Friday, February 14, 2014

Build a Floor, Place a Bet

My past four posts, beginning with Unraveling Retirement Strategies: Systematic Withdrawals, have described the four major classes of strategies for funding your retirement:
  • Systematic withdrawals
  • Purchasing a life annuity
  • Floor-and-upside, and
  • Time-Segmentation.
I'll repeat the charts for potential spending with each strategy from my previous posts. Note which strategies have a “floor”, or downside limit for spending, and which don't.

Also, note that the spending lines move to the right on the charts until you run out of money for all strategies except life annuities. For life annuities, you never run out of money and the line moves to the right for as long as you live.

For floor-and-upside, when you run out of money and where the spending line ends is largely determined by your TIPS bond ladder implementation and to a lesser degree by the stock market. It is determined by the stock market and your spending rate for systematic withdrawals and time-segmentation strategies, which is another way of saying that systematic withdrawal and time-segmentation strategies have more longevity risk.

While there may seem to be a dizzying array of alternative strategies for you to choose from,  I view them all as combinations of these four classes of strategies, or “tweaks” of one of them. I also view the four strategies as lying within two axes that plot longevity risk against the possibility of increasing retirement standard of living if investments perform well.
At the top right is the systematic withdrawals strategy with maximum upside spending potential and maximum longevity risk. In fact, longevity risk mitigation is the mere reliance on what has happened in the stock market in the past.

At the bottom left is the strategy of purchasing a life annuity only. Life annuities provide the greatest longevity risk protection—you cannot outlive your money—but zero upside spending potential.

Nearest life annuities, but with a smidge more longevity risk and some upside spending potential, is the floor-and-upside strategy that can provide secure real income for decades. That's still more longevity risk than a life annuity unless you build a very long and inefficient ladder. Floor-and-upside, however, doesn't have the “premium forfeiture” problem of annuities. You always own and control your bond ladder investments.

Time-segmentation lies near systematic withdrawals, but perhaps with different upside spending potential, downside spending risk and longevity risk. That's because the cash and bond allocation for time-segmentation strategies is determined by spending assumptions while systematic withdrawal strategies base bond allocation on how much overall portfolio volatility a retiree can tolerate. The allocations, and therefore their risk-reward profiles, may be different.

Annuities and floor-and-upside guarantee a minimum amount of income throughout retirement. Systematic withdrawals and time-segmentation do not.

Most strategies in the gaps among these four could probably be achieved by combining strategies.

To a large extent, you can determine the best strategy for your household by understanding how much longevity risk you are willing to accept in exchange for increasing your chances of improving your standard of living if the stock market winds blow favorably throughout your retirement years.

Think of having two accounts to invest your retirement savings in: one account guarantees your future retirement income and the other is a bet on the stock market improving your standard of living over time.

The secure account includes Social Security retirement benefits and other pensions. To these, you can add retirement savings invested in bond ladders, life annuities, or both to provide secure lifetime income.

The bet account consists of stocks, bonds and other risky assets that might improve your standard of living in the future if those assets grow a lot, and might not. It is unlikely that you will ever lose all the money in the bet account if it is properly diversified and not leveraged, although markets have lost 90% of their value and more in the past.

Here's how the strategies use these accounts:
  • Life annuities put all of your savings into the secure account.
  • Systematic withdrawal strategies put all of your savings into the bet account.
  • Floor-and-upside strategies fund the secure account with enough capital to generate 30 years or so of safe retirement income before putting whatever then remains of your savings into the bet account.
  • Time-segmentation strategies fund five to ten years or so of spending in the secure account and then invest the remainder of your savings in the bet account.
The first step to selecting a retirement funding strategy is to figure out how much of your future you feel you need to secure and how much you're willing to bet that the stock market will improve your future standard of living. (I wrote about this in Your Retirement Income: Will You Take the Bet?)

The amount of your retirement savings may limit your choices. If your retirement is underfunded, you may not want to risk what little capital you have. You may not be able to afford ten years of desired secure income, let alone thirty. If you accumulated an 8-figure nest egg, on the other hand, you can probably afford to purchase secure income and take a lot of risk.

