What’s up with Fed daylight overdrafts data?

I’m working on a bigger story to do with US dollar intraday liquidity, but until that’s done I wanted to quickly flag something specific to paid subscribers as a small preview.

It’s the Daylight Overdrafts and Fees page of the Federal Reserve website:

And specifically this bit:

What’s important about that?

Well, as the Wayback Machine can testify, the last time this chart was updated was on February 4, 2020. But given that the data is supposed to be updated quarterly, that’s a little bit weird.

As one of the few people in the world who tracks daylight overdrafts somewhat compulsively, I had clocked that the data was failing to update a while back. I didn’t think anything of it until I was on a panel with some money-market academic types who confirmed they too had spotted the anomaly, and were equally perplexed by it.

I tried to get to the bottom of the situation in December 2021 when researching this column about the stigma associated with the facility since 2008. Needless to say, nobody at the Fed ever answered my questions.

Then, lo and behold, when I finally took to writing about a different aspect of intraday liquidity last week I discovered the data had been mysteriously updated on March 22. Clearly the series was not discontinued after all.

Why all the mystery? It’s hard to say.

I suspect it may be related to the subtle growth in the facility’s use as the Covid crisis unfolded in March 2020.

Breaking into the granular data, it’s clear the most stressful two-week period occurred around March 11:

This ties in with my grander theory that the more continuous and global an RTGS system gets, the bigger its appetite becomes for a daily swingable float that’s funded by third parties or the Fed. Without it, payment system risk beckons.

For the most part, however, the excess reserves generated by QE since 2008 have been fulfilling that daily float role. In doing so they have prevented an absolutely huge expansion in the use of intraday overdraft facilities at the Fed (and other central banks).

This, however, also means that the de facto daily minimum float requirement for managing payment system risk is now analogous to the lowest comfortable level of reserves that the system can cope with more generally before freaking out.

As the Fed begins to tighten and remove excess reserves from the system, overdraft data should be watched carefully. (Let’s hope it remains regularly updated.)

My hunch is that if any pressure is detected the system will soon be forced to develop a better and smarter way of funding intraday imbalances associated with payment liquidity risk than just relying on excess reserves.

In time, I suspect, this could involve the creation of an actively traded intraday funding market. Some of the systems the crypto markets have designed to cope with similar pressures might provide an important template for this market’s design.

I’m specifically thinking of one particular product. But more on that later…