Perspective on Risk - Oct. 19, 2023
Treasury Market Musings; Follow-up On Japan's Zengin-Net; Asset Allocation; (Lots of) Other Reading
Treasury Market Musings
NY Fed Looks At 2023 Market Liquidity
Former colleague Michael Fleming of the NY Fed has written How Has Treasury Market Liquidity Evolved in 2023? for their Liberty Street blog. It is a follow up to a similar 2020 post.
We find that liquidity worsened abruptly in March 2023 after the failures of Silicon Valley Bank and Signature Bank, but then quickly improved to levels close to those of the preceding year. As in 2022, liquidity in 2023 continues to closely track the level that would be expected by the path of interest rate volatility.
… order book depth, measured as the average quantity of securities available for sale or purchase at the best bid and offer prices … points to relatively poor liquidity in March 2023... Depth in the five-year note was at levels commensurate with those of March 2020, whereas depth in the two-year note was appreciably lower—and depth in the ten-year note appreciably higher—than the levels of March 2020. Within about a month, depth for all three notes was back to levels similar to those of the preceding year.
The association between liquidity and volatility in 2023 has been consistent with the past …
El-Erian Is Less Sanguin Going Forward
El-Erian writes in the FT that the US bond market is losing its strategic footing. When Mo talks, folks at central banks tend to listen.
Last week’s unusual turbulence in US Treasuries points to a deeper issue than just the latest reading of the runes on inflation and the interest rate intentions of the Federal Reserve. The US bond market is losing its strategic footing, whether in economics, policy, or technical aspects.
… my primary concern lies elsewhere: the most influential segment of the world’s financial markets is losing its longer-term strategic anchors and is at risk of losing its short-term stabiliser ones as well.
The recent consensus on economic growth for the world’s largest economy has been erratic … Federal Reserve policy is further shrouded in uncertainty, with … the absence of an effective framework for monetary policy. … All of this occurs in a context of substantial fiscal deficits ...
This uncertainty also extends to longer-term supply and demand dynamics. Despite rising interest rates, there is genuine doubt about who will readily absorb the additional supply of government debt associated with high deficits.
Follow-up On Japan's Payments System Zengin-Net
Iza Kaminska has been following the payment system outage that I mentioned in the last Perspective.
Finally! We have more info on what went wrong with Japan's payments system Zengin-Net last week.
As we predicted in Politico's Morning Central Banker newsletter last week, it was down to interoperability failures between Japan's "More Time System" (which operates outside of working hours) and the "Core Time system". Due, of course, to updates.
Both systems operate in parallel during normal times, and in the event of a failure, only one operates.
According to a press-conference just given by Zengin-Net, over Oct 7-9, 14 institutions were supposed to update their relay computer to new models to better interoperate between the two systems.
But the update failed, impacting the ability of the relay computer to send messages between the Core system and the banks.
In addition, JP Morgan bank (the most important intraday liquidity hub in the world) also supposedly launched its "More Time System" relay computer with a new model at the same time. But its core system remained on the same old model. So, based on the failure of the core system, it didn't connect to the More Time system.
This (and this is my conjecture) would probably have led to the accumulation of a large number of unsettled intraday transactions over a holiday period.
And this (again conjecture) may have required additional temporary liquidity.
The system has been resolved apparently using "provisional measures" - which doesn't sound like a robust solution at all.
Further investigations are coming. But, as we noted in our newsletter, it's noteworthy that this sort of systemic "gridlock" risk was flagged by then deputy governor Masayoshi Amamiya on the case for and against a digital yen in 2019, with reference to the "More time System" introduction in 2018 - which essentially extended the deferred net settlement system for retail payments (which settles out via the realt-time BoJ net) to a 24 hour basis.
As he noted at the time:
“While the efficient use of liquidity provided by DNS is an advantage, it accumulates unsettled positions up to the designated time, which means that settlement is not final until this net position is settled. In this way, DNS harbors a systemic risk; if even one of the participating financial institutions fails to meet its obligations, it could potentially start a chain reaction affecting all other institutes.”
What we need to know is if these "system updates" are in anyway connected to BoJ plans announced in recently to develop ‘a test environment for the CBDC system’.
Today is the day the Digital euro enters its preparation phase, so it's an important point to find out.
For more on the Blind Spot's take on how this all connects to intraday liquidity, see my last newsletter: https://t.co/OQNNpOevfZ
For more background: How RTGS inadvertently killed system liquidity (FT)
Asset Allocation
Maybe You Only Need A Global Equity Index & A Local Bond
Jan Loeys, JPMorgan’s veteran asset allocation guru, says in a recent client note that this complexity is both pointless and counterproductive. Pointless, because investors need only two assets: a global equity one and a local bond one, with the relative amounts driven by their ability to withstand short-term drawdowns and return needs.
Counterproductive, because the clarity and simplicity that is afforded by sticking just to two assets makes it easier to judge risk on them, and much cheaper in terms of manager and transaction fees, as well as governance time.1
I don’t have access to the JPMC piece, but the FT has substantial excerpts in The art of keeping it simple, by JPMorgan’s Jan Loeys.
I think his argument simplifies to:
Over the long run equities outperform
You hold bonds only because you can’t tolerate drawdowns
Any structurally superior assets or strategies will be arbitraged away before the average investor can access the returns.
