What is LDV ?

Who benefits from LDV?

LDV benefits all participants in the securities finance industry.  Lenders are better able to exercise their corporate governance responsibilities and, since lenders recall fewer loans, overall securities lending volume and revenue increase.  Loan, borrow, and collateral portfolios are more stable, allowing agents and brokers to more effectively manage investment, counterparty, and operational risks.  Corporate issuers receive more proxy votes from long-term investors, allowing them to reach quorum more quickly and at lower cost, and counterbalance votes of short-term activists.  Higher loan volumes also improve financial market liquidity and price discovery.

 

What is Lender-Directed Voting, or LDV?

LDV is a new process that matches securities lenders' loaned shares to broker securities that would otherwise go unvoted, enabling lenders to direct proxies without recalling loans.  It substantially improves existing market practices, which require lenders to recall loan in order to vote proxies.  Recalls are inefficient in that they reduce overall lending and borrowing revenue, and create instability in loan, borrow, and collateral portfolios. 

Why haven't lenders voted on loaned shares in the past?

Historically, institutional securities lenders had to forgo voting rights on loaned shares because there was no mechanism to vote without recalls.  Recent technology and transparency improvements in securities finance markets, however, enable loaned shares to be matched with broker shares that would otherwise go unvoted.  In particular, the Agent Lender Disclosure Initiative made apparent the direct counterparty relationship between lenders and broker-borrowers and provided brokers with detailed loan data necessary to include lenders in their proxy allocation routines.

Are there enough unvoted shares to cover lender voting interest?

Approximately 60 billion U.S. equities go unvoted each year[1], while roughly 15 billion shares are on loan[2], suggesting that sufficient votes could be available to meet lender vote demand.  However, it is unlikely that lender voting interest will be fully covered for all issues, such as those with particularly contentious proxy events or that are hard-to-borrow in securities lending markets. 


[1] www.broadridge.com/investor–communications /us/Broadridge_Proxy_Stats_2010.pdf
[2] Data from RMA securities lending composite, assuming $20 average stock price

Does the broker have the lender’s shares on the proxy record date?

1.  U.S. Federal Reserve Regulation T (“Reg T”) defines the permitted purposes for the extension of credit in the borrowing and lending of securities. In general, all of these purposes involve settling trades through re-delivery of the borrowed securities. Most often, the broker’s need to borrow has arisen after failing to receive securities required for an impending trade settlement, either as the result of an operational breakdown or after a short sale.

2.  Given the broker-borrower’s mandatory compliance with Reg T, it can be argued that borrowed shares, which are re-delivered in the settlement of a trade, are not available on the broker’s books (as a technical matter, the position would be held at DTCC) in order to earn voting rights on the proxy record date. However, this argument would only be true per se if the settlement took place on the proxy record date, because an analysis of the ongoing process reveals that the proxy votes, not just the entitled shares, are properly treated as fully fungible on the broker-borrower’s books.

3.  Reg T does not require that the borrowed shares be returned to the original lender when a subsequent receipt of securities is used to offset the original failure-to-receive. At that point, the borrower can certainly return the securities to the original lender. Yet, an active borrower can also compliantly decide to close a loan of the same securities with a different institutional lender whose terms may have become less attractive or from another broker-dealer lender who may be viewed as more likely to recall shares at an inconvenient time in the future, especially if the shares were borrowed for an ongoing short position. Still another reason may exist to hold the securities if the broker considers the return on its cash collateral, received through a rebate from the lender, to be very attractive compared with other investment options. In all those cases, as well as for actively traded issues where there may be a high risk of ongoing settlement failures, the broker can simply keep the newly-received shares in its inventory, balanced against its obligation to the lender.

4. As a result of efficient management of its settlement obligations, a broker – perhaps all brokers – may well have borrowed positions on their books on proxy record dates. The brokers would have gained the right to assign proxies or even to vote at the next corporate meeting as a direct result of the original loans from institutional lenders. In effect, the proxies are fungible on the brokers’ books, along with the borrowed shares themselves subject, of course, to an equitable assignment of proxy rights in compliance with stock exchange rules. Yet, brokers are not expressly permitted to assign proxies to their institutional lenders. At this point, the Lender Directed Voting (“LDV”) argument gains relevance and substance.

