Evolving analytic approach
03 April 2018
Robert Levy, head of business development of Hanweck, explores securities lending opportunities through the lens of the borrow intensity indicator
Image: Shutterstock
As the securities lending industry collectively focuses on the imperative of having clean, actionable data to complement existing data sources and inform lending opportunities in a more difficult market environment, Hanweck has recently developed and launched the Hanweck Borrow Intensity Indicator. The tool offers a new approach to better understand both market microstructure, and broader trends within the equity financing market. The borrow intensity indicator is derived from real-time exchange-based data, and offers rapid reaction to changing financing market conditions. Intraday updates provide opportunity to spot fast moving markets after events such as earnings reports, news, or corporate actions. This can be useful to borrowers when evaluating the economics of a prospective trade, or an agency lender looking to maximise return for a beneficial owner with the next repricing. Borrow intensity is expressed in a format analogous to lending rebate and can be rapidly incorporated into firms’ valuation frameworks.
Borrow intensity indicators cover the breadth of the US equity universe with listed options, and offer a rich-term structure of term rates ranging from 45 to 360 days. Longer-term historical series reveal the demographics of security lending, highlighting and contrasting rate levels and trends in groups ranging from the hardest to borrow (high intrinsic) to general collateral. This information offers perspective for benchmarking securities lending as an investment product, showing trends in rates of return across a range of intrinsic value classes, and also the number of securities falling into each valuation group. This data can demonstrate recent or longer-term potential to a beneficial owner considering participation in or looking to expand a lending programme, and complements existing market information such as utilisation rates and lending margins.
Broad trends through the lens of borrow intensity
Securities lending-faced headwinds throughout 2017, with collateral utilisation rates declining in the record setting low volatility environment, and persistent bull market conditions that discouraged short activity. Despite poor utilisation, there was some revenue support in the form of higher intrinsic value for the securities that remained hard-to-borrow. What did this look like within Borrow Intensity Indicator statistics?
High intrinsic value groups
Two views are presented in the figure 1 of the segment of the market with 180-day borrow intensity below zero percent (meaningfully hard-to-borrow). In figure 1, the quintile view shows the fifth-tile, with the most extreme negative rates, pulling away from the other tiles to become increasingly hard-to-borrow, with average rates reaching minus 20 percent. The membership count of this quintile however, is declining with more days of smaller counts of 50 or less versus 100 and higher in the earlier part of last year.
The second view in figure 2, with securities bucketed by borrow intensity level, provides additional insight into trends from 2016 to 2018. Much of the decline in the total count of harder-to-borrow securities occurred from the decline in the mildly hard-to-borrow category of borrow intensity ranging from 0 to minus 2.5 percent (segment in blue). This category started at 200 names in January 2016, peaked at 800 names in mid-2016, and then steadily declined into 2018 to go below 200. Meanwhile, the higher intrinsic value categories had lower, but mostly steadier counts throughout the entire period, pointing again to a small core of very hard-to-borrow securities that was critical to maintaining lending revenue.
Easier-to-borrow groups and term structure
Figure 3 displays securities from mid-2015 through 2018 that had six-month borrow intensity greater than zero, including some mildly hard-to-borrow names and much more general collateral (that can have borrow intensity rates exceeding comparable risk-free rates).
This plot also displays a six-month overnight indexed swap (OIS) curve (solid black line) for ready comparison to a term interbank rate. Here, the bottom of the fourth ntile tends to float right above the OIS.
The upwards slope in all ntiles run in sympathy to OIS, but with lower ntiles (For example, increasingly general collateral), there is a steeper rate of increase as the data moves into 2018. This data is useful to help calibrate borrow intensity versus actual lending rates (both risk free and collateralised).
It is not clear why synthetic-term rates display the sharper rate of increase, but this phenomenon has been observed in both large cap single stocks and exchange-traded funds (ETFs), and in part reflects the difference between a pure interbank rate, and the collective balance sheet cost of option market-making participants.
Security-level borrow intensity indicator case studies
Moving from the macro to the micro, intraday updates of the borrow intensity indicators can reveal rising situations either with significant changes automatically detected in new securities that cross the threshold into the hard-to-borrow category, or rapid shifts in securities already well known as hard-to-borrow.
