Jay Kahn
@jstatistic
Followers
2K
Following
2K
Media
209
Statuses
2K
Economist covering repo, Treasuries, and the structures that shape modern market functioning. Alum @MichiganRoss, @SimonSchool, @Reed_College_. Views my own.
Washington, DC
Joined October 2014
No one really knows how big the U.S. repo market is. Or rather, no one πππ. We combed through the appendixes to a decade of annual reports for over 150 dealers and banks to get the full picture. The total? $12 trillion, almost doubling previous estimates.
12
104
366
John Leahy was a remarkable economist β one of the best in the world. I was honored to call him a friend for many years and had been looking forward to welcoming him to the Chicago FED next month as the director of research. May he rest in peace.
1
8
108
In my view the goal of economics is to help people (often policymakers) make good decisions. Unfortunately, we will never truly know what would have happened if they made a different decision.
Trolling aside, some earnest thoughts on causal inference: I think what annoys me about a certain viewpoint is: in my view, the goal of "science" broadly is to make predictions, and then test those predictions in reality
1
2
9
The problem is that more data doesn't give you causality. You *always* need an economic model. That's not ML. Tbf the mostly harmless crowd created this problem by pretending this wasn't true.
Yeah canβt believe these insane views held by *checks notes* Andrew Gelman https://t.co/IGUNVik6Nr
0
0
4
@momin_rayhan You donβt need LLMs. We made N PORT data available a couple of years ago before they were fully available. https://t.co/snBDlarhHJ
1
1
1
Excited to share my latest paper: βProviding safety in a rush: How did shadow banks respond to a $1 trillion shock?β (with Stefan Gissler and Marco Macchiavelli) Just out in the Journal of Financial Intermediation. We dive into how government money market funds (MMFs) handle
2
7
70
1
1
19
Very happy this paper was just accepted at Journal of International Economics!
Ron Alquist, @DiltsStedman, and I recently released a new version of our paper:Β https://t.co/2fQSmy1Z4H We show the use of the USD as reserves exposes US money markets to foreign countries' net export shocks, providing a role for the Fed as "central banker to the world." 1/14
0
1
16
TLDR: fact check your hot takes on repo with @jstatistic before posting!
Repo volumes have climbed sharply. But some of the posts about it are getting a little carried away! Primary dealer data--one of the few long-running sources--shows activity is high in $, but not above pre-crisis levels when scaled by GDP. h/t @aRishisays
0
1
7
You might want to scale by something else! Here it is scaled by total financial assets in Z1 (*excluding* Treasuries).
1
0
5
Repo volumes have climbed sharply. But some of the posts about it are getting a little carried away! Primary dealer data--one of the few long-running sources--shows activity is high in $, but not above pre-crisis levels when scaled by GDP. h/t @aRishisays
2
4
26
What if I told you that the original central bank had a repo facility, did OMOs, had RTGS, but only had ledger money (no physical currency)? What if I also told you this ledger money was widely used in Europe? Great show on the Bank of Amsterdam!
In case you missed it Will Roberds lays out for @DavidBeckworth how the Bank of Amsterdam was the first modern bank.
3
5
19
Seeing some renewed guessing about who holds the long duration in Treasury futures, the other side of trades like basis trades. No need to guess! We mapped it out in detail in our earlier paper. π
1/𧡠We just put out a new paper putting numbers to a mystery I've been working on since 2020: who is behind the almost $2 trillion increase in long Treasury futures positions? Spoiler alert: it's mostly mutual funds but the cool thing is why. Link here: https://t.co/VgTS8Y0qrr
4
6
28
Highly levered hedge funds play an important role as intermediaries in Treasury markets nowadays, absorbing 37% of issuance, much more important than we thought. Great forensic finance work by @jstatistic and co-authors. and excellent write-up by @RobinWigg
2
15
71
Very happy to share I've been promoted to Principal Economist at the Board!
21
2
240
In a new note, we demonstrate that the massive buildup in hedge fund Treasury positions in the past decade has had substantial effects on the national accounts data. This is due to underreporting to TIC of Cayman Islands Treasury holdings associated with these hedge funds.
0
0
1
"Where does Bloomberg get this stuff??" π
BREAKING: Hedge funds in the Cayman Islands held more Treasuries at end-2024 than US official data show, with their ownership likely to be $1.4Β trillion higher than shown, per Bloomberg
2
2
21
πΊπΈ Fed note showing that foreign holdings of US Treasuries by Hedge Funds in the Cayman Islands are massively under-reported, largely driven by the basis trade. This means that TIC data are wrong, but also national accounts overstating households holdings as a result.
In a new note, we demonstrate that the massive buildup in hedge fund Treasury positions in the past decade has had substantial effects on the national accounts data. This is due to underreporting to TIC of Cayman Islands Treasury holdings associated with these hedge funds.
4
36
114
This is an additional problem for Z1! Lending for the trade goes MMFs -> dealers -> HFs. MMF leg of the trade is captured in Z1 through N-MFP. But the hedge fund side that should show up through TIC is missing! So we get a discrepancy between total repo and reverse repo.
0
0
5