5. As noted, in addition to holding the shares cum voting rights, the broker also retains an obligation to its original lender. Indeed, one could argue that an institutional lender's ownership rights are stronger than those of other “beneficial owners” to whom the broker owes shares in the same securities. That is partly due to the distinction that can be drawn between the institutional lenders, who do not receive proxy assignments, and the broker’s own margin customers and hedge fund clients, who do receive proxy assignments. The distinction resides in the timeline of their property rights: the former owned the shares fully prior to lending them to the broker, while the latter required broker-financing in order to acquire their positions. Although we have seen that the institution’s shares may now be on the broker’s books, it is very likely that the financing customers’ shares are out on loan, i.e., hypothecated as collateral to source the broker’s own funding needs. And, in such cases, those positions are truly not in the brokers’ DTC account, although the brokers may well be assigning proxy rights to their accountholders. One can ably argue that those proxies would more equitably be assigned to the institutional lenders.

How can lenders instruct broker shares?

Brokers administer proxy allocation routines to distribute proxies to their customers.  Since broker shares are held in fungible bulk and lenders have beneficial ownership to loaned shares, brokers can include lenders in their allocation routines.  After brokers allocate proxies to lenders, standard proxy processes are followed to garner and submit voting instructions and submit them to corporate issuers.  For example, proxies are assigned to Broadridge accounts designated for the lenders, then are instructed by lenders or ISS on the lenders' behalf.

Could lenders also instruct custodians' unvoted shares?

Regulatory and operational considerations may pose challenges to matching custodians' unvoted shares with lenders’ loan positions.  In particular, custodian shares are not held in fungible bulk, as are broker shares, which presents difficulties when considering custodial allocation of proxies across lender accounts. Furthermore, custodians are not counterparties on loans, so the lenders are not beneficial owners to any of the custodians’ unvoted shares.

Does LDV contribute to “over-reporting,” since lenders’ shares were delivered to new buyers who now have the associated voting rights?

Existing proxy reconciliation processes are sufficient to address any potential "over-reporting" issues.  For example, brokers already use post-reconciliation processes to mitigate the risk of over-reporting that may arise from assigning proxies to margin customers whose shares may have been loaned or rehypothecated.

How do brokers decide which lender(s) are assigned proxies?

Beneficial owners and regulators have expressed concerns about voting opportunities being directed to preferred lenders or leveraged for beneficial loan terms.  In the same way that agent lending queues are designed so that lenders get equitable access to borrower demand, brokers need pre-defined and algorithmic “proxy queues” to ensure equitable assignment of voting opportunities.  Furthermore, on-going auditing and validation of proxy assignments may be needed to ensure against development of a “market for votes.” 

What if proxies are not available from a lender's borrower, but are from another broker?

Reallocation of the loans to brokers with available proxies would increase overall lender voting opportunities.  However, numerous other loan factors would need to be taken into account, such as counterparty risk assessments and credit limits, loan prices, and collateral types and quantity.  Considering these factors, loan reallocations may not be in the overall best interest of lenders and borrowers, and will have to be considered on a case-by-case basis.

How can lenders know, before record date, how many proxies they will be assigned?

To the extent that lenders receive proxies through LDV, they will not have to recall loans to regain voting rights.  However, broker holdings change daily and varying numbers of investors vote, so the number of proxies that can be assigned to lenders cannot be known with certainty until just before the meeting date, which is typically two months after lenders must make record date recall decisions.   The number of available proxies must therefore be forecasted, taking into account factors such as each broker's customer base, the scarcity of shares in the securities lending market, and the expected materiality of proxy ballot items.

Corporate Governance Blog

Monday, November 8, 2021

Exposing the Rogue Traders

The Case for a Cross-Border Stock Loan Registry, Part II


Author: Ed Blount

RogueTraderDigitalMarket

Master Criminals don’t usually confess in public. If prosecutors’ charges are true, Sanjay Shah is the leading figure in the largest reported tax swindle in history. Yet, Mr. Shah, unbowed, pleading his case to reporters, has openly admitted to borrowing the assets of widows and orphans in one country to kick-start a pyramid scheme of dividend capture trades, so as to swindle widows and orphans in other countries. Mr. Shah’s attorneys argue that his trades were not illegal. Mr. Shah, according to the reporters, claims everything he did was legal, and then he appeals to the Law of the Jungle:

 

“If there’s a big sign on the street saying, ‘please help yourself’,
then me or somebody else would go and help themselves.”

