The right insights, right now

Access the latest news, analysis and trends impacting your business.

Explore our insights by topic:

About Broadridge


The Green Roof

Making recalls automated, data-driven and ESG-friendly.

It is very easy to keep feeding the hype on topics such as automation, data and environment, social and governance (ESG) driven investment. What conference would be complete without sessions on each of those topics? However, there is one area of the securities lending business where these important but much-hyped topics come together in a very real way, recalls management.

Generating and managing recalls is one of the fundamental processes of the securities lending business. Securities lending transactions are normally transacted on an ‘open’ basis i.e. there is no fixed term to the transaction. The end of a transaction is generally dependent on either the borrower determining the securities are no longer required, generating a return or a lender determining they need the lent securities to be returned and generating a recall.

Though some trading systems have functionality to significantly automate the recalls process, in general there is a high degree of manual effort involved in recalls management including a great deal of reliance on human decision making. Recalls processing has two main areas:

  • The internal decision-making process regarding when to make a recall and what to recall
  • The management of the bilateral process between the party making the recall and the party being recalled

What generates the need for a recall?

The most common motivation for recalling a stock that has been loaned out is having a short position in the security lent. The short position would typically result from the lender selling the security or making a loan to another party (potentially at a better rate). Where the lender is using an agent it may be possible for the agent to re-allocate the loan to other funds but if it’s not possible to fully cover the short by re-allocation (or the security has been loaned on a principal basis) there will generally be a need to initiate one or more recalls of loans.

The other motivations for recalling a stock are:

  • Corporate actions – the lender may have policies for recalling securities if there are up and coming corporate actions such as dividends, share splits or rights issues. Particularly if the lender wishes to avoid tax complications and the risks related to dividends or exercising a particular option where there is some degree of choice in the corporate action such as whether to take up rights or not
  • Credit limit breaches – a credit limit breach at the legal entity, counterparty or fund level may mean trades need to be recalled to reduce the overall position in order to clear the breach
  • Changes to lending policies - a recall is also possible where the lender has changed policies about whether particular stocks are to be lent or what proportion of a portfolio are to be lent
  • General meetings or any other vote in relation to the issuer (this is discussed in more detail below in relation to ESG)

Using data to drive the recall process

Identifying the short — the fundamental data required to drive a recalls process, whether manual or automated, is good quality inventory data. This is necessary to view current or projected short positions. With that data it should be possible to automatically generate recalls but given that future short positions will be impacted by pending trades and potential settlement failures, an automated solution needs to give a lender a choice of algorithms that are to varying degrees ‘optimistic’ or ‘pessimistic’ regarding what to view as a short. For example, should a position include traded positions (optimistic) or just settled positions (pessimistic)? Should intra-day purchases be included (optimistic) or just intra-day sells (pessimistic), or one of many variations on this theme?

Identifying what to recall — the first step to identifying what to recall is identifying outstanding loans for the relevant securities which could potentially be recalled. Data to drive this process comes from both the outstanding loans and the historic data on recalls. Relevant factors include:

  • The loan type (whether fee, rebate or evergreen) – the recall algorithm for instance may prioritise cash collateralised trades over non-cash collateralised
  • The fee and rebate rates — typically loans with lower fees or higher rebates would be recalled first
  • The quantity — an algorithm may prioritise minimising the number of loans that need to be recalled by choosing the largest trades
  • The age of the loan — newer loans would typically be recalled later that older trades
  • The number of times a counterparty has been recalled — recalling a particular client too frequently is bad for client relations and may undermine the broader relationship 

The recall lifecycle

Recalls in general have a similar life cycle globally. However, in the US domestic market there are additional steps and variations such as callbacks and terminations, making it a more complex workflow. There are also market utilities available from DTCC (smart/track available in US), EquiLend and Pirum that help the bilateral management of the recall process. The following is a simplified generic semi-automated workflow:

  • The recall is identified
  • The recall is ‘generated’
  • The recall is in an unactioned state ‘pending’
  • The recall is ‘sent’ (via email, phone call or any other communication method)
  • The recall is disputed
    • The disputed recall moves to either a ‘cancelled’ state, or
    • The disputed recall moves to an ‘agreed’ state
  • The recall is cancelled – end of lifecycle
  • The recall is agreed – but the settlement date or quantity differs
  • The recall is agreed with no changes to its parameters
  • The return is booked
  • The return is cancelled (due to a cancel of the recall or other reasons)
  • The return is booked but the settlement status is ‘failing’
  • The return is booked but the trade has failed settlement
  • The return has settled – end of the recall lifecycle

Bilateral communication

Current processes for managing the communication relating to recalls tends to be manually intensive involving spreadsheets, emails, Bloomberg messages and phone calls.

There are also multiple vendor platforms for managing recalls that can simplify the process but there is no prevailing market standard. Establishing a market standard for the communication of recalls data is one of the areas that could potentially be achieved through initiatives such as the International Securities Lending Association’s Common Domain Model (CDM) initiative. This would be a good opportunity to make the recall lifecycle more granular to enable additional standardisation and automation of currently highly manual processes. Any standard introduced however needs to be global and incorporate the requirements of both the US and International markets.

The ESG dimension

Though one of the most publicised aspects of ESG is the environmental dimension, the governance aspect is equally, if not more important. In general, good corporate governance policies aim to avoid the problems that can arise from the separation of ownership and control in public companies, the well-known ‘agency problem’. Over the past 18 months of frothy and volatile markets, it has sometimes been very hard to see a direct relationship between good corporate governance and shareholder returns. However, there is a solid body of research going back decades that demonstrates that firms where there are clear mechanisms to monitor and hold management accountable, generate higher shareholder returns. Not to mention reduce the risk of general mismanagement or even outright fraud.

One of the key measures of a buy side firms’ commitment to investing in firms with good corporate governance is corporate voting. Voting on issues such as the constitution of the board and executive are vital to maintaining accountability. Corporate votes may also be on matters that impact other areas of ESG. Voting on matters related to diversity and inclusion, for instance, are important to social objectives of ESG. Feeding corporate voting data into an automated recalls process can improve both efficiency and firms voting records. Voting information can trigger either automatic recall of equities or flag loan trades as requiring a recall to allow a human to determine whether to make a recall. Using the right data provider allows an even more effective and targeted identification of the need for recalls. Data available includes information on the specific ESG objectives impacted by announced votes as well projected (but not yet announced votes) based on historical data.

It is likely the importance of voting rights as a motive for recalls will have an increasingly large impact, potentially leading to higher volumes of recalls in the future. Thus, putting more pressure on the industry to improve automation.


Though recalls are frequently treated as an afterthought to the core trading process, the costs of and impacts of a manual process can be significant. Failures in the recalls process can lead, amongst other things, to reputational damage, loss of voting rights, credit limit breaches and corporate actions losses.

Recalls processing is an area that would clearly benefit from more automation, particularly automating decision making though the better use of data. It is also a great example of an area where there is scope for data driven automation to add additional value, notably in ESG, without adding additional headcount.

Contact Us

Welcome back, {firstName lastName}.

Not {firstName}? Clear the form.