The size of your Social Security retirement benefit and any other pension you might have will also play a role in your choice. These are two sources of secure retirement income that might be large enough to enable you to bet more on stocks.

For workers retiring today, systematic withdrawals or time-segmentation will be the best strategy to have selected if we are on the verge of a long bull market. If the market performs badly, life annuities and floor-and-upside will turn out to have been the best choices.

(If only we knew.)

Since running out of money before we die is an outcome to be avoided at all costs, in my opinion, we are perhaps better served not by the strategy that will perform best if we guess correctly about future stock market returns, but by a strategy that takes the worst case scenario off the table. That would be purchasing a life annuity or implementing a floor-and-upside strategy.

Floor-and-upside has excellent protection against longevity risk and offers upside potential for our standard of living if we have saved enough to also fund the bet account. And, we maintain control of our capital.

Though floor-and-upside might not be the strategy that best fits your own finances, you'll only regret this choice if your neighbor bets everything on the stock market and is blessed with a raging bull market throughout retirement.


  1. Really enjoyed this series on retirement income strategies Dirk. Thanks for continued good posts. Brad

  2. It seems to me that the asset allocation for what money is left over for "the Bet" in the floor-and-upside approach might be effected by different factors than the asset allocation for someone using systematic withdrawals. Do you have any suggestions for what factors to consider in floor-and-upside?

    1. Great question. It is affected by other factors. Here's how I look at it.

      The reason you allocate part of your portfolio to bonds is to dampen portfolio volatility. With systematic withdrawals, you establish a kind of "mushy" floor based on the worst stock market performance we have seen so far and the probability that it or something worse will happen in the future.

      Once you establish a firm floor with a TIPS ladder, you can take far more risk with your equities. They are no longer serving as both a source of critical spending and a bet that you can have more.

      I consider a TIPS ladder to be part (or all) of our overall portfolio bond allocation.

      Say I decide that with an SW strategy I would want no less than 60% bonds because I can't tolerate more than a 15% loss of wealth in a bear market. If my bond ladder requires about 60% of my total investments, I simply invest the remainder in stocks and I have both a bond ladder to mitigate interest rate risk and an overall 40/60 portfolio that lets me sleep at night.

      If the ladder only requires 40% of my portfolio allocation, I invest enough of the remainder in bonds (funds or laddered) to bring my overall portfolio allocation to 40/60.

      The problem is when I need to invest more than 60% of my portfolio in the bond ladder. Then I have to decide whether the additional upside potential for my standard of living is worth getting rid of some of my floor and investing that money in equities, or whether I would rather have the floor than the upside, in which case I go with the lower stock allocation.

      Thanks for the question!

  3. Dirk,

    New reader to your blog and finding it, along with Wade Pfau's, to be indispensable. Thanks much for sharing your wisdom.

    My question is with regard to where you see rental real estate fitting into your four major strategies. I think it is on the upside half of the floor and upside strategy, but with a very different risk/return profile than a traditional stock/bond portfolio. Let's consider dividend growth stocks (dividend champions or aristocrats) for comparison's sake. My take on dividend growth is as follows:

    Volatility of Capital: High
    Volatility of Income: Low
    Risk of Capital Loss: High and on short timescales
    Risk of Income Loss: Low/Medium -- high principal loss is likely to be coupled with a dividend cut and/or suspension.
    Spending Plan: Spend dividends
    Investment Strategy: Grow income as dividends increase annually. Initial 3% dividend with 7% annual growth becomes 6% yield on cash after 10 years.

    For rental real estate (assume leveraged at high loan-to-value), the profile is like this:

    Volatility of Capital: Low
    Volatility of Income: High (depends on expenses each year), but could be mitigated by diversifying across multiple properties if resources permit.
    Risk of Capital Loss: Low and on long timescales; Insure against total loss
    Risk of Income Loss: Low on average, but with volatility year-to-year
    Spending Plan: Spend cash-on-cash return.
    Investment Strategy: Grow cash-on-cash return with inflationary rent increases. Grow principal with inflationary value increases and mortgage paydown. Initial 2% cash-on-cash grows to nearly 8.5% after 10 years (assuming fixed rate mortgage is initially 50% of rent income and 3% inflationary adjustments in rent and expenses).

