Monday, October 29, 2012

“Adaptive Investing: A Responsive Approach to Managing Retirement Assets”

An interesting article in the new fall issue of the Retirement Management Journal is Sam Pittman and Rod Greenshields’ “Adaptive Investing: A Responsive Approach to Managing Retirement Assets.” Unfortunately, new issues of the journal are not available online, but you can seek a working paper version of their article here.

Both authors are from Russell Investments, and this article updates readers on the Russell Investments approach to retirement income which I introduced here back in April when reviewing the excellent, Someday Rich: Planning for Sustainable Tomorrows Today (Wiley Finance)

The Russell Investments approach is sufficiently important that it is worth re-visiting. They think in terms of three financial goals for retirement: a low risk of outliving assets, consistent income, and financial flexibility. Sustainability, Predictability, and Liquidity.

Funded Ratio Management

To meet these goals, Russell Investments suggests thinking in terms of funded ratio management, which may also be known to readers as asset-liability management or liability-driven investing.

First, a retiree calculates their assets as one total number. This includes the financial portfolio, as well as the present discounted value of other potential income sources, including Social Security, pensions, annuities, part-time work, etc.

Next, a retiree calculates their liabilities as one total number. This could include current debt such as a mortgage or loan balance, but mostly what we need here is the present value of lifetime spending needs.

Next, divide assets by liabilities. This is the funded ratio. If it is 1 (or 100%) then sufficient assets are available to cover liabilities. If it is greater than 1, a retiree enjoys the flexibility which comes with having more than needed. If it is less than 1, a retiree is underfunded. Options here include reducing spending plans, working longer, or making a Hail Mary pass and taking on more risk with one’s investments. Regarding the last point, Russell strongly discourages this, arguing instead that rather than taking a chance that the funded ratio will get even worse, retirees should plan to spend less and not add more risk.

Assets, liabilities, and the funded ratio will fluctuate over time in response to portfolio returns, interest rates, inflation, or unexpected spending needs.

Managing Longevity Risk: The Value of Delaying Annuitization

Retirees must deal with uncertainty about their longevity. They can insure this risk through annuitization. But annuitization reduces the flexibility goal for their assets. There will not be excess funds available for other purposes. So rather than annuitizing right away, Russell suggests avoiding annuitization for as long as possible but keeping it in the background as a potential option.

As long as the value of a retiree’s financial assets stays larger than the cost of annuitizing a desired lifetime remaining income stream, then there is no need to annuitize. But the option is still there.

In practical terms, they way they suggest to go about doing this is to set aside assets to cover spending needs over the next 10 years (updating each year to maintain a 10-year horizon). Then, manage remaining assets so that they retain sufficient value to buy an annuity to cover lifetime spending needs after the 10-year horizon. If the value of those assets start to drop below the value of the liabilities (represented as the cost of the annuity to fund those needs), then seriously consider buying an annuity, or come to terms with the idea of lowering future spending.

Adaptive Asset Allocation

One other aspect of the Russell Investments approach is to adjust asset allocation in response to changes in the funded ratio. This connects asset allocation to the liabilities which must be funded by those assets. The stock allocation should be the lowest when the funded ratio is 100%. As William Bernstein says, stop playing the game if you’ve already won. As the funded ratio increases above 100%, a higher stock allocation can be justified for the surplus assets which exceed liabilities. Again, as the funded ratio falls below 100% the mathematical solution is also to increase the stock allocation, but they suggest not to do this. Rather, reduce liabilities so that the available assets are sufficient without upping the risk.

By adjusting the asset allocation in this way, it should be possible to better protect oneself from seeing their funding ratio fall below 100%. But do be careful about taxes, trading costs, and portfolio turnover. I haven’t seen the math behind this part, but I do wonder about the implications that one is buying high and selling low when overfunded. I wonder if the asset return assumptions guiding this part of the analysis account for mean reversion or whether they are independent and identically distributed. Mean reversion would work the other way, suggesting to reduce the stock allocation as the overfunding level grows.

At any rate, the adaptive asset allocation part is about making marginal improvements to the outcomes. The heart of the material is thinking in terms of the asset-liability model and managing the option of delaying annuitization. Those are two important lessons to incorporate into a retirement income plan.