 

------------------------------------------------------------------------------------------------------

 

 

Ed Blount, executive director of CSFME, reacts to the €150 billion “Cum-ex II Papers” controversy, as reported on October 21st.  Ed was the lead investigator on our 2010 Empty Voting project and testifying expert in the $450 million Lehman unsecured creditors vs. IRS suit, as well as two dozen litigation and compliance matters in securities finance since the onset of the credit crisis.

 

Q.        What would you say to a judge hearing this case?

A.         He’s not a trader. Criminals like Shah do not meet threshold standards for markets.  He’s just a thief.

 

            Just because the money is lying in the middle of the street is no defense for theft. As a society and an industry, we reject that premise. Money is never abandoned in the real world.

            We’re discussing standards for acceptable market behavior, and Shah is responding with a fairy tale, just like a rogue trader explaining away the fact that his books won’t balance.

            Civil society cannot tolerate Shah’s behavior, much less his pyramid manipulations. Securities finance has an existential risk, but there is a serious reputational risk here for all financial institutions. Borrowing pension assets to siphon from a government treasury, no matter how clever, is just as evil as stealing from its social safety nets.

          We all have to repay in taxes the money they steal. Real traders create wealth; they don’t steal it. We don’t need legal language to describe a thief.

 

Q.        How can we expose rogue traders like Mr. Shah in the future?

A.        Strip away their cover.

 

            Just like pickpockets, rogue traders work in places that are very busy with lots of distractions.

            Record dates are popular because of the many positioning trades made in advance of the entitlement postings. After record date, reversals may truly be as they appear, unwindings of dividend capture or similar legitimate trades. But some will be tax-driven, without economic substance. That’s what we’re looking for.

            If we can scan and tag the legitimate, socially benign trades, tax auditors will have fewer left to examine in our search for the fraudulent trades. Rogues, as a result, will have fewer places to hide.

 

Q.        What then?

A.         Look for signs of desperation or excess in the rest.

 

            Fees are a good tell. When Bear Stearns was failing, their cage managers would pay any price for liquidity.

            Collateral is another. Lehman’s “Repo 105” trades were overcollateralized to appeal to unsuspecting counterparties. But it sent a signal to the Market Posse that Lehman’s securitization pipeline was slowing, and their liabilities were coming due. I later described those “Liability Dynamics” for Bear Stearns, Lehman Brothers, and other distressed firms in the RMA Journal.

            In particular, I wrote, look for evidence of distress in the form of unusual margins, positive and negative, in the remaining trades.

            Then audit the outliers.

 

Q.        Where is the data for that analysis?

A.         Some of it is available from vendors, and other useful data can be derived from regulatory filings. Service providers have most of the data, but it’s spread out.

 

            Commercial data vendors such as FIS ASTEC, S&P Global (IHS Markit), and Datalend all have the ability to look for outlying loans and collateral. However, those vendors don’t have access to the Know Your Customer” (KYC) profiles, contracts, trust indentures, and other policy documents that guide and constrain a trading desk. And data vendors must always keep in mind their own business models. The profit opportunity from surveillance may not be as appealing as other areas into which their resources can be directed.

            Regulatory reports, like those required by the European regulations SFTR and MiFID II, can be foundational to the analysis. There’s now an enormous data lake of details on loans from the summer of 2020, when SFTR went live for securities loans.

 

Q.        Can’t government agencies use that data to catch rogues and fraudsters?

A.         They’re certainly trying, but they also have limits, both in reach and resources.

 

            Government agencies cannot exceed their jurisdictions, except with the cooperation of foreign agencies willing to share intelligence. Foreign agencies might want to help, but their own regulations constrain them. Some of the documents they would need are firewalled in other agencies. Only the actual asset owners can access all the records required, either for compliance or surveillance. But even if the foreign KYC documents and legs of the trade were known, domestic regulators would have to revise their data models to build or manage an account-level registry for cross-border loans.

           

Q.        As you say, Banking itself has serious reputational risk in this scandal. Are existing bank systems equipped to expose the rogues?

A.        Banks have great systems, but they’re not designed for surveillance.

 

            Compliance systems are regularly upgraded to shadow changing regulations. But there are very few regulations in the cross-border markets. Any systems to track best practices would require a redesign. Bank legacy systems are excellent for the work they do, but they are often two or three generations old.

