@ByrneHobart
@turtlesnapz1
Also: could be that he drops in, fixes a bunch of stuff, things work, word spreads, marginal effect of fixing stuff elsewhere > maintenance ==> switch, w. higher comp
New post: “Addressing overseas USD funding conditions & the Fed as global collateralized LoLR”,
discusses (1) optimizing CB swap line design for non-bank demands, (2) foreign CBs acting as localized primary dealers, softening international spillovers.
First step is indeed to understand that the
a) initial Japanese purchase of USD bonds and
b) the JPY-denom. reinvestment from swap counterparties due to a)’s FX hedges
are mechanically linked.
The one that piqued my interest: the large foreign buying of zero yielding JGBs. It more or less matched Japanese buying of foreign bonds over the last ten years.
I am talking my book, but I don't think you can really understand global finance unless you understand this trade
Enjoyed the recent BBG article on multi-manager risks, along with
@FundamentEdge
’s expanded take on it .
Raising these points and being responsive to their implications is good.
2/
Since there are many moving parts in setting up such a term premia framework, an extra post is published alongside, to slowly walk through the (not always intuitive) overlap of FX and FI markets.
do note that while this approach doesn't always capture every finest detail, I'm all butcertain it gets the overall picture right. And for the blue, orange and green stretches I know well, the local relations look eerily spot on.
27/27
Chapters 2 & 3 of the cooperation with
@Brad_Setser
. Focus is on the transition from Taiwan’s central bank to the life insurance sector recycling the country’s large trade surpluses & lifers’ management of overseas FI books.
franchise in half a decade?
We're nearing the end of year 2 of the ‘grand multi-manager era’ and returns are clearly pointing to oversupply. Capital-weighted MM net alpha above RF through October is about 2%, which falls into negative territory when excluding the five
11/
The even more crucial next step is to realize that the mindset that led to these purchases (esp. post 2015) stemmed from an inherently flawed understanding of fixed income markets & FX hedging.
From 4 yrs ago:
grasp how varying one or more variables changes risks in the system.
Reasons are kinda obvious:
The data situation is patchy. The desire for a fuller/higher frequency handle on most variables is perennial and always leaves uncertainty that is >> vs stress tests of single
7/
MMs equity," which Mr. KG deemed possible in Singapore. Put differently, it kind of matters to have an idea of whether the odds p.a. of a major incident are of order 1 in 5, 5^2, or 5^3, and my sense is that far too few G/LPs have given this enough thought, with a solid
6/
Exposures: $1.3tn equity GMV, 2.4tn total cash GMV, 5tn total gross notional exposures (incl. linear derivatives)
Volume footprint: In equities, largest firms each are party to 2-5% of ADV on most exchanges; sector as a whole is involved in around 90bn of trades globally
17/
It's something else to have modelling capabilities-- however imperfect--to play out industry-wide degrossing events from various starting points of:
· industry AuM,
· leverage,
· interconnectedness/position overlaps,
· market impact & price reversal,
4/
largest names.
There is no disputing the long-term track of the handful of firms that have done it for decades, and it has taken unreasonably long for the in most respects superior platform model to gain widespread traction.
12/
The other half is the depth of the alpha pool. Almost surely, E/DCM activity, retail particp., energy opps. etc. of ’20-‘22 won’t carry on at comparable levels. And the alpha trajectory from ‘15-’19 --before the capital influx-- flattened noticeably, even for Tier 1 firms.
Yesterday’s, with hindsight well-timed, post discusses why this makes sense and further steps to consider. So, what did the Fed do, does it matter and where/how does it differ from yday’s proposal.
If anything, however, the piece also underscores the handwaving character most discussion on crowding, concentration and interconnectedness takes, a condition ill-suited to the sectors' gravity.
It's one thing to qualitatively point to some fuzzy, remote crowding risk.
3/
I am joining forces with a market friend, the blogger Concentrated Ambiguity (
@stwill1
), for a series of posts that will do a deep dive in the balance sheet of Taiwan's life insurance sector and the balance sheet of the Central Bank of the Republic of China (Taiwan). Stay tuned
golden as possible, can fund-level de-risking and return of capital be expected if it leads to poaching of top PMs while undermining longer-term firm viability by failing to sift through enough new PMs of this year's vintage, in search of the 3 to 5 stars who will carry the
10/
Too many who should have known better got wrong-footed on this, and much of the hundreds of $ bn flowing from Japan into USD fixed-rate debt with short-term rolling FX hedges from ’15-’21 should probably not have happened.
If meant as trick question, then just one : the three-letter palindrome.