  1. While I understand the intent behind the 10 year time horizon, I'm not sure that I agree with that horizon as a choice. There are two reasons, to me, that it seems to be out of sync with life events:

    1) It may not be appropriate to incorporate what is, for the most part, intended to be a relatively static period of life. The authors allude to as much when they refer to mobility changes, health changes, martial status changes, etc.; however, depending on the age of retirement and the living situation that the retiree finds himself/herself/the family in, the living situation may alter dramatically within a few years. I'd use it as a rule of thumb, but not as a hard and fast rule, as sometimes the horizon may need to shrink.

    2) If someone has access to a large sum of money designed to last them for 10 years, they may spend more up front because of psychological biases. The denomination effect ( shows this in a small example, but it may be necessary to implement some sort of spendthrift plan to prevent psychological biases from justifying exorbitant purchases ("I NEED that 138" flatscreen TV") simply because the money is there. Look at how the client got to that situation. If it was through wise saving and spending, then there's no worry, but if it was through, for example, an inheritance or winning the lottery--yes, I know, an extreme example--then that approach may merit consideration.


    1. Jason,

      Thanks. I'm not sure that I see how the approach here would cause unique problems that would not exist in a more standard framework with a 30 year horizon or what have you. Spending needs may change over time, but that is supposed to be incorporated at the beginning. It is not just that you plan for only 10 years and not worry after that. As well, you may always have the temptation to spend too much. Are you suggesting that this is a reason to annuitize from the beginning in order to avoid that temptation?

    2. I think that you and Dirk captured it below. Take the need money to eat assets off of the table through an annuity and leave the rest for either dealing with shocks (e.g. the Big One) or lifestyle adaptation. It just seems a little risky to think that you can catch the timing if your asset base is shrinking to know when to buy the annuity. Remember, we hate to sell losers because of the endowment effect. Better, to me, to remove that temptation altogether.

  2. "As long as the value of a retiree’s financial assets stays larger than the cost of annuitizing a desired lifetime remaining income stream, then there is no need to annuitize. But the option is still there. "

    Imagine a retiree (or near-retiree) in early 2007 with adequate resources (but just adequate) set aside to purchase an annuity to fund a desired lifetime income stream. Just three years later, his portfolio is halved and interest rates are near zero, reducing annuity payouts twice.

    His problem is his inability to accurately forecast future portfolio values and interest rates, one we all share.

    Bernstein calls this "continuing to play the game after you win".

    Even if you invested the "adequate amount" of your portfolio in TIPs instead of stocks, you still face interest rate risk for the payout. The cost of that "desired income level" at any given time depends on future interest rates when you will need to purchase the annuity, which is, of course, unknown.

    You could purchase the annuity as soon as you knew you could afford the desired income stream to avoid both risks, but as you say, this costs you flexibility. This strategy meets their first two objectives, but not the second.


    1. Dirk, you are hitting on an important point. I could imagine that people who were fully funded in August 2008 would have found that their funding status plummeted over the next few months, that annuity rates may simultaneously bouncing around, and that the person is so shell-shocked from the market losses that they missed acting to lock in their needs. This sort of approach would need to be automated in some way with software that automatically checks the portfolio balance and annuity rates and provides a warning if action needs to be taken. Otherwise, people might end up missing their chance.

  3. I really like that perspective on annuities. Essentially, hold off on the annuity for as long as possible, and when you do, the monthly income will be much higher.

  4. You nailed the essence of the problem in citing Bernstein. The flexibility objective for most people is probably subservient to the sustainability and predictability when push comes to shove. For the person who has "won", maybe a combination of TIPS ladder till an attractive point of annuitization would work? The delay of annuitization is appealing since the mortality credit rises with age, does it not?

  5. $400 and Paid $1200 in one hour
    Giant investment is a private investment group operating on the world securities and currencies markets since 2002. Having started as a small investment group operating only on the United Kingdom stock market, we have gradually expanded our investment activities on all major financial markets of the world. Our representatives operate all over the globe thus providing us with a big picture view on the world of finance.
    Test Plan * 300% after one hour
    Standard Plan * 500% after one hour
    Premium Plan * 800% after one hour
    100% of funds get risky free
    Don't need to register. That can be safe from hacker
    Stable income
    Return principal after the plan completion

    Invest Now

    Principal Guaranteed