            It should be no surprise that regulators, led by ESMA, are pressing banks to build distributed ledger-based (DLT) infrastructures that can be used for market surveillance. The 2016 EU directive to disclose loan records by the Securities Financing Transaction Regulation (SFTR) was the opening gambit by regulators. Others have taken up the challenge. For example, compliance consultancies are exploring the use of blockchains and smart contracts to find fraudster networks that are operating in the covid-welfare distribution space. Some of those techniques may be useful.

 

Q.        How can all that data be compiled?

A.         Forensic and data analysts are very experienced in creating specialized databases.

 

            For the Wells Fargo account-opening fraud, the forensic auditors at FTI Consulting, Inc. reportedly analyzed 35 million documents and unstructured data from more than 300 custodians. Their report to the independent board members concluded that “events show that a strong centralized risk function was most suited to the effective management of risk.”

            That’s what securities finance needs now: a centralized loan database and registry to minimize the entire financial sector’s reputational risk from cross-border stock loan frauds.

 

Q.        What about the privacy and confidentiality issues?

A.         Bitcoin brought encryption into the mainstream, and Facebook’s profiling algorithms pushed data trusts into the spotlight.  Let’s use the new DLT tech if it survives testing.

 

            Traders are rightly concerned about exposing positions to raids in the GameStop squeeze era. I think that can be solved by lags and by puzzles. That is, if we are only building a compliance filter for banks, not a surveillance tool for regulators, there is no time sensitivity for the loan postings. A three-month posting lag would protect trading confidentiality but still be sufficiently current to filter the tax auditors’ screens for suspicious loans. Encryption can protect sensitive fields in the records, creating puzzles that defy access without the private keys.

            For asset managers and owners, a data trust can provide a legal framework to protect the resources used to validate borrower strategies and even to monitor style drift. Data trusts are a just-in-time solution for this problem. Let’s use the new tools available and reassure asset owners about the social propriety of their loans -- and let’s give banks an ESG feather for their bonnets.

 

The Massachusetts Institute of Technology (MIT) has named “Data Trusts” as one of the “10 Breakthrough Technologies of 2021”, along with Messenger RNA vaccines, lithium-metal batteries, digital contact tracing and Tik-Tok’s feed algorithms.
 
 

Q.        Your old firm, ASTEC Analytics, created the first benchmarking systems for securities loans in the 1990s. Can’t a powerful database like that do the job?

A.         Not likely. Purpose-built databases are inadequate for surveillance work.

 

            I’m not privy to the tech just introduced in Lending Pit II, but, as a rule, vendor systems in securities finance were optimized for benchmarking. Their table structures, key fields, and SQL queries weren’t designed for surveillance or even for compliance. However, the new distributed tech of blockchains and shared ledgers offers data scientists a fresh start and may well be flexible enough to platform the analytics. Down the road, smart contracts can help banks to sort through the data and tag suspicious trades for further inspection. That level of functionality wasn’t in my original design specs for the Lending Pit.

 

 Q.       Is this the project that brought you out of ‘semi-retirement’?

A.         Yes, I was intrigued to answer the critics, who have become quite vocal.

 

            Those who say that securities finance is a platform for widespread fraud and manipulation, where thieves routinely deploy the assets of widows and orphans to steal from government treasuries, are wrong. There are also those who claim that proxy votes are being unknowingly sold to hedge funds. Every industry has a few bad apples. But the men and women in securities finance, by and large, are honest and dedicated stewards of the assets they’re given to manage.

            We can create an answer to the critics’ charges of malfeasance, just as we did ten years ago when we educated the regulators about the mistakes of the empty voting critics. Personally, I’m not interested in ancient history. I’m sure others are working diligently on researching old loans. But I would like to leave a way for the new loans to be seen as legitimate, so the field is accepted as honest and I don’t have to scratch off the last 30 years of my CV.

 

Q.        Will we work with RMA, SIFMA, and DTCC again, as during the Empty Voting project?

A.         We will work with any organization that can help the Market Posse.

 

           We will certainly work again with trade groups and industry utilities and with our forensic consulting colleagues and the platform vendors for distributed ledger technologies.

