A “Hidden” Pension Subsidy in SURS

Posted by Jeffrey Brown on Jun 23, 2010

Filed Under (Retirement Policy, U.S. Fiscal Policy)

I have written before that SURS – the public pension system in Illinois covering most workers in higher education – is a highly complex system, and that this complexity often “hides” some of the generosity of the system from both participants and taxpayers.  (I put “hidden” and “hides” in parentheses because SURS does disclose the information – it is just that one needs a pretty high level of knowledge and financial sophistication to make sense of the fiscal implications of what is being disclosed).

Today, I want to talk about one very specific taxpayer “subsidy” to retirees that I do not think many people realize exists. This subsidy comes in the form of excessively-generous annuitization rates.

I will try to spare the unnecessary details, but a bit of detail is required. Individuals in the SURS defined benefit plan have their retirement benefit calculated in multiple ways, and then the participant gets the higher of these amounts. One of those approaches is the “money purchase” formula (which was eliminated for those hired after July 1, 2005). When I was doing some research on this issue a few years ago, I learned that – at least at the time – the majority of SURS retirees were receiving the highest benefit using the money purchase option.

I have written before about the problems I have with the way the Effective Interest Rate is set that is used to accumulate the value in the money purchase calculation, and I will not revisit that accumulation issue here.

Rather, the hidden subsidy I want to discuss today is with regard to how the accumulated account balance is converted into an annuity.

On page 17 of the SURS Traditional Benefit Package guide, it shows that someone retiring at age 65 and 0 months will get $1 per month in income for every $110.40 in accumulated “money purchase” account value. Thus, if your account is calculated to be worth $100,000, SURS will convert that to $905 per month for the rest of your life (=100,000 / 110.404).

For comparison, if one goes to www.annuityshopper.com and look up the price for a single life income annuity with no payments to beneficiaries for a 65 year old male resident of Illinois (as of 6/21/10), you will find that a $100,000 account would buy you only $633 per month in benefits. If you are a woman, it would buy you only $579 per month (because women live longer).

This is an absolutely enormous subsidy relative to private market annuity prices. The SURS annuity conversion rate is 43% higher than the best annuity price available from the private market for men, and a 56% bonus for women.

Now, the sophisticated among you might argue that there are two reasons that SURS can provide better pricing than the market. The first is that SURS does not suffer from adverse selection like in the private annuity market (the phenomenon in which only longer lived people buy annuities, thus driving up the price). But research has shown that adverse selection can account for about a 10% reduction in payouts in the private market. The second reason is that SURS may have lower administrative and marketing costs. These are estimated to be about 5-10% of the value of an annuity in the private sector. So, at best, these two factors can account for about 20 percentage points – less than half — of this difference.

Where does this enormous subsidy come from? It comes from the fact that SURS is pricing these annuities assuming that it will continue to earn about 8% per year in its investments going forward. (That distinguishes it from the use of the ERI to accumulate account balances based on past performance of the SURS funds).

To guarantee 8% going forward is financial insanity. In essence, SURS is promising annuitants the expected rate of return on a stock portfolio, but leaving all the risk with the pension fund (and ultimately either future taxpayers or future pensioners).

Were SURS a private insurance company, pricing annuities in this way would be a pretty good way to ensure a “junk” rating from the rating agencies.

At the bottom of the table on page 17 of the SURS material, it states “this table is determined by the SURS actuary and is subject to change based on the actual experience of SURS.” Because SURS has disclosed that it is subject to change, then SURS ought to be able to invoke more realistic pricing assumptions without violating the impairment clause. Of course, it is probably limited in doing this only for those that have not annuitized yet.

Of course, it is worth noting that the 43% and 56% numbers above would apply only to the income generated under the money purchase approach. In reality, making this change would likely result in benefits being higher under the standard formula (2.2% of final average compensation for each year of service). So the good and bad of it is that neither benefits to annuitants nor the cost savings to the SURS system would be as large as this suggests because the other formula acts as a floor.

In closing, I am not necessarily saying we should cut benefits -  I will leave that to the politicians to decide. But if we want to subsidize our annuitants, let’s make a conscious decision to do so and be very transparent about it so that both they and the citizens of Illinois know we are doing it. It is not good for a democracy to hide the subsidy behind actuarial complexity where we can pretend it does not exist.

8 Responses to “A “Hidden” Pension Subsidy in SURS”

  • Jim says:

    Obviously, the politicians will have to decide, but given that they created the current mess by stealing from the retirement systems’ funding for decades, I’m not sure they are competent or trustworthy to find a solution. Your analysis is interesting, but what would be the impact of full funding throughout SURS history on the ability of the system to sustain Money Purchase?

  • Subhash Bhagwat says:

    So what you are saying is that the private insurers are a reliable and desirable example to follow. If my impression is correct, SURS has, by law, no profit motive. Private insurances have profits as their prime reason for existence. Is SURS really “subsidizing” us? And if it has been doing it how is it that until not too long ago it seemed to have enough surplus? Now, you may point out that the surplus is a result of artificially high interest rate assumptions, as some people have argued. Well, if that were true then the returns on investment reported in their annual reports must be wrong and therefore criminal. Where is all this leading to? Enlighten us with details of your calculations, please!

  • Frank Goudy says:

    Of course there are many ways to determine a proper pay out for annuities.. Comparing them to private companies is one, but only one.

    Bu the author does have a point.. Perhaps an assumed 8% return is historically generous.

    Post war (Jan 1, 1946 to January 1 2010), the DIJA returned a compounded retrun of 6.46%.

    From Jan. 1 1970 to Jan.1 2010 the return was 6.67%.

    During MY employment period when MY money was supposedly invested the return was 9.33%. Thus, during MY employment period the 8% assumed return rate was less than what the DIJA returned.

    I have used the DIJA because it is a pretty solid indicator of economic growth and returns. If a pension system can not do as well over a period of time, then heads should roll.

    So if one were to use the money purchase formula for the period of years that one was employed (in my case 28 years) then the returns could vary- perhaps considerably.
    But given the long haul, it does appear that anything above 6.5% is indeed suspect.

  • Andrew Szakmary says:

    The prescribed rate of interest SURS uses to set its annuity factors (which, by the way, is and always has been 8.5%) has nothing to do with private annuity rates. According to the Illinois Pension Code, it is supposed to be set by the SURS Board to equal the long term expected return on the system’s investments. Thus, unless SURS sincerely believes that its future investment returns will be much lower than its historical returns, which have indeed averaged close to 8.5%, it has no legal and constitutional basis for reducing the prescribed rate of interest. In addition, as I understand it, the prescribed rate of interest must, by law, equal the assumed rate of return that is used to determine the funding status of the system. If they reduce the prescribed rate, then the system will “officially” be even more underfunded, and required state contributions would be even higher than currently.

    Aside from constitutional issues, there are very good theoretical and practical reasons beyond those you mention why SURS annuity factors should NOT track market rates. Obviously, right now, those market rates are very low because interest rates are at record lows, but I am certain that if you look back 30-40 years, you would find that sometimes the market rates are higher than SURS rates, and sometimes lower, depending on the level of interest rates. The key issue you do not seem to understand is that, as a participant in traditional SURS, I am forced to take an annuity at the point that I retire; if I don’t, then I forfeit ALL state contributions and ALL health insurance coverage in retirement. I do not have the choice of taking a lump sum distribution and waiting for a more favorable interest rate environment to annuitize, as would someone who is in a defined contribution plan. Thus, in Finance terms, your arguments completely ignore the value of the timing option which people who purchase private annuities invariably have, and which SURS participants do not have. Another, related, issue is that if the annuity factors are set using market rates and thus fluctuate drastically over time, people retiring several years apart would receive drastically different pensions based on the vagaries of market interest rates. Does this make sense from a public policy perspective?

  • Jeffrey Brown says:

    I’ve written in other posts about the fact that had our legislatures and governors made the contributions that they were supposed to make over the years, the system would be in MUCH better shape. In fact, using the flawed GASB standards that we currently use, SURS would be considered fully-funded. Even if we used the more appropriate discount rates, under which the system would still be under-funded, we would be MUCH better funded than currently. But that is not the point I was making in this post. Rather, I am just trying to point out that the complexity of the current system does everyone a disservice – participants, state employers, taxpayers, the system itself – by providing hidden benefits that are economic costly but that nobody understands. I would make the same point even if we were over-funded …

  • Jeffrey Brown says:

    No I am not saying anything about private insurers being a desirable example. I simply compared pricing, and then I also pointed out that EVEN IF we eliminated all private sector administrative costs AND private insurer profits AND we also assumed that SURS participants have the same mortality characteristics as the general population (an assumption that works against the point I am making), the SURS annuity pricing is STILL subsidized. In other words, it is *better* than “actuarially fair,” better than what a not-for-profit insurer would provide, etc. The reason is that SURS uses an inappropriately high discount rate assumption that treats a risky portfolio return as if it were riskless.

  • Jeffrey Brown says:

    Actually, my main point is that the return on the DJIA is largely irrelevant when one is making *future* commitments to pay out *fixed* liabilities. Promising fixed future payments using a rate of return generated from a risky asset portfolio creates an asset-liability mismatch that *creates* risk. That risk is being borne by taxpayers (and/or future SURS participants if you believe that the cost of the current system is likely to lead to benefit cuts for future participants).

  • Jeffrey Brown says:


    First, I do not necessarily disagree with you about what the Pension code says. I am not a legal expert (a point I have disclosed several times), so I will defer to others on the definitive answer with regards to who has the power to change what. I am simply making a point about proper valuation and disclosure of whatever it is we choose to do.

    You raise an interesting point about volatility of payout rates, an issue that the private market has indeed seen. But there are ways to deal with this. For example, the money could be gradually annuitized over time, thus averaging out interest rate fluctuations. Also, if one shifts gradually into bonds over time – to the point where one is largely in bonds at the point of retirement – then even if interest rates fall, driving annuity prices down, then the total amount of income would not change by much because the value of the bond portfolio would rise due to the declining interest rates. This may be too much to ask the typical person to think about, but I think there is a good case for “automating” some of this, such as by having a life-cycle fund that gradually shifts the portfolio into deferred, fixed (or even better, inflation indexed) annuities over time.

    Oh, and by the way, my calculations in my post actually UNDERSTATE the size of the “implicit subsidy” because it treats the SURS annuity as a fixed annuity, rather than one that rises over time due to COLA adjustments. I realize, of course, that SURS claims that part of the annual employee contribution goes to pay for this increase, which is why I left it out.

    Thanks again for your comments. While it may not always seem so, I actually enjoy your challenges to me in your comments because it does make me think hard about how to clarify my points!