Skip to content
All library documents

Why Bank Funding Uses Overnight Rates Instead of Continuous Funding

Article Quant Q&A · Author: Phil H

Summary

The document considers whether banks might replace overnight funding with continuously or intraday indexed funding as trading activity grows. It cites a historical triparty repo example in which daily unwind mechanics created a temporary funding gap that clearing banks covered, suggesting that funding needs can arise within the day. The answer, however, argues that continuous funding would offer unclear business benefits and require changes to banking systems and accounting.

The explanation gives a structural reason for the overnight convention: intraday payments and unsettled trades can leave banks temporarily out of balance, while end-of-day requirements make overnight borrowing useful for meeting obligations. The answer says that other market rates follow the overnight rate through arbitrage. These are concise, historically framed claims rather than a detailed analysis of modern settlement systems, regulatory rules, or the feasibility of intraday funding products.

Key ideas

  • The document asks whether continuous or intraday funding could replace overnight bank funding conventions.
  • Daily repo unwind mechanics can create intraday funding needs even when trades are not yet settled.
  • Overnight borrowing helps banks manage temporary intraday imbalances and end-of-day balance requirements.
  • The answer cites infrastructure costs and uncertain customer demand as barriers to continuous funding.
  • It argues that arbitrage links other market rates to the overnight convention.

Tags

Full text
# Could banks move to continuous (rather than overnight) funding?


# Could banks move to continuous (rather than overnight) funding?












For a long time, the dominant tenors for money market and FX instruments were 6 months and 3 months, and banks slowly moved to commercial trades at those tenors but funding overnight. If this is a step in the direction of increasingly short-term funding as trade frequencies and volumes increase, could banks move to continuous funding? That is, the book would be continuously balanced, excess lent continuously and funding performed continuously, with a continuously indexed swap instead of an Overnight Indexed Swap (OIS)?

One can imagine an hourly or minute-to-minute fixing in FX already.

[Update Sep 2012]: Via Deus Ex Macchiato, this from the FSOC annual report:

> Currently, triparty repo trades unwind every day, meaning that the clearing bank returns cash to the lender’s account and returns collateral to the borrower’s account. Trades are not settled until several hours later. For several hours each afternoon, dealers require funding of their entire triparty repo book that lenders do not provide. This $1.7 trillion funding need is provided by two clearing banks. This is a potentially unstable situation.

This suggests that accounts are already balanced hourly or less.

## Answer by André Bittencourt (score 1)

https://quant.stackexchange.com/a/71694

Probably not.

The first question, why? Why would banks do that? Would you this attract more clients and business? Or would it be irrelevant?

How? Continuous (or at least intraday funding) would require a big update in banking infrastructure. Also, how would be the 1/100th of 1/10th of cent would be registered in the books? Ok, both are not big problems, but should be solved, anyway.

Why overnight? Banks pays overnight rate because during the day balance sheets can be not netted. You can buy something and not pay until the end of the day, or even two days later. Banks borrow and lend in the overnight, because they legally can't have "negative" money in the end of the day (which can happen intraday), they might have to pay fines if this happens. It's cheaper to borrow money from banks that have surplus.

By arbitrage argument, the rest of the market follows the overnight rate convention.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.