Security Lending Times recently published about Hanweck's Trading Indicators. Hanweck's Borrow Intensity Indicators (BII), launched in March last year, and the most recent offering, Issuer Volatility Risk (IVR). BII generates synthetic lending rates on a real-time basis for the optionable universe of listed equities (OPRA coverage), describing term rates with maturities from roughly one month to one year. BII originates in real-time implied borrow data that is part of the core Hanweck Options Analytics.
An additional area of research and development for Hanweck has been debt-equity models entirely based on market data (for example, no dependency upon reported financial statements). By not requiring less frequently reported corporate information, the model can react with the speed of financial markets.
IVR has evolved from deployed analytics that Hanweck has provided on a customized basis since 2007, and intraday updating dataset. Both these trading indicators extract information from entire option chains in order to deliver signal compressed into more compact indicators. In the case of IVR, we condense information across strikes and time to generate either an implied probability of default based or an implied (synthetic) credit spread by assuming a fixed recovery rate.
The motivations in using debt-equity type models for cross-asset class hedging and arbitrage are straightforward. For example, a high-yield bond trader may wish to hedge credit risk with equities or equity derivatives. But why would anyone care in securities lending?
We would argue that by relating information in the more liquid and transparent exchange traded markets to less transparent markets, it provides another channel of information to equities collateral management areas to more rapidly anticipate greater intrinsic value in securities such as fixed income exchange-traded funds (ETFs).
In this article, we will apply both IVR and BII measures to the iShares iBoxx High Yield Corporate Bond ETF (HYG), looking at equity collateral value through Borrow Intensity, and relative value to credit through IVR. Both speak to liquidity and flows that impact execution, trading, and investment in this ETF, of direct concern to market participants with shorter-term focus. Also, longer-term investors can find these metrics valuable, for example, when evidence arises that short term market dynamics may be causing the ETF to overshoot on either rich or cheap directions, departing from fundamentals and creating a better entry opportunity.
Cross-asset class ETFs present a special challenge to collateral traders in terms of predicting liquidity and borrowing costs since the constituents are not equities. But if a securities lending desk notices, for example, the HYG ETF overshooting cheaper than the IVR levels, this is another indicator of trading momentum that will create shorting and increasing need for locates.
Issuer volatility risk, analytic approach
Trading indicator impact
• Short term: informs estimates of market impact and liquidity
• Long term: rich/cheap view across the capital structure (Credit ETF) versus its equities counterpart
Momentum and trends in valuation and implied term rates inform overnight rates
IVR derived implied probability of default in real-time from the equity derivatives market. It is in some respects similar to a ‘single-name’ VIX in the manner it reacts to a wide range of strikes and skews, but the measure is particularly sensitive to the downside or tail risk.
IVR is generated for a single name: the single equity underlying an option chain. Aggregate measures can then be constructed as shown below for HYG, either based on the weights of an ETF or index, or any arbitrary basket.
• Synthetic credit spreads from equity options
• Based upon a constant elasticity of variance (CEV) model
• Transformed with predictive analytic processes to separate signal from noise
Synthetic credit replication with equity options
Figure 1 shows in the upper panel a synthetic replication of HGY entirely from equity and equity options data (no corporate bond information). In this example, the synthetic is generated in terms of credit spread (IVR_spread) shown in a reversed Y-axis so that price and yield spread can move in the same direction. The lower panel shows the historical closing prices of the HYG ETF, which comprises over 950 high-yield bonds of nearly 300 unique issuers.
Process of credit replication with IVR:
• Map corporate bond to related equity
• For equities with listed options, calculate IVR on the option chain
• Construct a weighted aggregate of IVRs on mapped entities
Synthetic showed distress prior to the actual ETF in Q4 2018
The IVR of HYG series has very high correspondence with the actual ETF, and in Q4 2018 it appeared that the equity derivative market moved first both in the process of rapid spread widening (lower HYG prices), and then recovery with spread narrowing (higher HYG prices). IVR moved ahead of HYG, with the major move down (spread widening) on 29 October last year to 437bp (circled in blue), and the second leg down on 24 December last year to 530bp. HYG didn’t see the first corresponding move lower until reaching 82.71 on 23 November last year (lagging by weeks), and then the second move down to the December low of 79.63 on 24 December last year (entire area of changes circled in red). Both series recovered rapidly after Christmas of 2018, with the synthetic series moving up very sharply to reduced risk level.
The fact that the equity derivative series moved more quickly can partly be explained simply by the fact that it has more transparent and liquid underpinnings—the exchange-traded listed equity options market. On the other hand, the markets for the bonds within HYG are opaque, trade less frequently, and tend to be model priced between trades based upon characteristics such as sector, rating, and duration.
Those trying to seek bids for the bonds may notice that markets have widened considerably, but otherwise, information filters back out to the market relatively slowly compared to exchange-traded products. This will only get worse if recent proposals such as a 48-hour delay on block bond trades are brought into effect. These are all motivations to see an alternative market view of HYG and other credit-based ETFs.
BII on HYG
Figure 2 shows the 45-day and 180-day BII on HYG, where Borrow Intensity is expressed in the format of a rebate rate. In this time period, the 45-day term rate reached a negative (hard-to-borrow) level of -1.5 percent with the 180-day term at the same time of roughly 0 percent. This was wider and more rapidly changing spread (45 day to 180 day) than usual for HYG (see chart area circled in blue), which tends to be a leading indicator for market distortions arising from some combination of tightness in the collateral market and a higher level of put buying. The pattern around the turn of the year was somewhat erratic, but the BII synthetic rates appear to show evidence of shorting by 15 November 2018, which dissipates by roughly a month later on 18 December. Combined with the IVR information also showing distress from the equity options market, there was evidence of signals anticipating near-term distress in HYG. In fact, HYG declined approximately 2.6 percent over the next month (from 15 November to 18 December), and 4.7 percent in total by 24 December last year before reversing after the holiday. Notably, by this time the BII picture had turned completely upside down with 45-day rates above 180-day rates.
The ETF didn’t command a greater borrow premium again until mid-January as HYG bounced strongly higher almost completely retracing December losses by 11 January 2019, when the 45-day rate went back below the 180-day, but both stayed more synchronized in this next episode of maintaining a lending premium.
Value of linking information across markets
We focused here on using Hanweck IVR and BII to increase the understanding of the flows and valuation of the high yield bond ETF HYG. Fixed Income ETFs are increasingly important as liquidity expands to meet institutional investment needs. The bond market does not have as efficient a market structure as equities in either transparency or execution, and these ETFs serve a valuable function. At the same time, there is a tension in creating liquid ETFs in the equity market that holds comparatively illiquid bonds.
This creates both challenge and opportunity for equity investors, market makers who participate in the creation/redemption process, and for securities lending areas trying to maximize value in managing ETF collateral.
It’s valuable to connect information across the derivatives and cash markets, and even across asset classes. The analytics considered here can transform information hidden in the equity options markets—where there can be market leading information—into data that informs both equity strategies and securities lending.