It used to always make sense, but today there are fewer purchase options and in most cases the cost has increased to the point where they would be better off saving in a 403(b). But not always. I still come across situations where it not only makes sense to make the purchase, it would be crazy not to.
I use several different models to determine whether the buyback makes sense, but the easiest to understand is the one that tells the client "how long do I have to live to break even?" Break even being the number of months one would receive income from a pool of money, earning a specific interest rate and that pool becoming exhausted. There is no right number of months, it really depends on the client, but believe it or not, I've come across some service credit purchases that allow the client to make back every dollar in just over four years of retirement, assuming no interest (longer with different interest assumptions).
Recently, I ran a scenario for a client who had already decided against the buyback. I won't give away the state where this plan resides and we'll name my client George. George had less than four years that could be purchased for about $28,000. He thought this to be too much money. However, the annual increase in benefit was just over $7,500. I didn't have to run any numbers to determine this was an incredible deal, but I ran them anyway.
I wanted to see what rate of return George would need if he invested the money on his own and then provided the same income (adjusted for a similar inflation benefit) to himself and lived to about normal life expectancy. Using 86 as normal life expectancy I calculated that George would have needed a return of almost 21% annually (with no fluctuation in that return) to provide the same benefit the buyback of service credit would provide. This became the biggest buyback slam dunk I've ever encountered. Not buying back the service credit would be leaving potentially hundreds of thousands of dollars on the table. Not only was it worth it, but the client should certainly borrow if they didn't have the funds.
I should state that I rarely run into these "slam dunk" service credit purchases anymore and when I do they are usually of the "repurchase" variety - meaning the client had taken money out of the system many, many years ago and is eligible to re-purchase those years. Normally, when a person takes a distribution of their defined benefit balance after leaving an employer, it is only their own contributions plus earnings that they receive - not the employer contributions. So when they ask to repurchase the credit, they have the employer contributions restored, lowering the cost of the buyback.
Buying service credit can be a smart idea, even if the payback period or internal rate of return is not great. But generally I like to see a payback period under 12 - 15 years (depending on if it's a single life or joint life) and an internal rate of return to average life expectancy over 5%. Client risk profile is very important in this decision, so my numbers will vary with the client.
If you are eligible to repurchase past service credit, you should look into it. If you are eligible to purchase years of service credit for which you were eligible, but didn't work - get the data and have someone run the numbers. While "airtime" is becoming more rare and not usually worth it anyway, you might also look to see if it makes sense for you.
Be careful who you ask to run the calculation, a product salesperson has a vested interest in you NOT doing the buyback. A buyback can be like found money, don't leave it on the table.
Scott Dauenhauer CFP, MSFP, AIF
@403bAdvocate
