Korea's convertibles create new shares at a 24% discount. Cash raises, 12%.

A Korean company that needs equity has two ways to get it. It can sell new shares for cash — a rights issue to existing holders, or a public offering. Or it can let a convertible do the work: a bond, a warrant, or a debt-for-stock swap that turns into shares at a price fixed in advance.

Both create new shares. Both dilute everyone who already owns the stock. The question a shareholder should ask is the same in each case: how far below the market did those new shares get printed?

The answer, measured on the day the shares appear

For every issuance event since 2020 that carried a usable price, we took the company’s own closing price on the day the shares were created and asked what discount the new shares came at. Positive means the shares were issued below market — dilution at a markdown.

A cash raise prints new shares at a median 12.5% below the going price. A convertible prints them at 23.8% — nearly twice the markdown.

That is the number that matters to a holder, and it is the opposite of reassuring. The instrument that companies reach for when they want financing to look painless — no discounted rights offer, no headline placement — is the one that quietly creates shares at the deeper discount.

Is it the instrument, or the companies that use it?

Here is where this kind of number usually falls apart. Convertibles are not spread evenly. Smaller, more volatile, more cash-hungry companies lean on them, and those companies might dilute more heavily whatever route they took. If so, the 24% is a fact about the companies, not the convertible.

So we held the company fixed. Of the names in the sample, 456 used both routes — the same issuer, at different times, raised cash and converted a bond. Inside each one, we compared that company’s own convertible discount to its own cash discount.

• In 288 of 456 (63.2%), the convertible discounted deeper than the same company’s cash raise.
• In 166 (36.4%), the cash raise was deeper.
• The median company converted its bonds 7.7 percentage points below where it sold its cash shares.

The gap shrinks when you control for the company — from an 11-point spread to about 8 — but it does not vanish. Most of it is the instrument. A convertible, in the same company’s hands, still creates shares at a visibly deeper discount.

Why the raw prices oversell it

There is a louder version of this story, and it is wrong. Line up the issue prices themselves and convertibles look like they print at barely half the price of a cash raise. That comparison is a trap, and it is worth seeing why.

An issue price on its own says nothing about a discount. A convertible that turns into stock at 3,000 won on a 3,300-won share is a small markdown; a cash raise at 6,000 won on a 12,000-won share is a large one. The convertible has the lower price and the smaller discount. Reading dilution off the raw price rewards exactly the wrong issues — the cheap-stock names where convertibles cluster — and manufactures a gap that is mostly a fact about share prices.

Measured properly, against the market on the day the shares appear, the real gap is about 8 to 11 points of discount. Smaller than the raw prices shout, larger than zero, and pointed the same way every time you tighten the lens.

What the discount is

None of this makes the convertible a trick. Its discount is structural, not stealthy. A convertible fixes its conversion price in advance; the shares appear later, usually because the stock has risen enough to make conversion worthwhile — which guarantees they arrive below the market of the day. The 23.8% is that mechanism working as designed.

But “as designed” is the point for a shareholder. A cash raise announces its discount up front, in a document everyone reads. A convertible’s discount is set months earlier and paid out later, in shares, on a day nobody is watching for it. Same dilution, delivered where it is harder to see — and, in the same company’s own hands, delivered deeper.