March 25, 2020
In my post last week, I looked at how the 2020 Bear Market will impact folks saving for (early) retirement. But I deferred my recommendations on how current retirees will optimally adjust to the new realities. So, here we go, a new installment of the Safe Withdrawal Series, now 37 posts strong!
Nothing I write here today should be shocking news to people who have read the other 36 parts, but having it all summarized in one place plus some new simulations and perspectives is certainly a worthwhile exercise. In a nutshell, I argue that if you’ve done your homework before you retired, not even a bear market, not even this bear market will derail your retirement. Depending on what approach people chose, some retirees might even increase their spending target now.
Let’s take a look…
Recap: How bad is the current bear market for retirees?
I know that the S&P 500 daily all-time high occurred on February 19, 2020, but looking at the month-end values, just as I do in my withdrawal rate simulations, the stock market peak was actually on 12/31/2019. Since then, the market dropped by about 30% to the low-point on March 23, and about 25% to yesterday’s closing on March 24.
If you heeded my advice here on the blog and kept some bonds for diversification you should have significantly cushioned the fall. But not all bonds are created equal. In the chart below, I plot the performance of a few fixed-income ETFs:
- Intermediate Treasury Bonds (e.g., the iShares IEF ETF). This is the asset class that I use as my bond allocation in all of my simulations.
- The U.S. total bond market (e.g., the BND ETFs). This is the index of all U.S. Treasury bonds (all maturities) plus investment-grade corporate bonds. (Side note: Investment Grade = Credit Ratings AAA all the way through BBB)
- U.S. Investment-Grade corporate bonds (e.g. the LQD ETF). Drop the government bonds and concentrate on corporate bonds, but again only the higher-grade bonds.
- U.S. High-Yield Bonds a.k.a. Junk Bonds (e.g. the HYG ETF). Instead of the investment-grade, use the lower-rated bonds with a higher yield but also more risk. Credit Rating BB and below!
- U.S. Preferred Shares (e.g. the PFF ETF). Not exactly bonds. Much riskier but also potentially much higher yields than a bond!
- I also added the S&P 500 index ETF (iShares IVV) for comparison.
I was surprised about badly the non-Treasury ETFs performed. Not a pretty picture, especially the major nosedive in March. In March, most of the riskier bond ETFs are down roughly by the same percentage as the equity index. Not much diversification from LQD, HYG and PFF in March!
So, let’s look at how a 75/25 portfolio would have performed year-to-date through March 24, see the chart below. Not a big surprise here, the more risk you took with your fixed-income allocation the worse your performance. I also distinguished between rebalancing vs. not rebalancing, and you can clearly see that rebalancing would have slightly hurt you when equities keep going down and IEF keeps going up, i.e., you shift money from Treasurys with positive momentum to equities with negative momentum.
Also, the charts didn’t consider withdrawals yet, so a retiree who started withdrawing money on 12/31, 1/31 and 2/29 would be looking at about an additional drop of about one percentage point!
In any case, if you had kept enough safe Treasury Bonds in your portfolio then the drop in the portfolio still looks OK. If you had replaced too much of the safe assets with higher-yielding assets you could have lost a lot more. But remember, when I debunked the infamous”Yield Shield” approach I very explicitly warned of chasing after higher ETF yields, especially on the fixed income side:
“[T]he three bond alternatives to the low-yielding IEF all have higher correlations with the equity indexes. Not only that, moving to higher and higher yields (IEF to [BND] to LQD to PFF) you increase the equity correlation in exactly that order!” (Part 29)
“We can say with near-certainty that the fixed-income portion of your portfolio will behave very similarly to the 2008/9 episode once the next big bad bear market comes along. In other words, in the next big recession, higher-yielding corporate bonds will underperform the safe and low(er)-yielding government bonds.” (Part 30)
“The fixed income recommendations of the Yield Shield are just plain bad. The credit premium that gives you the higher yield will backfire in a bear market because this is essentially backdoor equity beta.” (Part 31)
So, how would a retiree who had the misfortune to retire on 12/31/2019 have fared in this bear market? That depends on how they approached their retirement. Let’s take a look at several methods:
Method 1: Fail-safe withdrawals based on historical simulations
Let’s go through a simple numerical example to be used in the Google Sheet as introduced in Part 28.
- A $2,000,000 initial portfolio, 75% Stocks, 25% Bonds (Interm. U.S. Treasury)
- A 30-year horizon
- The retiree plans to have one quarter ($500,000) of the initial capital at the end of the horizon (in real, CPI-adjusted $!!!)
- No additional supplemental cash flows for the duration of the 30 years
The reason I like this example is that it’s a nice generic example for both traditional and early retirees. You might be a traditional retiree at age 65 and plan to supplement your Social Security and pension with your portfolio over the next 30 years and plan to leave a small bequest to your kids. Or you might be an early retiree at age 40, and you try to bridge the 30 years until you receive (full) Social Security benefits and pensions, so you plan to have 25% of the initial capital left at that time.
To implement this into the Google Sheet, here are the parameters in the main sheet
And here are the results, specifically, the historical fail-safe withdrawal rates (by decade):
It looks like the 4% was not safe in this example. 3.58% was the true safe withdrawal rate that survived even for the worst historical retirement cohorts (i.e., September 1929 and the mid-to-late-1960s).
If we now assume you had used this very, very conservative withdrawal rate, where would you stand on March 24? Here’s the calculation, see the table below. Wow, this bear market did a trick on the portfolio. It’s down by $350k or about 17.5% in real terms.
If you keep withdrawing that same retirement budget of 3.58% of the initial amount this now represents a 4.33% annual withdrawal rate relative to the 03/24/2020 portfolio value. Whoa, danger zone! We’re withdrawing more than 4%! A 4.33% withdrawal rate had a historical 13.6% failure rate!
But is that so bad? Do we really have to reset our withdrawal amount to 3.58% of that new $1,651,856 portfolio value on March 24, 2020? That’s $4,928 per month, more than $1,000 lower than the initial withdrawal.
Fortunately, that’s a fallacy. So, I like to go back to my Google Sheet and now recalculate my withdrawal analysis for someone who has a $1.65m portfolio, a 357-month horizon, a $500k final asset target, and the same 75/25 asset allocation. Notice that the final asset target is now a bit larger than 25% because 500/1,652=30.3%!
So, I punch in those parameters into my Google Sheet and now look at the Fail-Safe withdrawal rates. But after this steep equity drop we don’t look at the overall fail-safe rate, but conditional on the equity market having dropped 24% since the peak. Then, not only is the current effective WR of 4.33% safe in the historical simulations. You could even increase the withdrawals to 4.52%, or about $6,222 per month. That’s 4% above the initial withdrawal amount. How is that possible? Very simple, if you calibrate your initial withdrawal rate to the historically worst-case scenario, then as long as the current market conditions don’t reach the depths of the 1930s or 1970s, you can slowly move up your withdrawals.
So, the lesson here is that the 4% Rule fails only if we’re at or very close to the equity market peak. After we’ve already fallen by 20+% we can be a lot more generous with our safe withdrawal rates. Historically, you can use withdrawal rates well in excess of 4% if the stock market is down as much as today!
Method 2: The naive 4% Rule (and scary “5% Rule and crazy 7% Rule)
Sometimes it’s instructional to go through an example illustrating how not to do it. Let’s assume you’d never heard of my blog and you had just followed the naive 4% Rule. Or the really bad advice out there from the personal finance clowns advocating that you can use even a 5% or even 7% initial withdrawal rate “if you’re really flexible.” Let’s study how that would have worked out for the 12/31/2019 retiree, see the table below.
- Not surprisingly, the portfolio values differ by only a few thousand dollars because there were only 3 withdrawals so far in December, January and February.
