SEPARATE ACCOUNTS FUNDING AND NEW MONEY

The investment of qualified plan funds under an insurance contract can involve certain disadvantages relating to the commingling of the employer's pension funds with those of other employers or with other funds of the insurance company. Separate accounts funding and the new money technique are methods by which two of these disadvantages can be mitigated.

Separate Accounts
Separate accounts funding was developed to avoid commingling pension assets with all of the general assets of the insurance company because an insurance company's general assets have traditionally been invested in long-term, low-return investment vehicles, sometimes subject to legal restrictions on investments. With separate accounts funding, the insurance company makes available one or more special accounts, separate from its general asset accounts, solely for investment of pension money. Such separate accounts may have a specified investment philosophy; for example, one type of account might be invested in common stocks, another in bonds, another in real estate mortgages, another in money market instruments, and so forth. The qualified plan sponsor is then given the option to invest various portions of the plan fund in one or more of the separate accounts. The funds so designated are commingled with similar funds of other qualified plans, as in a common or collective trust fund. If an individual plan fund is large enough, an insurer may also offer the employer an individual separate account only for that employer with whatever mix of assets the employer designates.

New Money
Because many of the insurance contracts available for qualified plan funding involve the mingling of an employer's pension contributions with money invested in previous years, the investment return credited on a given year's contributions is not necessarily equal to the return that could have been earned if the amounts contributed had been newly invested that year. This can present a problem under contracts giving employers discretion as to how much to contribute to the fund during a given year, as is generally true under DA and IPG contracts. Where the composite return provided under the insurance contract is greater than that available for new investments, the employer will tend to maximize contributions, while contributions will be minimized when new outside investments can obtain a higher rate of return than under the contract. To alleviate such problems, insurers have developed methods to credit each block of qualified plan contributions with the rate of return at which the funds were actually invested. There are a number of ways of accomplishing this; such techniques are referred to as new money, investment year, or investment generation methods.

New money methods must not only credit plan contributions for a given year with the current investment rate for that year, but must also take account of the fact that contributions from past years are constantly being reinvested at current rates. Thus, a plan contribution made in Year 1 is credited with the new money rate for Year 1. For Year 2, the Year 1 contribution is credited not with the Year 2 rate, but with a composite of the Year 1 rate and the Year 2 rate, to reflect the fact that a portion (but only a portion) of the Year 1 money will be reinvested in Year 2.

Two broad methods of accounting for these changes are the declining-index method and the fixed-index method. Under the declining-index method, the amount of money associated with a particular year gradually declines as the amount originally invested in that year is reinvested in later years. Thus, under the declining-index method, the amount invested in Year 1 in the previous example would decline in Year 2 to reflect the fact that part of it was reinvested in Year 2. Under the fixed-index method, the amount associated with the initial investment year remains the same, and the rate of return credited to that amount is adjusted to reflect reinvestments without changing the original principal amount. The two methods generally produce the same results.