If literal, by rough size: Cathay UB, Fubon CB, CTBC, E.Sun, Taishin on the local side. HSBC, SC, Calyon, ANZ, JPM on the global end.
· risk mgmt. cultures
· counterparty flexibility
Reaching a close-to-reality rendering of this is non-trivial, but fiddling around with crude sketches is a worthwhile exercise on its own and the only way to get a feel for what happens next after a "10, 15 or 20% joint hit to
5/
This is one of those things that is either obvious, or it requires half an hour of carefully walking through each transaction and cash flow.
In short: HTM investors should think in “term premia” rather than “yield curve shapes”, and a steep yield curve ≠ arbitrage always.
The BBG article, PB reports and a few extra figures give a decent tear sheet of multi-strat, multi-managers:
AuM: ~$350bn
Employees: 16.000, with ½ in investment roles
16/
institutions, lighter on interaction effects.
"Now what?", which holds on many levels and is not very appealing when the answer is often just "less of it". E.g., is a long-term LP going to trim their MLP/Citadel allocation based on such thinking & risk not getting back in?
8/
Are those pouring funds into (sub-)tier 2 platforms realistically expected to be dissuaded? Are bank counterparties expected to toughen terms or retrench, sending the biggest fee-payers to laxer rivals? And from a fund's standpoint, after making sure you're as operationally
9/
Multi-managers in blue are unmistakably the by far densest region, with all of the graph's top 5 ‘influential’ firms by the node size measure located in it.
This is getting pretty long, so let’s save the magnified looks for another day. As preface & caveat to it,
26/
Two tweaks:
Using a very sophisticated technique--so called "manual inspection"--coloured landmarks are placed at the centres of the various clusters, with Gaussian kernels dropped on each, allowing node colours to be interpolated.
Further, node size is scaled by the
23/
But now, with capital mobilization for new entrants still plenty and billions in committed but so far uncalled funds at established players likely tempering the upside, greater clarity around the downside is essential.
13/
The other half is the depth of the alpha pool. Almost surely, E/DCM activity, retail particp., energy opps. etc. of ’20-‘22 won’t carry on at comparable levels. And the alpha trajectory from ‘15-’19 --before the capital influx-- flattened noticeably, even for Tier 1 firms.
Btw the whole pod/SM discussion can pretty much be reduced to "Capital Returns". What do you think will happen as Tier 2/3/4 MMs keep accumulating capital & talent to scale up? From crowding alone, returns will deteriorate; it isn't rocket science. Nothing good lasts forever.
totality of incoming flows, thus representing a mix of firm size and interconnectedness.
This produces the graph from the opening, displaying the 850 or so core firms that make markets move day by day. The colour assignment is the following:
24/
· infusing fresh information into prices and
· greasing its wheels through mid-frequency inventory management,
is still underappreciated today. Certainly in the broader discourse, and with a couple exceptions likely on here.
15/
The TBAC presentation on Treasury futures has deservedly received a lot of attention and contains great background & aggregate figures on the basis issue. 1/
each day, or around 20-25% of lit ADV in mid-cap upwards. Corresponding figures for cash fixed income is around 120bn, and for notionals of linear derivatives, 700bn.
18/
To simulate extra transitions, consider the following thought experiment: Let each firm (in turn) hypothetically close down and each front-office individual probabilistically jumps to other firms, proportional to the presence of the latter in the direct network of the former.
21/
A dated chart giving a sense of the numbers involved. FX hedges provided by reserve managers are labelled “FX swapped debt” and were around USD 1trn in the 2010s.
One last thing I learned from Concentrated Ambiguity (the blog of
@stwill1
) -- a lot of the world's hedges are supplied out of the world's central bank reserves.
FX hedged yen reserves, for example, generate dollar funding for Japan's institutional investors ...
Ahead of the concluding sixth chapter in the Taiwan project, a quick look back at the story so far, especially chapter V. (
@Brad_Setser
's summary here last week).
Taiwan’s central bank almost certainly has an undisclosed forward book over well over $100b – and has been intervening far more heavily after 2009 than would be inferred from its disclosed reserves.
This is joint work with
@stwill1
professionals.
Core idea: firms that hire from another are similar in some respects, and this similarity has downstream effects on strategies, positions and ultimately P&L. The issue is that, even though fast paced, actual job switches are insufficient to map this.
20/
***
With the tirade out of the way, on to the more constructive, fun? part of justifying the above.