Consider the case of frontier communications (FTR) in figure 4 below. FTR demonstrated high intrinsic value throughout the second half 2017, with 45-day borrow intensity ending the year at roughly -22.5 percent (expressed in the format of rebate rate). FTR experienced a major earnings miss after the close on 27 February this year, and the stock traded down sharply the next day. FTR had experienced similar misses and stock price behavior in the past—for example after earnings on 31 October last year.
The financing reaction in the latter event was far more severe. The 45-day borrow intensity showed a morning opening level of roughly minus 19 percent, with rates progressing rapidly more hard-to-borrow throughout the day with minus 30 percent at noon, and closing the day at minus 36 percent. Even longer-term borrow intensity indicators, such as the 180-day, showed parallel momentum—often a sign of a longer-term dislocation. This proved to be the case with FTR moving into March with 45-day borrow intensity moving below minus 50 percent.
Securities lending data is relatively opaque, and borrow intensity indicators at the intraday frequency can fluctuate, particularly in less liquid names due to the aggregation of data from varying levels of market liquidity throughout the day. To assist in discerning true event signals and new lending regimes in less liquid names, the indicators are paired with confidence series that can corroborate or question the borrow intensity valuation.
Confidence is based on millisecond observations of the liquidity of the underlying option markets, and this data is weighted and aggregated into 20 minute-intraday updates.
The combined information set gauges the strength of new updates, and can also inform recent changes coming through corresponding overnight rates.
This is visible in the data for Cara Therapeutics (CARA) seen in figure 5. The 180-day borrow intensity was fluctuating around zero percent in early May 2017, and the confidence indicator at that time ranged from 13 to 30, where confidence is measured on a scale of 0 to 100, with 100 being most confident.
The CARA stock price showed considerable volatility in the latter part of June 2017 with the market anticipating and then later reacting negatively to disappointing research trial results. During this period, borrow intensity moved from minus 6 percent to minus 26 percent, with confidence gradually stepping up from 21 percent to 40 percent. The increased confidence and high momentum of the borrow intensity indicators suggested significant dislocation. CARA stayed persistently hard-to-borrow throughout the remainder of 2017.
Using borrow intensity to sharpen focus in securities lending decisions
The examples here demonstrate the range of views possible with Hanweck Borrow Intensity Indicator data, from single security to broad demographics of lendable securities. At the core, this data is based upon transparent and neutral data derived from quoted derivative markets. Borrow intensity indicators complement existing securities lending data, such as utilisation and percentage on loan, and borrow intensity convergence or divergence with other market data can signal changing market conditions and new lending regimes.
More broadly, this data is an example of evolving analytic approaches to creating alternative data sets, and it is exciting to apply such techniques to inform the securities lending market.
Borrow intensity indicators cover the breadth of the US equity universe with listed options, and offer a rich-term structure of term rates ranging from 45 to 360 days. Longer-term historical series reveal the demographics of security lending, highlighting and contrasting rate levels and trends in groups ranging from the hardest to borrow (high intrinsic) to general collateral. This information offers perspective for benchmarking securities lending as an investment product, showing trends in rates of return across a range of intrinsic value classes, and also the number of securities falling into each valuation group. This data can demonstrate recent or longer-term potential to a beneficial owner considering participation in or looking to expand a lending programme, and complements existing market information such as utilisation rates and lending margins.
Broad trends through the lens of borrow intensity
Securities lending-faced headwinds throughout 2017, with collateral utilisation rates declining in the record setting low volatility environment, and persistent bull market conditions that discouraged short activity. Despite poor utilisation, there was some revenue support in the form of higher intrinsic value for the securities that remained hard-to-borrow. What did this look like within Borrow Intensity Indicator statistics?
High intrinsic value groups
Two views are presented in the figure 1 of the segment of the market with 180-day borrow intensity below zero percent (meaningfully hard-to-borrow). In figure 1, the quintile view shows the fifth-tile, with the most extreme negative rates, pulling away from the other tiles to become increasingly hard-to-borrow, with average rates reaching minus 20 percent. The membership count of this quintile however, is declining with more days of smaller counts of 50 or less versus 100 and higher in the earlier part of last year.