My own history working with industry utilities goes back to 1974 when I opened the first ADR account at the brand-new Depository Trust Company. That’s me on the left, in my double-knit pinstriped suit, proudly showing our pin-feed position report to Bill Spencer, president of First National City Bank; to Dick Banks, my boss; and to our DTC rep. NSCC was one of ASTEC’s first consulting clients, for a technical assessment of challenges onboarding the custodian banks into the brokers’ Continuous Net Settlement system. In fact, the first time that I ever saw a personal computer was a Heathkit that Jack Nelson, NSCC’s first president, had on his conference table in 1982.

More on that another time, perhaps.

 

Q.        Who else will be helping the Market Posse?

A.        In our next blog, I will introduce Eric Poer, lead forensic accounting partner at FTI Consulting, to explain how he and his team exposed the largest consumer fraud in U.S. history.

 

           FTI Consulting, on behalf of the independent directors on the Wells Fargo board, was engaged by the law firm of Shearman & Sterling to organize and examine data held in scores of systems going back many years. Eric Poer will explain how his team approached the challenge of collecting, analyzing, and collating the data to determine the breadth of the sales practice issues and determine their origins. FTI’s analysis was conducted in the cloud in virtual sandboxes, then collated with unstructured data (i.e., emails, interviews, etc.) and summarized in expert reports and exhibits.

 

          Eric, with 25 years of experience in regulatory inquiries, investigations, and disputes, was also my analytical partner in our support of the tax claims pursued in federal bankruptcy court by Lehman’s unsecured creditors committee. Together, we will explain how we mapped the cross-border loans by Lehman’s subsidiaries to show their economic substance and confirm the legitimacy of the $450 million in foreign tax credits that the U.S. Internal Revenue Service had disallowed.

Print

Corporate Outreach Milestones

MILESTONES FOR LENDER DIRECTED VOTING

May 8, 2014: Council of Institutional Investors; - CII Elects New Board, Names Jay Chaudhuri Board Chair. http://www.bloomberg.com/news/2014-01-31/north-carolina-treasurer-may-cede-pension-control-5-questions.html )

February 2014:  Swiss Minder Initiative implies the value of LDV. http://www.ipe.com/switzerlands-minder-initiative-will-cripple-securities-lending-experts-warn/10000947.article.

January 2014FL SBA begins their SecLending Auction Program with eSecLending.

November 27, 2013 – CSFME staff call with Glass Lewis Chief Operating Officer. He gave his commitment for cooperation and support for LDV, and most importantly, he suggested that perhaps we should discuss with a Broadridge/State Street/Citi the scenario that permits Citi to forward an “Omnibus Ballot” of proxies to State Street, which State Street would then take and assign the proxies to their pension lenders/LDV participants, which would then be incorporated into a single ballot and sent to Broadridge. This eliminates the secondary ballot issue. While this description is oversimplified, Glass Lewis was fairly certain the parties involved could operationally create such a combined ballot. Responding to the question on cost, the Glass Lewis executive stated that the cost depends on the number of voting policies a fund has. Most funds have one policy; therefore, depending on the client, the cost would be $.75 – $2.00 per ballot.

October 21, 2013 – CSFME staff call with ISS Chief Operations Officer. He committed his cooperation and support to advance LDV’s implementation into the markets. He responded to the question about cost: “It depends on the client and the services they use. $6-7 per ballot on average.”

June 25-28, 2013 – CSFME staff attended ICGN Annual Conference in NY, NY. Spoke with executives of CalSTRS; ICGN Chair and Blackrock about LDV.  We received favorable comments and encouragement from each.

June 6, 2013: CSFME meets with Chief Investment Officer for NYC Pension Funds. While very much in favor of the LDV concept, the comments that the NYC Pension Fund Boards are for the most part followers in new initiatives and would prefer a roll-out by other funds first.

April 5, 2013: ‘SEC gives CSFME limited approval for LDV going forward’ providing brokers assign proxies only from their proprietary shares.

March 26, 2013 – CSFME and its legal team presented the case for LDV to SEC Commissioner Dan Gallagher. Present by phone and speaking on behalf of LDV were representatives of FL SBA who spoke about the difficulty of timely recall of shares on loan following release of record date and issues on agenda; and a representative from CalSTRS who spoke about their recall policy affecting income.