Despite the buzz the last 1.5 yrs, it’s, imo, still a good bet that relative to when the history books will be written, the centrality of platforms to liquid markets due to
14/
(While the details are in the paper,
@TheStalwart
&
@tracyalloway
were kind enough to host my co-author
@Brad_Setser
last fall. The resulting podcast remains the perhaps gentlest intro to the overall topic.)
1)This plus
2)Extend QE to allow for direct reserve manager sales of USTs/MBS already in Fed’s custodial accounts to the Desk – no need to clog dealers even momentarily.
3)Encourage foreign CBs to intermediate global repo funding for USTs/MBS by transferring private sector..
My new idea is a facility that lets large countries with substantial holdings of Treasuries and Agencies repo them for cash. Would reduce demand for swap lines from countries with lots of reserves, and would reduce their need to sell Treasuries into a difficult market
Taiwan’s central bank almost certainly has an undisclosed forward book over well over $100b – and has been intervening far more heavily after 2009 than would be inferred from its disclosed reserves.
This is joint work with
@stwill1
@nosunkcosts
@Brad_Setser
Difficulty with JPB is assessing their hedge ratio for foreign bonds held indirectly via trusts. On balance sheet foreign currency bonds are hedged ~50%, while trusts likely fully hedged.
4. All combined, the following picture emerges. It will provide the basis to slice the nationality based numbers into more functional categories in the next post.
@Value_Ideas
@YellowLabLife
@FundamentEdge
Smoothed over a cycle, and assuming that most hires are first-time pod PMs and not laterals, this is a reasonable approximation historically.
While all true, actually *seeing* where they fit into the bigger picture, recognizing connections and relative proportions, is oftentimes superior to isolated facts. The most macroscopic means I'm aware of looking at this is through the flow of front office buyside
19/
Interesting survey & results. Putting aside all contractual minutiae, it indirectly asks for takes on
1. the street’s/MMs’ risk aversion/utility and its
2. short-term risk-bearing capacity in relation to the trade’s size, i.e. the extent of capital constraints.
Totally fair coin.
Heads you win $50mm, tails nothing happens.
Coin gets flipped in an hour. No credit risk.
You can shop the proposition to Jane, SIG, Citadel or anyone else for that matter.
What price does it trade for?
but are (on the bond side) ultimately far eclipsed by two depository institutions: Norinchukin and Japan Post Bank, which together hold foreign bonds of close to $1 tn. As the single biggest buyers, the effect of their purchases (and FX hedges!) on markets is hard to overstate.
freeing banks to focus on domestic issues while limiting spillovers foreign liquidations would cause. From the Fed’s perspective, the transaction is just another repo, this time not with a domestic bank but a foreign CB, equally collateralized (& margined) by Treasuries.
This is a follow up to what began as joint project with
@Brad_Setser
last October.
Some housekeeping and looking back is in order. This includes publication of the project as a proper paper and how the central bank of Taiwan, after initial deflections, eventually did confirm our
I have spent almost twenty years of my life looking at China's balance of payments data in one way or another.
Right now neither the current account numbers nor the financial account numbers quite fit together.
1/x
Since there are many moving parts in setting up such a term premia framework, an extra post is published alongside, to slowly walk through the (not always intuitive) overlap of FX and FI markets.
@KrisAbdelmessih
@W98AB
Quoting a fmr. colleague: “What Blackrock does [in equities] in a quarter, the larger platforms do individually in a busy two weeks.”
Re CB currency swap lines, it’s broadly agreed that transmission is much tougher today, as large shares of the ultimate demand is from the buyside and not banks themselves. See for instance here
How will it be used, what will its effects be?
Fed’s statement contains little color on *how* it would like overseas CBs to utilize the facility + the current design misses some features I’d thought would be included – so tough to make precise predictions. Statement
Fed’s Desk and not rely on bank intermediation, even at a later time. This would further reduce pressure on dealers in that knowing that FX managers will (for some time) sell directly to Fed would allow focus to be placed elsewhere. Pricing of direct sales could be somewhat of an
(2) FX-unhedged foreign demand. This is largely a story about foreign fixed income investors taking on a rather uncharacteristic risk: FX exposure. USD appreciation has helped these institutions immensely and invited further inflows. Closing of such positions (or hedging
One interesting follow-up, probing those two factors more directly, would be to swap variables:
How large would $ X mm have to be for it to trade at a discount of, say, 10% compared to 0.5X?
A second puzzle is the reappearance of regular late-day depreciations of the TWD. It has long been clear (see prior reporting by
@hwang61
,
@ariespoon
,
@samsonellis
& Argin Chang) that market forces alone would not give rise to such a pattern, hinting at active CBC involvement