- But there is a big difference in the current effective withdrawal rates. Your 4%, 5% and 7% initial withdrawal rates have now grown to 4.85%, 6.08% and 8.55% as of late March.
- If you were to recompute the withdrawal analysis now (using a horizon of 357 months and an unchanged $500k final asset target) then those new effective withdrawal rates would have some pretty hard to digest failure probabilities in historical simulations: about 11%, 52% and 82% failure rates. (and yes, these failure probabilities already take into account the equity drawdown. The unconditional failure probabilities would have been significantly larger!)
- If you now wanted to adjust to the new realities and pick the historical fail-safe withdrawal rate calibrated to today you’d have to lower the withdrawal amounts by between 7% and 47%. Not everybody will be flexible enough to do that! Though, tightening the belt by 7% seems doable, so the naive 4% Rule wasn’t that bad. But belt-tightening under the scary 5% Rule and crazy 7% Rule will be harder to stomach.
Method 3: Sustainable withdrawal rates based on the Shiller CAPE
If you remember Part 18, I illustrated why I’m really intrigued by CAPE-based withdrawal rules. Let’s go through a simple example with one such rule, where we set the withdrawal rate to a+b/CAPE and a=1.5% and b=0.5. I’ve previously even advocated more aggressive rules with a=1.75%, but let’s be a little bit more cautious for this example.
So, every month-end we look at the CAPE and calculate a “provisional” withdrawal rate. For example on 12/31/2019, the CAPE was 30.85 and thus the SWR was:
CAPE-SWR = 0.015 + (0.5 / 30.85) = .0312 = 3.12%
Provisional because the CAPE withdrawal rates are calibrated to target capital preservation, while in this example we’re happy to deplete 75% of the capital. To translate the capital-preservation WR to a capital-depletion WR we do a simple calculation, very similar to the Bogelheads VPW approach. In Excel, we use the PMT function to compute the monthly withdrawal with capital depletion as:
Which is just above $7,700 a month on 12/31/2019 or 4.62% of the initial capital. Then you iterate this forward with the changing portfolio values and CAPE ratios, and also a shrinking horizon (though shortening the horizon from 360 to 359 and 358 and 357 months makes very little difference). The results are below:
- Even though the portfolio drops by a total of 17.6%, the monthly withdrawals drop by 11.6%, so significantly less than one-for-one. About 0.65% for every 1% in the portfolio drop.
- Again, the reason is the higher expected portfolio return in light of more attractive equity valuations after the large drop that brings equity multiples more in line with historical averages.
So, in a nutshell, if you use the CAPE-based rule your portfolio took a hit, but because of the lower CAPE ratio and thus the increased equity expected return you can now withdraw a larger percentage. A larger percentage of a smaller portfolio will still work out to be a net cut in your retirement budget. So the CAPE-rule does require a certain degree of flexibility. It’s because the CAPE-based withdrawal rate is calibrated not to the worst-case scenario, but the average-case scenario. So, you can afford a higher initial withdrawal rate than the fail-safe but you also have significant downside risk. It’s for the more flexible retiree!
Is now the time for retirees to shift to 100% equities?
This is a question I’ve gotten several times now via email. The idea here is that, if you’re sitting on a nice big fat bond cushion now, say 25-40% of your portfolio, wouldn’t it be tempting to shift those bonds into stocks now that the market is so beaten down? I don’t blame you for an itchy trigger finger. If the worst is over and this current 24% drop is all we have to suffer and the market will recover soon, then maybe now is the perfect opportunity to time the market.
A word of caution, though. If we look back at previous bear markets, the initial drop is rarely the trough of the bear market! Case in point the 2000 stock market peak and the bear market that followed. In the chart below, notice how the trough was not reached until late 2002. There was even a point in the spring of 2003, 2.5 years(!!!) after the start of the bear market, where it looked like we’re going hit another low point. In other words, even the post-9/11 panic selling was not the trough. The uncertainty about the weak recovery and talk about a “double-dip” recession caused further declines even worse than 9/11!