    Based on this, my take is that rental real estate belongs on the upside part of the floor and upside strategy, but with lower risk to principal and higher income volatility than dividend growth stocks.

    What's your take?

    1. Rental real estate certainly isn't flooring, since rental income is uncertain, but I would also question how much upside it provides. The long term return for real estate has been just a tad higher than the rate of inflation. I am a landlord and consider managing real estate more like running a business than investing.

      Nor would I consider the risk of loss of capital low, even over the long term. A friend held about $5M in rental properties for 20 years and declared bankruptcy in 2008. (At least no one calls him to plunge toilets, anymore.)

      I find it very difficult to compare an individual rental property with a large company that pays dividends. The latter would likely be far safer.

      Rental real estate has a number of challenges. It's nearly impossible for the typical retiree to adequately diversify. Most properties will be owned in the same geographical area. Much of the profit depends on vacancy rates and expenses. The properties are highly illiquid, especially in today's market. Management is expensive and there are demands on your own time.

      In addition to difficulty diversifying within your property holdings, unless you have an awful lot of wealth, you will find your overall portfolio diversification affected (you'll own too much real estate relative to other asset classes).

      And then there's the toilet problem.

      Sure I can't interest you in a REIT?

  4. I think most retirement withdrawal strategies fail to consider that, in our old age, we will be less inclined and less ABLE to handle our finances. For example, my parents both suffered varying degrees of dementia as they got into their late 70s and beyond. So I feel like my "floor" needs to be really solid by about 75 or so. That's why I plan on building a gradually increasing floor so that after 75 I don't have to do anything on the chance that I'm just not able to.

    It's a hard thing to plan for because it's hard to imagine not being able to make those decisions. But it happens to many of us.

  5. Thanks for mentioning that, Tom. I completely agree. I think one of the things you need to build into a retirement plan is an easier-to-manage late retirement phase strategy.

    This might be needed because we lose some of the mental acuity needed to manage a complicated strategy. Also, in households like mine, one of us is very interested in finance and the other has no interest at all. I have to think about how my wife, for example, would manage our finances if she survives me.

    We've both about decided your ladder suggestion would work best for us.

    Thanks for commenting!

  6. How about hiring a robo advisor and set your asset allocation as needed over time? Not much require for you to do

    1. This comment has been removed by the author.

    2. I think, like many things in life, you get out of it what you put into it.

      If your retirement plan only tells you how to allocate your assets, then you don't have a retirement plan – you have an investment plan.

    3. That's what I was referring to. Use a robo advisor for your investments as opposed to an advisor, most of whom are salesmen. Even fee only who may or may not be CFP's aren't investment experts. They take 1 course on investing to get the designation. I have an attorney who did my estate plan. I don't need an advisor who's a jack of all treades, master of none. I've got my insurances as well.

    4. Also, some robo advisors now have retirement income withdrawal strategies as well.

    5. Steve, there is a lot more to retirement planning than investing and withdrawing from a retirement portfolio. I suggest you read some of my more recent posts.

      Good luck, and thanks for commenting!

  7. Dirk,
    I will, thanks. I've been in the business for 30 years, and read Pfau, Kitsch and all the others who publish.
    I was just sharing my opinion. Maybe those not in my situation could use a legitimate advisor (meaning not a sales guy).

    1. I am totally sympathetic with your argument. There are way too many insurance salesmen, stock brokers and bankers masquerading as financial planners. A CFP is better than no CFP, but if that's their only training it isn't enough. I don't meet a lot of planners that impress me. On the other hand, there are good ones out there if you look for them.

      I also have a computer science degree and know enough about expert systems to have significant doubts about robo-planners.

      The future post I'm referring to will probably be on the blog in about two weeks. I talk about things that retirees should consider beyond sequence risk. Let me know what you think.

      Thanks for writing.