A common belief: If corporates diversify away from banks, their liquidity risk falls.

What is often overlooked is that banks do not treat “funding diversification” as neutral. They actually price the spillover.

The hidden friction is 𝗰𝗼𝗻𝘁𝗶𝗻𝗴𝗲𝗻𝘁 𝗱𝗿𝗮𝘄𝗱𝗼𝘄𝗻 𝗿𝗶𝘀𝗸. When market-based lenders retrench, bank credit lines become the backstop, often at the worst possible time for bank balance sheets.

That changes 𝗲𝘅-𝗮𝗻𝘁𝗲 bank behaviour:

• 𝗛𝗶𝗴𝗵𝗲𝗿 𝗻𝗼𝗻𝗯𝗮𝗻𝗸 𝗿𝗲𝗹𝗶𝗮𝗻𝗰𝗲 𝘀𝗶𝗴𝗻𝗮𝗹𝘀 𝗵𝗶𝗴𝗵𝗲𝗿 𝗿𝗼𝗹𝗹𝗼𝘃𝗲𝗿 𝗰𝘆𝗰𝗹𝗶𝗰𝗮𝗹𝗶𝘁𝘆, especially in institutional term-loan markets (e.g., CLO-funded TLBs with bullet maturities).
• Banks anticipate that a nonbank shock will 𝘀𝗵𝗶𝗳𝘁 𝗳𝘂𝗻𝗱𝗶𝗻𝗴 𝗼𝗻𝘁𝗼 𝘁𝗵𝗲𝗶𝗿 𝗯𝗮𝗹𝗮𝗻𝗰𝗲 𝘀𝗵𝗲𝗲𝘁𝘀 via credit line drawdowns, creating capital and funding pressure precisely when they want optionality.
• The result is simple: 𝗹𝗲𝘀𝘀 𝗮𝗻𝗱 𝗺𝗼𝗿𝗲 𝗲𝘅𝗽𝗲𝗻𝘀𝗶𝘃𝗲 𝗹𝗶𝗾𝘂𝗶𝗱𝗶𝘁𝘆 𝗶𝗻𝘀𝘂𝗿𝗮𝗻𝗰𝗲 for firms that look most likely to “run” to the bank in stress.

In the data, moving from bank-reliant to nonbank-reliant borrowers is associated with materially smaller credit line shares and meaningfully higher drawn and undrawn spreads.

The uncomfortable implication: the safest-looking capital structure in good times may be the one that loses 𝗯𝗼𝘁𝗵 refinancing and liquidity capacity in bad times.

Here is my question
• to those who actually underwrite bank loans: Where, in your credit process, 𝗱𝗼 𝘆𝗼𝘂 𝗲𝘅𝗽𝗹𝗶𝗰𝗶𝘁𝗹𝘆 𝗽𝗿𝗶𝗰𝗲 𝘁𝗵𝗶𝘀 “𝗱𝗼𝘂𝗯𝗹𝗲 𝗯𝗮𝗰𝗸𝘀𝘁𝗼𝗽” 𝗿𝗶𝘀𝗸?
• to folks in treasury departments: 𝗗𝗼 𝘆𝗼𝘂 𝗮𝗰𝗰𝗼𝘂𝗻𝘁 𝗳𝗼𝗿 𝘁𝗵𝗶𝘀 𝗶𝗻 𝘆𝗼𝘂𝗿 𝗹𝗶𝗾𝘂𝗶𝗱𝗶𝘁𝘆 𝗺𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁?