Guest Post: A Recent Retiree Considers Safety-First Retirements

Nothing motivates careful thinking about retirement income strategies than actually being on the cusp of one's own retirement. This is the case for Grinder12, who recently told his story at the Bogleheads Forum in such a compelling way that I've asked him if he was willing to provide an updated version as a guest blog post here. There are two schools of thought about retirement income: probability-based and safety-first. For those considering the safety-first school of thought, the major question is about whether it is worth losing much of the potential upside for greater downside protection. In today's guest post, Grinder12 considers arguments against safety-first (which in this case means building a 25-year TIPS ladder) as well as his responses to those arguments. He's articulated all of these issues very well.

Advisor Cautious About TIPS Ladder Strategy
by Grinder12  
I've recently I felt emboldened to go forward with a plan I've been thinking about for a while, which is investing a lot of our money in a TIPS ladder. But our investment advisor is questioning our plan (they're fee-based and are focused on helping us) and has sent me an email itemizing their concerns about a TIPS ladder strategy. I have drafted a response to the advisor's cautions (which are partially paraphrased), below.  My responses are based on the digging, reading, thinking and crunching I’ve been doing since receiving their email, but I am not as sure of myself as it might appear in those responses, and am wavering in the face of their arguments. I guess I'm really just looking for someone knowledgeable to bolster my confidence in this approach or identify holes in my thinking. Any feedback is welcome. Thanks.

Strategy 
Construct a TIPs ladder using the bulk of our traditional IRAs (78% of our total financial portfolio) to fund our projected living expenses over the next 25 years (We're retirees, age 60 and 70). Then we would use the monies in our Roth (about 15% of our portfolio due to conversions a few years ago) to invest in an aggressive equity growth fund (DFA Global Equity “Fund of Funds”).  Our strategy is to feel safe for the next 20-25 years with our TIPs ladder, then beyond that rely on wife's pension, husband's social security and wife's annuity income (most but not all inflation indexed), plus whatever the Roth has grown to in 25 years. The assumed real rates of return are 0.8% for the TIPS ladder and 4% for the Roth equity. I've assembled a detailed bond purchase plan that matches our TIPS income with our projected liabilities including basic expenses, travel, auto purchases, major house expenses and annuity purchases, with a little extra built in. We have tracked our expenses for 20 years so we have a good idea of what those expenses are.

Advisor Caution #1 
The fact that we are giving up the possibility of the considerable additional gain that equities have historically provided over a fairly long period like we are using.

My Response: 
Yes, there's no question that the expected final portfolio value of this plan is MUCH lower than the expected value of a plan with significantly more equity/less TIPS at the start. But we have no children to leave an estate to and our entire investment focus is on defending our current standard of living into the future against worst case scenarios, not average conditions. I'm sure you'll agree that this is a very different goal than optimizing a portfolio to maximize wealth, based on those long term average conditions. It is also very costly, in terms of foregone expected equity gains. We are just fortunate to be in a position to be able to afford that cost. When I run probabilistic scenarios, I'm not even comfortable with a 95% portfolio survival probability. If I can afford to increase that survival probability further, I will.

Advisor Caution #2: 
Being 20-30% in equity should generally keep us even with or slightly ahead of inflation. We are using TIPs for this but are giving up any net difference equities provide above this. The advisor acknowledges we are giving up the possibility of declines as well.

My Response: 
Giving up the expected net gains of equities is a price we're willing to pay. Avoiding declines at this stage of our life is what our primary goal is.

Advisor Caution #3: 
Being so heavily weighted in TIPS, we risk deflationary adjustments in the TIPS principal.

My Response: 
Yes, the TIPS adjusted principal goes down with deflation. But if the adjusted principal falls below the original principal, you get the higher of the two at maturity, and we're always holding to maturity, so there is a deflation floor. And of course, living expenses are presumably tracking downward with deflation. But they've alerted me to the fact that our April 2028 and April 2029 maturities have a much bigger gap between the adjusted principal and the original principal so they have the most deflation risk. We could replace those with the January maturities of the same years to reduce that risk. In the big picture, I'm thinking it's probably not worth it. But something to think about for the future.

Advisor Caution #4: 
We are extremely heavily weighted in US Government bonds, and in US currency. Yes it is the leading currency currently but it is being rapidly debased and over 25 years other currencies may take on a leading role in a global economy.

My Response: 
If this is referring to potential dollar devaluation and hyper-inflation, am I wrong to think that TIPS being held to maturity would be among the safer places to be in that situation?

Advisor Caution #5: 
In general, having this much of our investable assets in one type of instrument, ie TIPs can lead to unforeseeable problems. They are proponents of diversification to spread the risk of any given asset/asset class being in a major downward trend for extended periods.

My Response: 
That 78/15% TIPS/equity allocation number in 2014 probably doesn't accurately reflect the allocation story here. For example, the TIPS balances will be falling every year, while the equity balances will hopefully be climbing, so that TIPS/equity balance is projected to change a lot over time. Since our income needs are protected, we don't assume any need to rebalance for a long time...the plan allows us to avoid equity withdrawals and basically forget about the Roth equity for 20+ years. If you present value all our income inflows over the plan period, our true asset allocation in 2014 looks like this: Wife's pension 27%, Husband's Social Security 18%, Wife's annuities 4%, Property 6%, TIPS 26%, Equity 16%, Gold 1%, Cash 1%. If you just consider the PV of the financial asset income it's 58% TIPS, 36% Equity, 3% Gold and 3% Cash. So I think the plan is really more diversified and "into" equities than it might look when using simply a first year allocation snapshot.

Advisor Caution #6: 
If we’re not careful we could be paying taxes on inflation adjusted principal on TIPs annually even though our principal isn't received until maturity.

My Response: 
Yes, we need to make sure that we keep the TIPS in the IRA account where this is not a problem. The TIPS "shadow income" problem only applies to TIPS being held in taxable accounts.

Advisor Caution #7: 
For most of their clients, during times of rising interest rates, being in intermediate and long bonds can be very unpleasant.

My Response: 
I have to keep coming back to the point that we’re holding the bonds to maturity, so we aren’t  concerned with how their market value bounces around. That's a key feature of this approach. Now if we had an unfunded catastrophic expense, there'd be more risk of being caught on the wrong side of interest rates. But we are heavily insured for liability, LTC and medical costs. If one of us dies, we'd each lose some income stream but expenses are projected to fall in a matching manner so the survivor is kept whole. I'm hard pressed to think of a scenario that could force us to liquidate at a disadvantage, not that it couldn’t happen. But even if our income/liability mis-matching starts accumulating a difference, it would only be worth a year or two of accelerated redemptions or ladder rebalancing which would have a low interest rate risk. And of course TIPS are less inflation sensitive than nominal bonds.

Advisor Caution #8: 
They feel the biggest issue may be how inflation is measured by the Government and whether it will reflect our own experience of inflation over such a long period.

My Response: 
This is a crucial question. Fortunately we have 20 years of our own expense history to compare with CPI data. When I asked my wife what her perception was of our standard of living today versus 20 years ago, her feeling was that it had increased a little. That's my feeling too. When I inflated our expenses in 1994 to 2014 dollars using the CPI-U, our current expenses are about 1% lower than 20 years ago. If I use the estimated chained CPI-U (0.25%/yr lower), our current expenses are about 3% higher today in real terms. That 3% increase is consistent with our feeling that we're living a little better/spending a little more than 20 years ago. So I'm comfortable that the chained CPI-U reasonably matches our experience of inflation.

Addendum:  That said, when matching liabilities and income it’s important to think through the future inflation issue and try to account for any significant income or expense categories that might not track with inflation.  So I have put this issue under a microscope below.
I’ve already determined from the expense history analysis that the CPI has tracked our standard of living reasonably well, so I’m comfortable starting from the default assumption that our expenses will continue to move with the CPI measure.  From that point, I consider each expense category to see if there might be a reason to assume a different future growth rate as we grow older.  Below is a summary of our major categories of expenses and income, how I think they will change in the future versus CPI inflation, and whether I've explicitly forecast for that category or just assumed inflation.  The thinking is to adjust our forecast where the negative impact could be significant or where the expense is very lumpy, and to not adjust for those expenses that I think might decline, mainly because I don’t know exactly when those decreases would kick in.  It's a conservative approach, but as long as we can afford it, I’d rather be looking for ways to invest excess ladder cash than be looking for ways to get the cash when the ladder income is short.  Of course you can always adjust the ladder (at a cost) or find ways to get the cash so I don’t want to overstate how exact your forecast needs to be.  But the better the forecast, the lower the adjustment costs.  All income streams need to be considered in this process as well, since they directly affect the required funding from the ladder.  For the estimates of taxes, social security and medicare premiums I used ESPlanner. 
 
There doesn’t seem to be a lot of research on how expenses change as we age but here are two relevant links:
 
http://theretirementcafe.blogspot.com/2013/12/does-retirement-get-cheaper-as-we-age.html
 
http://wpfau.blogspot.com/2011/08/reality-retirement-planning-new.html