March 13, 2013 – CSFME meet staff of Senator Rob Portman and Congressman Steve Stivers of Ohio. These meetings were for the purpose of lining up political support, should the SEC resist the LDV concept. We also met and spoke with CII Deputy Director Amy Borrus for one hour and 15 minutes for a scheduled 30 minute meeting.  She expressed great interest in the value of LDV to long-term beneficial owners.

January 17, 2013 – CSFME conference call with CoPERA Director of Investments.  Among CoPERA’s concerns were: (1) How are agents/brokers notified re: LDV? (2) Who moves or approaches first lender to agent or agent to lender? CSFME responds  that a side letter is needed between lender, agent and broker.

November 8, 2012 – CSFME conference call with Council of Institutional Investors (CII) detailing LDV. Some in attendance were opposed to securities lending because of their desire to vote 100% of recall. This position would be irrelevant giving CalSTRS’ change to policy on proxy recall.

October 24, 2012, 2PM – CSFME presents LDV to Broadridge Institutional Investor Group. At this meeting, a representative of CalSTRS states: “We would view brokers willing to provide proxies more favorably than those who would not.” We were also informed by CalSTRS that they were looking to change their 100% recall policy. A representative of SWIB led a discussion on International Voting Issues, and apparently was chairing 3 meetings to determine the following: 1. who is voting internationally? 2. What are the issues in the international markets? 3. How do we increase and improve international processes?

October 24, 2012, 11AM – EWB/KT conference call with ICGN.  Executives stated that the argument for LDV may not be as strong in a non-record date market, and asked what would be the cost for LDV.  They further stated that they would like to see the U.S. go with LDV first and would need more information and operational detail.

October 13, 2012 email note from Elizabeth Danese Mozely to Broadridge’s Institutional Investor Working Group: “TerriJo Saarela, State of Wisconsin Investment Board, will provide commentary on their fund’s interest in international voting and an update on her participation in the Council of Institutional Investors’ working group on international voting.  Our discussion will include the differences in process for voting abroad, share blocking, attendance at the meeting via proxy or Power of Attorney (POA), best practices available through the various laws and regulations, etc.”

September 18, 2012: CSFME contacts Blackrock/ICGN Chair for a brief on LDV.

August 13, 2012 – CSFME conference call with OTPP.  Discussion of LDV was not timely in that their SecLending Program stopped lending securities through agents in mid-2006. State Street is their custodian and they were using a tri-party repo through Chase to Lehman, until the Lehman collapse. All the assets sat at Chase. It was not clear who had voting rights. At the time of this discussion in August 2012, OTPP was thinking formulating an SLA because they do not have the capacity to lend securities on their own. We have had no discussion with them since.

August 2, 2012 – CSFME contacts Ontario Teachers’ Pension Plan (OTPP) regarding LDV.

March 19, 2012 – CSFME conference call with executive in charge of securities lending for Franklin Templeton

February 22, 2012ICGN sends LDV letter of support to the SEC, signed by Chairman of the ICGN Board of Governors.

September 30, 2011CalSTRS sends LDV letter of support to the SEC, signed by Director of Corporate Governance Anne Sheehan.

July 18, 2011Florida SBA sends LDV letter of support to the SEC, signed by Executive Director and Chief Investment Officer.

November 2011 – CSFME introduces Council of Institutional Investors editor to LDV.

July 5, 2011 – CSFME sends a Comment Letter to the Securities and Exchange Commission regarding LDV.

October 2010 – CSFME releases report: Borrowed Proxy Abuse: Real or Not? This report and the SEC’s Securities Lending and Short Selling Roundtable prompted the question from beneficial owners and regulators regarding the need to recall shares on loan to vote proxies, why can’t lenders receive proxies for shares on loan when we get the dividends? From this question, the idea for Lender Directed Voting was born.

January 2010 – SEC issues rules that brokers no longer have the discretion to vote their customers’ shares held in companies without receiving voting instructions from those customers about how to vote them in an election of directors. http://www.sec.gov/investor/alerts/votingrules2010.htm. The rule, periodically, contributed to the difficulty of corporate meetings attaining a quorum.

Fall 2009/2010 – Four public pension funds join CSFME in Empty Voting studies/LDV initiative; FL SBA, CalSTRS, SWIB and CoPERA.

September 29-30, 2009 - SEC Announces Panelists for Securities Lending and Short Sale Roundtable; http://www.sec.gov/news/press/2009/2009-207.htm