In late October, Genworth announced the establishment of a $475 million financing facility to securitize the non-economic portion of AXXX reserves associated with a block of universal life insurance policies with secondary guarantees (ULSG). This marks a milestone as the first securitization to finance the redundancy in statutory reserves underlying ULSG products, where such reserves are calculated under Actuarial Guideline 38, otherwise known as “AXXX.”
Several securitizations have been brought to market over the past few years to reduce the cost of holding statutory reserves associated with level premium term insurance products. These are known as “Regulation XXX” securitizations, based on the name commonly used to refer to the NAIC Model Regulation, the standard that defines the reserve methodology for level premium term products. Up until now, however, companies have been unable to adapt the structure used for Regulation XXX securitizations to an AXXX environment. Now that the first AXXX deal has hit the market, we expect to see more AXXX transactions.
Securitization is an attractive option because it provides an opportunity to boost profitability by converting equity financing to less expensive debt financing (with such debt financing getting “operating” leverage or off-balance sheet treatment if structured properly) while potentially still experiencing significant tax benefits. Because of the complexity and lifetime coverage of ULSG products, there are many moving parts that need to be handled when developing an AXXX transaction.
What follows is an examination of some of the major issues and challenges related to structuring an AXXX securitization that would be considered by various reviewers of a transaction, including regulators, rating agencies, financial guarantors, and investors.
What is reinsured? Securitizing redundant reserves typically involves a reinsurance transaction with a captive. In an AXXX transaction, either the entire ULSG product risk or just the secondary guarantee risk may be ceded to the captive. The insurer ceding the business needs to make sure that if it cedes just the secondary guarantee risk, it gets the appropriate reserve credit it is expecting.
Determining the level of redundancy in AXXX reserves. While most products will have some non-economic portion of reserves, the level of redundancy will vary significantly based on the product design. Also, the statutory reserve requirements have changed over time. Policies issued after June 30, 2005, are subject to more stringent rules than policies issued prior to that date. For new policies issued in 2007 or later, new rules take effect that will likely reduce reserves for most product designs.
Defining “economic reserves.” The term “economic reserves” has been used in securitization transactions to define the level of reserves that can be funded by policyholder premiums. The excess of the statutory reserves over economic reserves are intended to be financed by the capital markets. There is no generally accepted economic reserve methodology for AXXX transactions, although consistent with this concept for XXX transactions, there seems to be movement towards having the assumptions locked in at issue.
In order to assess risk to investors, the counterparties will need to consider the capital levels in the transaction as well as the economic reserves. Even if assumptions are locked in, however, the methodology must be flexible enough to account for the dynamic relationship between the premium funding levels and the face amount.
Lapse assumptions. The lapse rate should be assumed to fall quickly as the secondary guarantee moves in the money. For fully paid-up secondary guarantees that are in the money, reviewers of the transaction will expect the lapse rate to be zero. For in-the-money secondary guarantees that still require a premium to be paid to keep the guarantee, reviewers will expect a lapse rate of 1% – 2%. Consideration should also be given to potential life settlement ownership of the business, as the expected lapse rate would be zero for such policies.
Mortality assumptions. The impact of mortality depends upon the structure of the transaction. From an investor perspective, higher than expected mortality will be harmful in some transactions while beneficial in others. Several issues should be considered when setting a mortality assumption.
o Consistency of recent experience with mortality assumption.
o Slope of mortality rates, particularly at older attained ages.
o Compliance with risk classification systems, including the effects of any table-shaving programs.
o Assumed mortality improvement.
o Underwriting approach for older age applicants.
o Wearing off of preferred risk underwriting benefits.
o Effect of term conversions.
Premium patterns. The flexibility provided to policyholders in determining the level and timing of premiums is one of the more challenging issues to assess. Different premium patterns will need to be analyzed, including low-funded patterns (the policyholder is using the policy for term coverage), short-funded patterns (the policyholder pays for a lifetime guarantee either in a single premium or a limited number of premiums) and lifetime-funded patterns (the policyholder pays for a lifetime guarantee with a level premium). Reviewers will be concerned if the transaction is dependent upon assumed inefficient policyholder behavior, so they will consider whether any assumed premium patterns over-fund the guarantee.
Investment returns. Several different investment philosophies will apply for assets in the captive backing each of economic reserves, excess reserves and capital. Reviewers will also give consideration to investment returns in the ceding company that are assumed to increase over time, perhaps reflecting an assumed reversion to historical interest rates that are higher than current rates. Overstated investment returns will result in overstated projected crediting rates, which will thereby understate the financial impact of the secondary guarantees.
Non-guaranteed elements. Changes to non-guaranteed elements such as COI rates and crediting rates will affect the secondary guarantee risk. These risks may be addressed either with sensitivity testing or with structural design penalties.
Taxes. Although tax issues are not unique to AXXX transactions, the transaction needs to be structured in the most tax-efficient manner to reflect the tax situation of the insurance organization. This will be the key consideration in deciding whether the captive should be situated on-shore or off-shore.
Dividend rules. The insurance company funding the captive will look to have its capital returned as soon as possible whereas investors and financial guarantors would prefer to have the capital remain in the captive until the entire principal has been repaid. Dividend rules need to be negotiated and will likely include various lock-up mechanisms that stop the payment of dividends in the event of various adverse events.
Liquidity risk. AXXX transactions extend longer and run off more slowly than XXX transactions, and they will typically exceed the maximum duration for which securities can be issued. Some mechanism will be required to allow investors to be repaid at the maturity of the notes, even if the debt cannot be re-issued.
Modeling. Investors and the rating agency will depend on a robust model of projected insurance cash flows to perform their reviews. Development of robust models for AXXX business is challenging because of the complexity of the products as well as the complexity of the AXXX calculations.
While there is a long list of challenges and issues, Genworth’s transaction shows they can be addressed. We should expect to see more AXXX transactions in the future.