HF Investors Want More For Their Money
Investors and Hedge Funds Are Quarreling: Should the Risk-Free Rate Heighten Their Hurdle? (Institutional Investor)
The risk-free rate of return — what an investor could earn on a super-safe investment like Treasuries — is about 4.5 percent and investors want hedge funds to adjust their hurdle rates, the performance threshold they have to cross in order to charge a performance fee.
So what this tells me is 1) investors will not be allocating as much to lower-risk, lower-return strategies when they can get roughly the same return risk free, 2) if this comment is true, and not just negotiation rhetoric, investors are more focused on absolute returns than the diversification benefit and marginal return contribution of their HF investments.
Other Reading
Deposit Convexity
Deposit beta has been a big discussion point following the run on the regional banks. But economists at the FRB-Dallas take it to the next step with Deposit Convexity, Monetary Policy, and Financial Stability.
Banks and researchers conventionally model the response of deposit interest rates to market interest rates as constant, implying that deposits have nearly constant duration. Contrary to this standard assumption, we show empirically that the “beta” of deposit rates to market rates increases as market rates rise, causing the duration of deposits to fall. The amount of duration risk delivered to bank balance sheets via this channel from March 2022 to September 2023 is comparable in magnitude to the amount of duration risk absorbed by each of the several large-scale asset purchase programs the Federal Reserve has undertaken since 2008. Dynamic betas present a significant challenge to bank portfolio hedgers by introducing large and dynamic risks that are difficult to model and impractical to replicate on the asset side of the balance sheet. As a result, deposit convexity amplifies monetary policy transmission and increases financial fragility, mechanisms that recent banking stresses have highlighted.
Our analysis shows that, unless properly accounted for in bank deposit models, dynamic betas will increase run risk when interest rates are low. Importantly, to avoid overhedging duration risk, banks would need to consider these dynamics ex-ante.
So, the speed with which the Fed raised rates did substantially contribute to duration mismatches.
I really liked this paper, and anyone involved in liability modeling or the interaction of monet6ary policy and financial stability ought to internalize this.
The Quants Have Come For The Bond Market
Analyzing 563 trillion possible models, we find that the majority of tradable factors designed to price bond markets are unlikely sources of priced risk, and only one novel tradable bond factor, capturing the bond post-earnings announcement drift, should be included in the stochastic discount factor (SDF) with very high probability. Nevertheless, the SDF is dense in the space of observable factors, with both nontradable and equity-based ones being salient for pricing corporate bonds, and a Bayesian model averaging–SDF explains corporate risk premia better than all existing models, both in- and out-of-sample, and captures business cycle and market crash risks.
Real-time Bond Prices
Bloomberg Says It Is Using Machine Learning to Deliver Near Real-Time Bond Prices (Institutional Investor)
Bloomberg announced on Tuesday that it was launching the Intraday Bloomberg Valuation Service (IBVAL) Front Office, which uses machine learning to analyze “billions” of market data points to predict the prices of approximately 30,000 U.S. high-yield and investment-grade corporate securities that are eligible for TRACE (Trade Reporting and Compliance Engine). The securities are priced as often as every 15 seconds, far more frequently than the 2.5 million bonds and loans tracked by Bloomberg Valuation Service (BVAL) and which are priced at the end of each day based on real-time market observations or, for less liquid securities, comparable relative values.
CLNs Can Qualify As Credit-Risk Transfer
Fed Publishes Letter to Morgan Stanley on Credit-Linked Notes (Bank Reg Blog)
Last month the Federal Reserve Board issued FAQs addressing the treatment of credit-linked notes under its capital rules.
As I mentioned on Twitter (X?) a couple weeks ago, and as Mayer Brown pointed out in their memo yesterday, these FAQs were released the day before the Federal Reserve Board approved a request by Morgan Stanley to treat CLNs as synthetic securitizations under the capital rule.
Securities Lending
SEC adopts rule enhancing securities lending transparency (Jim Hamilton’s World of Securities Regulation)
The SEC has adopted a final rule intended to increase transparency in the reporting of securities loans. Those covered by new rule 10c-1a will be required to provide information about securities loans to FINRA by the end of the day the loan is made; the information will then be made publicly available on the next business day. This rulemaking fulfills a Dodd-Frank Act mandate to increase the transparency of information available to brokers, dealers, and investors.
Economic Effects of Global Warming
Hotter summer days heat up Texans but chill the state economy (Dallas Fed)
Analyses drawing on data from 2000–22 indicate Texas is especially vulnerable to hotter summers. For every 1-degree increase in average summer temperature, Texas annual nominal GDP growth slows 0.4 percentage points.
With this year's summer temperatures 2.5 degrees above the post-2000 average, estimates for Texas suggest, all else equal, the summer heat could have reduced annual nominal GDP growth by 1 percentage point for 2023, or about $24 billion. Other calculations suggest a somewhat lesser impact of nearly $10 billion in real (inflation-adjusted) GDP, about 0.5 percent of annual output.
The impact of an increase in summer temperatures on Texas GDP growth is twice as pronounced as the change in the rest of the U.S. because summers are generally hotter relative to the rest of the country. At the same time, the effect of rising summer temperatures on job growth is more subtle, though the effects vary widely across sectors. As climate change’s effects intensify over the next decade, heat waves will become more commonplace and severe, and Texans will need to adapt.
Someone Doesn’t Like Derivatives (or Math)
