There is an article in the Financial Times today concerning equity funded collars [1]. The equity collar structure is used by a counterparty $A$ which wants to build up a position in a stock $S_t$. Let $B$ be the investment bank arranging the transaction, then the structure is as follows:
- $A$ enters into an equity collar with $B$ on shares $S_t$, i.e. $B$ sells a put with strike $k$ to $A$ and buys a call with strike $(k+\varepsilon)$ from them;
- Given the collar has positive delta to $B$ (short put + long call), to delta-hedge it the bank borrows stock $S_t$ which it then sells to $A$.
Another way for $A$ to build a position in the stock would be a margin loan, in which $A$ buys stock $S_t$ with a loan from $B$ which is collateralized by the bought stock (i.e. a repo-like transaction). Cash margin calls ensue if the stock price starts falling below predefined levels.
Now, the article claims that
[...] banks like [collars] because instead of taking on the credit risk of the borrower [as in a margin loan] , they take on the market risk of the underlying stock, which they can hedge as the share price fluctuates.
I understand the delta-hedge on the collar is imperfect (partial hedge, discrete rebalancing) plus there is gamma exposure, hence I understand the market risk coming from the collar structure (although it could be argued a margin loan also has market risk given it is collateralized by the stock). I also understand the credit/counterparty risk coming from the margin loan transaction, as a counterparty defaulting in the midst of a fast depreciation of the stock would leave the bank exposed to the gap between the latest margin call and the value of the stock collateral.
However I struggle to understand why there is no credit/counterparty risk in the collar structure. Can anybody explain?