Why Credit “Breaks” All at Once

In calm markets, credit ETFs feel easy. You can buy or sell in seconds. The price updates all day. The screen looks clean and steady.

At the same time, many of the bonds inside those ETFs trade rarely. Some bonds trade once a day. Some do not trade for days. Quotes can be “runs” or “indications,” which often means “best guess.”

Then stress hits. The ETF still trades, sometimes even more than before. But the bond market starts to feel jammed. Bid lists get longer. Dealers reply slower, or not at all. Spreads widen in a jump. Prices do not slide. They gap.

That is the shock people notice: the ETF looks liquid while the bonds feel stuck. If the ETF holds the bonds, why does the ETF keep moving when the bonds do not?

One story says, “ETFs are more liquid.” That is true for the share, not for the bonds. It names the effect but not the cause.

Another story says, “ETFs cause the move.” That also stops early. An ETF is a wrapper. It cannot change how corporate bonds trade. It can only change how quickly people can trade exposure to those bonds.

The strange part is the timing. Credit does not always leak stress. It can hold together for a while, then break fast. Price discovery looks like a cliff.

That “all at once” feeling comes from plumbing. The ETF is a fast market sitting on a slow market, linked by a gate that depends on dealer balance sheets.

A credit ETF has two layers of liquidity.

The top layer is the ETF share. Shares trade on an exchange like a stock. Many buyers and sellers can meet at once. Market makers can quote all day.

The bottom layer is the bond market. Corporate bonds trade over the counter. Trades are negotiated. Each bond is its own small pool. Even in good times, this market is not built for a constant two-way flow.

The link between the two layers is creation and redemption.

Only certain firms can do this directly. They are authorized participants (APs), usually large broker-dealers. APs act as the conversion gate between the ETF share and the bond basket.

In calm times, this gate helps keep the ETF price close to the value of the bonds it holds. If the ETF gets a bit cheap, someone can buy shares and redeem for bonds. If it gets a bit rich, someone can create shares with bonds and sell the shares. This is often called arbitrage, but it is not free. It uses balance sheet and funding.

When fear rises, many sellers choose the wrapper first. Selling the ETF is one click. Selling a long list of bonds is slow.

That choice creates a mismatch in speed. The ETF share market can process heavy flow right away. The bond market often cannot. The ETF can trade smoothly even when very few bonds trade.

This is the key point: ETF shares can change hands without any bond trading. The tape can look active while the bond market stays quiet. Smooth trading in the share does not prove smooth trading in the bonds.

The pressure shows up when the flow is one-way.

If selling keeps coming and buyers step back, market makers can end up holding more ETF shares than they want. They can hedge some risk, but credit hedges are never perfect. The longer they hold, the more they care about the gap between the ETF price and what the bonds can be sold for.

At some point, the market maker wants to convert the ETF position into the bonds, so the position can be closed or moved. That is where the AP gate matters.

When the ETF trades cheap enough, a basic stabilizer exists. An AP (or a firm working with an AP) can buy ETF shares, redeem them, and receive bonds. The trade makes sense only if the bonds can be financed and sold at a profit relative to the ETF price.

But redemption creates a new problem. The AP receives a basket of bonds. Those bonds are not cash. They are inventory.

Inventory is where corporate bond liquidity lives or dies. Carrying bonds ties up the balance sheet. It needs funding. It triggers risk limits. In stress, those constraints tighten together.

A bond position that was easy to hold yesterday can become hard to hold today, even if the bond itself did not change. Funding can cost more. Haircuts can rise. Internal limits can shrink. The position eats more capital.

So the AP faces a constrained set of outcomes:

There is no rule that forces APs to redeem. They can choose not to. If they step back, the gate narrows. When the gate narrows, the ETF can drift farther from the bond marks.

The bond market depends on dealers being willing to stand between buyers and sellers. In corporate credit, dealer ability is shaped by balance sheet use.

In good times, dealers can “warehouse” bonds. They buy bonds, hold them, and later sell to real-money buyers. In stress, warehousing gets expensive. Risk limits bite. Funding feels less stable. The dealer’s best move can be to quote less size, quote wider, or not quote.

This looks like a sudden loss of liquidity. But it is often a balance sheet valve closing. The system is saying: holding this risk now has a higher cost.

Once that happens, the ETF price becomes the fastest place for the market to clear. It moves until it finds buyers willing to take the other side given the new inventory cost. The bond market clears later, and it often clears in chunks when prints finally happen.

A gap is what you get when the next true buyer is much lower than the last one.

In credit stress, the “true buyer” is often not a person with a view. It is a balance sheet with room. If the room disappears, the price does not drift down in small steps. It jumps to the level that compensates whoever still has capacity.

So the jump is not mystery behavior. It is the cost of turning a fast-trading wrapper into slow-trading bonds when the middle layer is constrained.

The mechanism is a mismatch: a fast wrapper around a slower market, joined by a conversion gate that works only when inventories can be carried.

This explains why credit can feel fine, then break fast. It also explains why the ETF can keep trading while the bond market feels stuck. The share market clears by price. The bond market clears by balance sheet.

It does not explain everything.

It does not identify the spark in each episode. Shocks differ. It does not tell you what “fair value” should be, since it is about plumbing, not fundamentals. It also does not cover every ETF design choice, since some funds rely more on cash creation or cash redemption, which can shift where the strain lands.

And it does not deny real credit risk. Sometimes spreads gap because the odds of loss truly change.

Still, when the pattern repeats - smooth trading up top, stuck trading below, then a sudden gap - the same constraint often sits at the bottom: a liquid share built on an illiquid market, with a gate that narrows when balance sheets get tight.