that the Bank hold retained earnings against risks greater than those of an asset holding the second highest credit rating from an NSRO (e.g., AA from Standard & Poor's). The proposed rule eliminates this requirement. We suspect, given the inadequacy of the notice, that the effective capital requirement for junk AMA under the proposed rule is a fraction of what it is under the current rule. If the Finance Board decides to republish the rule, we strongly urge it to give commenters adequate notice of the precise capital impact of FHLBanks holding noninvestment grade securities or loans. Credit Enhancement Requirements -- Proposed § 955.5(b) The preamble states that the proposal "would retain the requirement in.....the current rule that the credit enhancement must be provided by a member or housing associate of the Bank," and would "allow a Bank greater freedom to design specific AMA Products." 68 Fed. Reg. At 39,029. In fact, however, the proposed rule would do the opposite: allow the member to shed all but minor credit risk by transferring that risk either to the FHLBank or a mortgage insurer. Economic cost; third party credit enhancement. The current rule requires that the member retain the first loss position, and that any insurance cover losses only after the member has taken losses on a credit enhancement that was sufficient to achieve an investment grade rating. This requirement was absolutely central to the Finance Board's legal justification and policy rationale for adopting the final rule in 2000: The Finance Board believes that the recourse model, under which the seller of a mortgage retains all or part of the credit risk, is a more economically efficient system for bringing the benefits of the capital markets to the mortgage industry. Under the recourse model, entities that underwrite the loans benefit from good underwriting and therefore are economically disciplined to reduce credit risk. In contrast to the insurance-based secondary market model, under which Fannie Mae and Freddie Mac are paid a premium to insure against credit losses, the recourse model allows an originator to take on more credit risk (so long as that risk is adequately capitalized) and to profit from successful management of that credit risk. Thus, credit risk is dispersed among the many potential originators in the Bank System, and even further dispersed through the permitted insurance and credit derivative structures. In general, [the AMA rule] enables the Bank and the member to take best advantage of their core competencies by: (1) requiring the member to bear most of the economic cost and the management burden associated with lowering the credit risk of AMA assets to levels comparable with investment grade rates securities; thus (2) leaving the Bank with AMA assets that have a risk profile similar to the securities that have historically been a normal part of Bank operations. 65 Fed. Reg. at 43975 it, the proposed rule would allow a FHLBank to hold substantial - g strange that Finance Board has proposed to eliminate such a critical hout any meaningful notice or discussion. To the extent that the egally valid, it must have been based on a theory that the retention of member made AMA the functional equivalent of an advance, and lental" activity. The proposed changes destroy that notion. oly, the section of the preamble discussing these changes concludes, not intended to alter the economics of the current risk-sharing simply cannot understand the factual basis for such a statement. nce Board should decide to republish the rule, we urge it to explain the asis for this sea change in its approach. igations of members. Under the proposed rule, members must continue to secure advances. The Finance Board seeks comment on whether to matter, Congress was quite strict in specifying the types of collateral lity of this step aside, it is also poor policy and will inevitably lead to HLBanks. Transactions 8 HE RA RA Section 955.5(c) provides: “Grandfathered transactions. Assets acquired in transactions otherwise authorization, shall not be subject to the requirements of this section.” This provision is described in the preamble as “minor technical revisions" to the existing grandfather rule, relating to “AMA-type assets acquired by a Bank in accordance with prior resolutions of the Finance Board." 68 Fed. Reg. at 39,030 (emphasis added). Grandfather provisions are commonly understood to allow existing activities to continue, even when a change in applicable law would have ended or amended them. If “prior” in the preamble is meant to convey "prior to adoption of the rule,” and bind the Finance Board's interpretation of the rule, then § 955.5(c) would be a grandfather provision. As drafted, however, section 955.5(c) does not contain any “prior” reference and thus is not a grandfather provision in any sense of the term, but rather is a grant of authority to the Finance Board allowing it to exempt any Bank from any provision of the rule without notice and comment. (Under this reading, the reference to "prior" in the preamble presumably would mean "prior to acquisition of the asset.”) We urge the Finance Board to clarify the matter in the event this section of the proposed rule is republished. AMA-Qualified Interests in Whole Loans As noted above, we oppose expanding permissible AMA assets to include interests in whole loans in addition to whole loans. The proposed rule codifies the recent practice of allowing a FHLBank to deal with members of other banks, using that member's own Bank as a conduit. The FHLBank Act nowhere authorizes a Bank to deal with banks or thrifts that are not its own members and shareholders. This practice is a clear evasion of the Act, which we oppose. $955.8-- Administrative and Investment Transactions Between Banks The Finance Board's current AMA rule allows one Bank (hereafter, "passive Bank") to effectively outsource its administration of an AMA program to another bank (hereafter, “dominant Bank"). The pricing function can be included in this delegation, subject only to the caveat that the passive Bank must retain a right to refuse to acquire AMA at prices it does not consider appropriate. The Finance Board seeks comment on whether to eliminate this one caveat. Allowing a FHLBank to effectively outsource the pricing of its assets to another institution presents a serious safety and soundness risk under the current rule. Removing the one line of defense for the passive Bank effectively places it at the mercy of another competitor at that. This step would raise innumerable potential safety hese issues would be less serious if, according to the FHLBank Act, ed capital requirements for AMA the proposal envisions the FHLBanks taking on considerably more risk sed rule contains no explanation of the effective capital charges that ad the Finance Board for its announced decision to withdraw the I am writing to reaffirm and underscore Freddie Mac's commitment to completing the |