Why do two bonds from the same company trade at different spreads?

One issuer. One balance sheet. One set of business risks. And three bonds quoting 220bp, 320bp and 520bp. If they all default together — and they do, because default is an event at the issuer level — why are they priced so differently?

Same PD, different LGD

The default probability is shared. What differs is the loss given default, and the expected loss formula converts that straight into basis points.

An illustrative issuer with an implied 3% annual default probability:

Read the columns. Expected loss explains 60bp of the 100bp gap between secured and unsecured, and 60bp of the 200bp gap between unsecured and subordinated. The rest is risk and liquidity premium — and notice that the premium grows as you descend the structure.

That growth is not arbitrary. Junior claims are not merely exposed to more loss on average; they are exposed to the uncertainty about enterprise value. As the worked waterfall in this unit showed, a modest change in what the business turns out to be worth barely touches the secured recovery while swinging the junior recovery from something to nothing. Investors charge more than proportionally for that convexity.

The other reasons two bonds differ

Seniority is the biggest term but not the only one:

• Maturity. Longer bonds carry more cumulative default probability, so credit curves normally slope upward.
• Optionality. A callable bond quotes wider on a Z-spread basis because part of the spread pays for the option you sold. Compare on OAS, as Measuring spreads insisted.
• Size and liquidity. A small, rarely traded issue carries a bigger liquidity premium than a large benchmark issue from the same company.
• Structural position. A holdco bond and an opco bond can both say "senior unsecured" and rank quite differently.
• Covenant package. Better protection supports a better recovery assumption, which is a lower LGD, which is fewer basis points of expected loss.

When the credit curve inverts

Normally, longer means wider: more time, more cumulative risk. For a distressed issuer this flips. If the market's central question becomes "can this company get through the next eighteen months," then the short bond carries most of the survival risk, and short-dated paper trades at an enormous spread while longer paper trades tighter — because a company that survives the crunch will probably be restructured in a way that treats all its debt similarly.

An inverted credit curve is therefore the market pricing near-term survival as the binding question. That is a description of what a price structure means, not a prediction about any outcome, and certainly not a signal to act on.

The practical habit

When you see two spreads from one issuer, ask in order: same seniority? same maturity? same option features? same entity in the group? same issue size? Only once those match are you comparing credit views rather than structural differences. Most apparent mispricings dissolve at question three.

Try it now

Separate the effects:

1. Pick an issuer and compute expected loss for a secured, unsecured and subordinated claim using PD of 3% and recoveries of 60%, 40% and 20%.
2. Note how many basis points of spread difference the arithmetic alone justifies.
3. Now the observed side, and notice what is missing. Spreads by seniority for one issuer come from dealer quotes on its individual bonds; a published market curve, like the high-quality corporate curve below, is one yield per maturity built from many issuers, with no issuer and no seniority in it. So compare your arithmetic with the quoted spreads in the table at the top of this lesson, 220bp, 320bp and 520bp, gap by gap. Anything left over is premium, and describing it that way, rather than calling it cheap or expensive, is the correct professional posture.