Q. When I retired from the federal government I was allowed to apply the "three-year rule" to my monthly annuity payments to ensure they would be tax free until my equity -- i.e., my employe contributions to the system -- had been returned.

This year I began receiving Social Security payments based on contributions I have made (and continue to make) to the system as a self-employed person.

Because of other income, I shall be liable for federal income tax on half of these Social Security payments.

However, I already have paid federal tax on my total contributions to the Social Security system, so why isn't the "return of equity" principle applicable in this case? Isn't this a form of double taxation?

A. This isn't a simple question, and as you probably guessed, it doesn't have a simple answer. Taxation of half of Social Security benefits for upper-income recipients indirectly applies a "means" test to Social Security benefits -- something that was not contemplated when the system was designed.

For most beneficiaries, a tax on half the benefits can be explained, if not entirely justified, by the fact that half of the contributions made over the years -- the half contributed by employers -- was not subject to income tax at the time of the payment.

As you point out, however, this situation did not hold for the self-employed; all of the payments to Social Security by a self-employed person were made with after-tax dollars.

However, although you paid a larger percentage of your earnings to Social Security than an employe pays out of his own earnings, you always paid less than the combined total of the employe and employer contributions.

And an employe could, with some credibility, claim that if the employer did not have to pay Social Security tax for the employe, he might be willing to sweeten the wage figure a little -- so the employe is in effect contributing to the employer half of the payments. Certainly, an employer counts his Social Security tax bill as an expense of doing business.

I don't know if Congress considered this seeming inequity when the members worked on the Social Security Amendments of 1983. This is the bill that put the tax bite in place, as part of the rescue operation for the Social Security retirement fund.

If they did, I'm sure they were turned off by the enormous complexity of trying to differentiate between employe/employer contributions and self-employed contributions. Perhaps in your case all of your payments were for income from self-employment.

But it doesn't take any great stretch of imagination to recognize the problem of calculating ratios for people who had both employe wages and earnings from self-employment, either at the same or different times. And do you introduce any computation for the actuarial value of contributions of one type made in the 1950s versus those of another type in the 1970s?

Discussion about the apparent inequities in the taxation of Social Security benefits could continue indefinitely. I suspect the only productive approach is to recognize that drastic action was necessary -- and to be grateful that our income is high enough to require that we pay this extra tax. There are a lot of people who would be delighted to exchange places -- tax liability and all.

Q. Some of your fellow media members have me confused concerning contributions to a spousal IRA. It was my understanding that contributions may be distributed in any way desired as long as the total annual contribution doesn't exceed $2,250. A few articles I have read indicated that the contribution to the non-working spouse's account is limited to $250.

A. The confusion arises because of the $250 difference between the annual limit for a single worker ($2,000) and the limit for a worker and a nonworking spouse ($2,250). In fact, your interpretation is almost right. The $2,250 may be distributed between the worker and spouse any way they wish as long as not more than $2,000 is deposited in either account.