-yield portfolio, the AAA-rated portion might represent around 70% of the total capital structure. The first-loss equity piece might be relatively large too. But these numbers change for every portfolio and every choice of how many tranches to create.
The conceptual framework is remarkably flexible. You could do the same analysis for a traditional excess-of-loss insurance program. Or for a portfolio exposed to any risk, not just credit risk. The attachment points would be different, but the approach is identical.
Why This Matters
Chapter 13 is the Rosetta Stone for everything that follows in the book. Once you see that capital structure equals insurance program, the rest of structured finance makes way more sense.
Securitization? It’s just credit reinsurance done through securities instead of insurance contracts.
CDOs? Multi-layer credit reinsurance sold to capital market investors instead of reinsurance companies.
Understanding this equivalence is what separates people who can design and evaluate structured products from people who just memorize terminology.