Is it possible to achieve higher safe withdrawal rates, compared to a stock/bond portfolio, from a 100% bond portfolio? With bond prices at historical highs? That’s what I wanted to find out. I started thinking in this direction after my look at bottoms up retirement planning and calculating what low real rates are actually needed to achieve retirement success. This is mainly because what really kills SWRs in risky portfolios is negative withdrawals early in retirement and inflation as I discussed in my ‘Insist on 1966′ post. So I thought it would be useful to look at what the lowest risk retirement model is and what are the SWRs (safe withdrawal rates) from such a model. Then we can compare all other models, using riskier assets, to this minimum risk model and decide on the tradeoffs. The results are rather surprising.
I’ve always been somewhat surprised just how low the recommended SWRs are. The 4% SWR rule for a 30 year retirement that has proven to work historically for even the worst retirement periods has a worst case 30 yr period return of about 7%. That’s a 4% withdrawal rate from 7% returns! Pretty low when you think about it. Basically, because of negative returns, particularly early in retirement, what they call sequence of return risk, SWRs are much lower than the worst case portfolio returns. But it gets even more interesting. Lets look at what the SWR would be for a portfolio if you could get guaranteed 0% real return. Then lets compare those ‘ideal’ SWRs to those from risky portfolios. In the table below I show such a comparison. BTW, the 0% real return SWRs are simply 1/retirement period.
The two highlighted SWRs are the most commonly recommended asset allocations in retirement; 60/40 stock bond allocations or 70/30 stock bond allocations for a 30 year retirement period. As the table shows you don’t get much of a benefit in SWR from risky portfolios until you go beyond a 25 year retirement period. At the same time I think it would be surprising to many investors that you could have an SWR of 3.33% for a 30 yr retirement period with a 0% real return. At this point you should be asking, ‘great, but where in the world can you get a guaranteed 0% real return?”. Turns out we can get pretty darn close with a laddered bond portfolio of US Inflation Protected Bonds (TIPS). TIPS offer a guaranteed return of principal and inflation protection of that principal which takes the two biggest retirement risks off the table from the start. And beyond zero real returns, so far over their history, US TIPS have offered positive real returns over the entire term structure. The next thing to I tool a look at is what SWRs are achievable with a portfolio of US TIPS at today’s prices?
To calculate the SWRs for a laddered portfolio of bonds, in this case US TIPS, takes a little doing. I started with prices and data on the available US TIPS from the WSJ quote page. I used prices at the end of day on Sept 23, 2013. Download those into an excel spreadsheet and build a model for various time periods with a target of having equal payouts for each year of retirement, basically keeping real standard of living the same, just like the traditional SWR models. Fortunately, you can use the solver function in excel to do the hard work for you. The results are shown in the table below and you can access my spreadsheet here and play with the model for yourself.
So, for retirement periods of 20, 25, and 30 years, you get higher SWRs from a 100% US TIPS portfolio than from a 60/40 stock bond portfolio! Let that sink in for a minute. And that is with the historically low real rates offered today. This again shows just how detrimental negative returns are for retirement portfolios and is what keeps SWRs for risky portfolios so low. And if you, like many others, think future returns and SWRs will be lower than the past then these numbers should be even more compelling. At the minimum these super safe SWRs should be used as a benchmark when considering riskier asset allocations.
What about the downsides? Of course there are always downsides or drawbacks to any approach. For me, the biggest one of a 100% US TIPS portfolio is that there not as much upside as there is with a risky portfolio. If you hold the bonds to maturity the indicated SWR is all you get. No adjusting SWRs higher if markets perform well. If real rates only go higher from here the bonds also need to be held to maturity in order to protect principal. On the other hand, if/when markets experience another major/correction crash, real rates will undoubtedly go much lower, the TIPS could be sold with nice capital gains, and you could then implement a risky 60/40 portfolio at much lower valuations. So there is the potential for some upside. The other downsides are that the portfolio is a bit harder to implement that just buying ETFs for stocks and bonds, and TIPS do bring up some tax complications for investors. None of these are insurmountable and if people are interested I can provide some more info in the comments to this post.
In exchange for potentially less upside in total wealth you get, most importantly, a much higher probability that your portfolio will last throughout your retirement period. No worrying about stock market fluctuations and the like. In other words, greater piece of mind. Of course, it doesn’t have to be all or nothing. A TIPS portfolio such as this can be used as a base for a retirement portfolio and combined with risk assets (mainly stocks), to provide upside for the investor. Or you could even add other bonds to the TIPS ladder for some bond upside. For the investor who wants to take even more risk off the table a TIPS portfolio can be combined with a deferred income annuity to take longevity risk complete off the table.
In short, its very possible to build a retirement portfolio out of 100% US TIPS that has higher safe withdrawal rates than those provided by riskier portfolios. This can be a great option for conservative retirees or those who do not want to take any stock market risk.