Will there ever be a good time to go all-in? Let’s plot the historical fail-safe withdrawal rates for the numerical example above (360-month horizon, 25% final asset target) for different asset allocations (75/25, 90/10, 100/0) as a function of where the S&P 500 sits relative to its recent all-time-high. It turns out that historically, 75/25 still performed better unless the S&P500 was more than 50% below its recent peak. Even then, you’d prefer a 90/10 over a 100/0.
In summary, I don’t want to hold you back when you think this is a great market timing opportunity. Maybe you’re a U.S. senator with inside information and you know things will turn around soon. Make sure you share your info from your secret Senate meeting with your old buddy Big Ern! Use a prepaid phone, though! 🙂
But for everyone else, this could backfire! Even with the nice bounce in the stock market on March 24, there’s no guarantee that we’re out of the woods yet! There have been too many false rebound signals in history!
Clarification (3/25/2020, 8:15 AM PDT): This analysis is for retirees. I’ve pretty consistently advocated for retirement savers to be at or close to 100% equities. If you’re still saving for retirement and you shift into 100% stocks now, sure you also face the risk of a renewed meltdown. But savers would then Dollar-Cost-Average. Retirees would face Sequence Risk. So, in summary: for savers, 100% can work pretty well, for retirees 100% equities seems way too risky!!!
Talking about 100% equities…
While we’re at it, here’s a quick additional word about 100% equities in retirement. I know that many of my fellow FIRE bloggers maintain a 100% equity allocation in retirement. For everyone who retired, say, 5 years ago, that turned out to be a great decision in hindsight. Even with the big drop so far, you will still be far ahead of someone who kept a more conservative allocation. Congrats and good for you. You deserve the extra money as compensation for the volatility we’ve been through. But please don’t advertise this as something that other retirees should replicate! Maybe put this disclaimer on your blog:
Disclaimer: I’m what you call a professional personal finance blogger. If you don’t have income from a blog and/or from your spouse’s six-figure income you probably don’t want to try this 100% equity allocation in retirement yourself!
The reason is that your blog and/or your spouse’s income is effectively like a bond allocation. If you were to call the sum of those discounted cash flows an implicit bond allocation you’re no longer at 100% equities. You might be down to 50% equities! In contrast, for people who are fully retired – both spouses – and who don’t have a blog or podcast, or other income, going through this Bear Market with 100% equities is really downright irresponsible!
Oh, and by the way, just to clear up any ambiguities: Personally, I can’t live off my blog income alone. Advertising and affiliate income generate about 10% of our retirement budget. Even if I could fund my entire retirement from my blog income, I would never recommend my readers to hold 100% equities.
If you’ve planned your retirement well, there’s nothing to worry about (yet). If you tailored your withdrawal strategy to withstand the worst market events in history then as long as the current Bear Market is not getting worse than, say, the Great Depression or the 1970s then you should be OK. In fact, quite the opposite, if 2020 turns out to be less dramatic (depth and length!) than the Great Depression and the 1970s, you’ll likely be able to walk up(!!!) your withdrawal amounts once the dust settles. In contrast, if you were careless and listened to some crazy advice about how you should just wing it with a 5% or even 7% withdrawal rate, then, good luck winging it! (not that I believe any readers here fall into that category!)
And yeah, I know, some people will argue that this could be worse than the Great Depression. We will certainly see some ghastly Q2 GDP numbers. But the Great Depression took more than a decade to recover. This will hopefully be a temporary disruption in 2020 with a decent restart, maybe as early the second half of 2020.
Hang in there, everybody and stay healthy!
Thanks for stopping by today! Please leave your comments and suggestions below! Also, make sure you check out the other parts of the series, see here for a guide to the different parts so far!
Title Picture Source: Pixabay.com