The second view in figure 2, with securities bucketed by borrow intensity level, provides additional insight into trends from 2016 to 2018. Much of the decline in the total count of harder-to-borrow securities occurred from the decline in the mildly hard-to-borrow category of borrow intensity ranging from 0 to minus 2.5 percent (segment in blue). This category started at 200 names in January 2016, peaked at 800 names in mid-2016, and then steadily declined into 2018 to go below 200. Meanwhile, the higher intrinsic value categories had lower, but mostly steadier counts throughout the entire period, pointing again to a small core of very hard-to-borrow securities that was critical to maintaining lending revenue.
Easier-to-borrow groups and term structure
Figure 3 displays securities from mid-2015 through 2018 that had six-month borrow intensity greater than zero, including some mildly hard-to-borrow names and much more general collateral (that can have borrow intensity rates exceeding comparable risk-free rates).
This plot also displays a six-month overnight indexed swap (OIS) curve (solid black line) for ready comparison to a term interbank rate. Here, the bottom of the fourth ntile tends to float right above the OIS.
The upwards slope in all ntiles run in sympathy to OIS, but with lower ntiles (For example, increasingly general collateral), there is a steeper rate of increase as the data moves into 2018. This data is useful to help calibrate borrow intensity versus actual lending rates (both risk free and collateralised).
It is not clear why synthetic-term rates display the sharper rate of increase, but this phenomenon has been observed in both large cap single stocks and exchange-traded funds (ETFs), and in part reflects the difference between a pure interbank rate, and the collective balance sheet cost of option market-making participants.
Security-level borrow intensity indicator case studies
Moving from the macro to the micro, intraday updates of the borrow intensity indicators can reveal rising situations either with significant changes automatically detected in new securities that cross the threshold into the hard-to-borrow category, or rapid shifts in securities already well known as hard-to-borrow.
Consider the case of frontier communications (FTR) in figure 4 below. FTR demonstrated high intrinsic value throughout the second half 2017, with 45-day borrow intensity ending the year at roughly -22.5 percent (expressed in the format of rebate rate). FTR experienced a major earnings miss after the close on 27 February this year, and the stock traded down sharply the next day. FTR had experienced similar misses and stock price behavior in the past—for example after earnings on 31 October last year.
The financing reaction in the latter event was far more severe. The 45-day borrow intensity showed a morning opening level of roughly minus 19 percent, with rates progressing rapidly more hard-to-borrow throughout the day with minus 30 percent at noon, and closing the day at minus 36 percent. Even longer-term borrow intensity indicators, such as the 180-day, showed parallel momentum—often a sign of a longer-term dislocation. This proved to be the case with FTR moving into March with 45-day borrow intensity moving below minus 50 percent.
Securities lending data is relatively opaque, and borrow intensity indicators at the intraday frequency can fluctuate, particularly in less liquid names due to the aggregation of data from varying levels of market liquidity throughout the day. To assist in discerning true event signals and new lending regimes in less liquid names, the indicators are paired with confidence series that can corroborate or question the borrow intensity valuation.
Confidence is based on millisecond observations of the liquidity of the underlying option markets, and this data is weighted and aggregated into 20 minute-intraday updates.
The combined information set gauges the strength of new updates, and can also inform recent changes coming through corresponding overnight rates.
This is visible in the data for Cara Therapeutics (CARA) seen in figure 5. The 180-day borrow intensity was fluctuating around zero percent in early May 2017, and the confidence indicator at that time ranged from 13 to 30, where confidence is measured on a scale of 0 to 100, with 100 being most confident.
The CARA stock price showed considerable volatility in the latter part of June 2017 with the market anticipating and then later reacting negatively to disappointing research trial results. During this period, borrow intensity moved from minus 6 percent to minus 26 percent, with confidence gradually stepping up from 21 percent to 40 percent. The increased confidence and high momentum of the borrow intensity indicators suggested significant dislocation. CARA stayed persistently hard-to-borrow throughout the remainder of 2017.
Using borrow intensity to sharpen focus in securities lending decisions
The examples here demonstrate the range of views possible with Hanweck Borrow Intensity Indicator data, from single security to broad demographics of lendable securities. At the core, this data is based upon transparent and neutral data derived from quoted derivative markets. Borrow intensity indicators complement existing securities lending data, such as utilisation and percentage on loan, and borrow intensity convergence or divergence with other market data can signal changing market conditions and new lending regimes.
More broadly, this data is an example of evolving analytic approaches to creating alternative data sets, and it is exciting to apply such techniques to inform the securities lending market.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
