Bravo to Todd Tibbetts of Sandoz Corp. for his Sept. 4 commentary, "A better way than high-cost securities lending," which exposes the investment risk assumed by plan sponsors participating in securities lending. Hopefully Mr. Tibbetts' commentary will initiate an investigation by plan sponsors into the inner workings and evolution of securities lending, which is overdue.
Securities lending is not necessarily a bad thing that should be avoided. It has played a major role in increasing the liquidity and improving the efficiency of the global stock and bond markets. Also, securities lending has, for a very high percentage of time, added incremental returns to plan sponsor's portfolios, offsetting pension expenses.
But as Mr. Tibbetts skillfully points out, things have changed. Because of increased competition among securities lenders, the game has changed from simply receiving Treasury bills as collateral and collecting a fee. In the past, brokers would call securities lenders for securities to cover fails. Nowadays, the situation is reversed with the securities lenders calling the brokerage community to push out securities. Brokers can now give out cash for borrowing securities and receive a negotiated rebate rate from the securities lending agent.
It boils down to simple supply and demand. If demand for securities is from the brokerage community (and it exists more today for foreign securities and small-capitalization domestic equities than for large-cap domestic equities), then the securities lender can earn a positive return with virtually no risk. But if securities are being pushed out for cash, investment risk must be assumed to earn a positive spread over the rebate rate. Plan sponsors need to evaluate whether they want to participate in this activity. If they do, it should be treated with the same caliber of fiduciary standards applied to all investment decisions.
After the 1994 experience with rapidly rising interest rates, some securities lenders that aggressively invested cash collateral received from lending out plan sponsor securities were unable to earn a positive spread over the rebate rates. Some large plan sponsors who carefully monitored the results of securities lending activity quickly saw what was going on and pulled the plug or reduced exposures when the "free lunch" wasn't free anymore. But many more were caught off guard and I'm sure that many are not even aware that there was a problem.
In many situations where a commingled fund is employed by the plan sponsor, the trust - not the plan sponsor - makes the decision whether to use securities lending and also decides who manages cash collateral. Not surprisingly, the cash collateral management assignment often goes to the same investment management firm retained by the plan sponsor originally, provided a short-term management capability exists. In some cases, there may be an additional management fee levied for the short-term management. This might be viewed as an additional fee and should be thoroughly reviewed by plan sponsors. But the more important point is that a plan sponsor does not usually receive any report from the manager separating out the results of securities lending from the equities or fixed-income management. The plan sponsor contractually gives the responsibility of securities lending oversight to the trust. If there is a problem, there is a good possibility the plan sponsor will not be informed. Even in a commingled index fund, a blow-up in the cash collateral management part of the securities lending program can translate to the bottom line as an "acceptable negative tracking error" for the fund's performance. But a plan sponsor should not be duped into believing that. There are two distinctive investment decisions here, one for the asset manager and one for the securities lending program. Each belongs in a different expected risk/reward category and should be evaluated as such. Plan sponsors should be requesting separation of these two results.
The traditional fee structure of securities lending sounds good, for the agent. A 30% to 50% piece for the agent and a 70% to 50% piece for the plan sponsor is all right -if there is any money made. If the collateral management has losses, that's the plan sponsor's problem. This is an asymmetrical distribution of risk-adjusted returns favoring the securities lending program. Again, only the plan sponsor using a separate account will know about a problem when there is a request for funds to make up for any losses. For a commingled account, who is objectively negotiating the securities lending fee arrangement, performing due diligence and monitoring results?
The fee structures will likely change for the better for plan sponsors. There are now instances of revenue guarantees offered by agents and third-party lenders (non-custodial) to larger plans. This creates a separate evaluation issue for the plan sponsor, i.e., assessing the creditworthiness of the provider. Large plan sponsors should review their current arrangements and request bids from providers periodically. Plan sponsors using commingled funds should request disclosure of securities lending arrangements and evaluate the short-term investment management capability. The cash collateral investment guidelines need to be thoroughly reviewed. The underlying securities and strategies should be sensible. There should be a strong link between the lending desks and the collateral management desks. Securities should not be pushed at unrealistic rebate rates, forcing the cash management team to take unacceptable risks.
The bottom line is that the nature of securities lending has changed. There is a level of investment risk being assumed which did not exist in the early days of securities lending. Plan sponsors need to bring themselves up to speed in this area and develop appropriate standards for monitoring and review.
Mercer Investment Consulting Inc.