100% Stocks in the Retirement Phase?

[Radagast]

I am adjusting my thinking a little bit. Instead of "at least 15% bonds" perhaps the more important of my original guidelines was the "not more than 50% in US stocks" rule. It is then up to the individual to decide what the remaining 50% should be, international stocks, real estate, bonds, bullion, or something else. Once you go past 50% US stocks there seems to be a clear case of not only more risk, but an increasing chance of actually worse returns. 50% US, 33% international, 17% bonds is definitely a better allocation than 100% US stocks, but it is not that clear for 50% US / 50% international stock.

I still sympathize more with the slice and dicers, but within the next 12 years (probable FIRE time for me) I may come around to a 100% global stock portfolio, more probably with 10% I-bonds for bad times.

On a related note, I am amazed how little improvement a total bond index fund had on outcomes over the 1972-2016 period. If 100% US stock had a 3.2% safe withdrawal rate, no amount of total bond took that over 3.3%, and in almost every case it led to less money at the end. It was even worse for US+International, with no improvement in safe rate of withdrawal, but noticeably lower returns. I was already vaguely aware of that and would more generally recommend 10% long term treasury bonds and 10% I-bonds instead, but I have had the point driven home for me. In fact I am going to be potential flame bait and say that calling gold a "bond" and using 5% of that, 5% treasury bonds, and 5% I-bonds is a far more effective "deep risk fender" and should perhaps be used in place of a total bond index fund. I notice they are all still uncorrelated with stocks :).

[mathjak107]

my portfolio has always been dynamic . it is never set allocations  in stone .

as the big picture evolves the investments evolve .

so while 50% equity is my retirement target   as i will be 64 and wife 66  right now we are 35% equity . but 1/3 the bond budget is a high yield fund which acts as a proxy for stocks right now . it has better value , higher returns and only .56 beta compared to the s&p 500 .

that will change over time ..  same apply's if bond rates rise . the bond funds i do own will change to different types of bond and income funds that stand up better in rising rates or rising inflation .

it was a lot easier to buy and die when we were in a bull market for bonds since 1980 except for what were some speed bumps to where we are today  . when bond rates rise things may not work out as well using a buy and die strategy if you have a bond heavy portfolio that is all interest rate sensitive . .

[Radagast]

Based on thinking and discussion in this thread, the following are my updated retirement allocation guidelines in order of importance:
1. At least 50% of portfolio should be stocks.
2. Not more than 50% of portfolio should be US stocks.
3. At least 10% but not more than 40% should be bonds.

Those are the ones I think are most important. If you stray from those ranges, I think you are taking some combination of a lot of unnecessary extra risk and/or greatly lowering expected returns.

Some people argue that bonds will only drag you down in the end. I think there is a 90% chance those people are correct, which I why I think at least 10% bonds is a good idea. Additionally, most backtests for bonds are very simple because that is the only option. If you expand "bonds" to include FDIC certificates of deposit, I-bonds, longer term bonds, or even gold bullion (this seems a lot like the long term bonds issued in many parts of the world now, even the US) you can probably come up with something that has a better chance of being useful in more situations. Take the CD's for example. If you redid the backtests using CD's with a 1% higher yield than treasury bonds, that would have shown bonds to be a lot more useful.

Upgraded aspects of my OP which I still agree with but don't feel are as critical now:
- The optimum allocation seems to be in the area of 45% US stocks, 30% International stocks, and 25% bonds, plus or minus 10% to each. The optimum 3:2 ratio of US to international was fairly persistent.
- If you want a higher ending $$$ value similar to 100% stocks, do not eliminate bonds. Instead, keep the bond allocation and take more risk on the stock side by including, for example, more emerging markets stocks, small cap stocks, and value stocks in accordance with the first three guidelines. If stocks generally have higher returns than bonds because they have more risk, then riskier areas of the stock market should also continue to have higher returns, though to a lesser degree.
- Use https://portfoliocharts.com/portfolio/financial-independence/ to make sure your portfolio supported at least a 3.7% sustainable past withdrawal rate.

[arebelspy]

Based on thinking and discussion in this thread, the following are my updated retirement allocation guidelines in order of importance:
1. At least 50% of portfolio should be stocks.
2. Not more than 50% of portfolio should be US stocks.
3. At least 10% but not more than 40% should be bonds.

Solid.

I can't fault anyone playing within that range to their comfort level.

[kenaces]

Today I watched a presentation by Rob Arnott and he mention the 50bps benefit to rebalancing and I also stumbled across the below quote on wealthfront.com:

"Analyses by David Swensen, the Chief Investment Officer of Yale University, and Burton Malkiel, Emeritus Professor at Princeton University and Wealthfront CIO, with Charles Ellis, Founder of Greenwich Associates, found that over two separate 10 year periods rebalanced portfolios earned an average of 0.4% more per year versus portfolios that were not rebalanced.*

Rebalancing will not always increase returns over all periods, but it will always reduce your portfolio risk over time and ensure that your portfolio remains consistent with your target allocation.

*Sources: David Swensen, Unconventional Success (2005) 195 & Burton Malkiel & Charles Ellis, The Elements of Investing (2013) 69."

[mathjak107]

most time frames rebalancing to bonds hurt performance . generally taking money from the asset with greater long term growth potential and putting it in one of lesser growth potential cuts gains .

 except for those with volatility aversion i never saw the logic during the long term accumulation periods of mitigating temporary downturns in the short term by permanently giving up gains in the long term  .

to get better returns doing that you better be a good market timer and that rules most of us out .

[kenaces]

most time frames rebalancing to bonds hurt performance . generally taking money from the asset with greater long term growth potential and putting it in one of lesser growth potential cuts gains .

 except for those with volatility aversion i never saw the logic during the long term accumulation periods of mitigating temporary downturns in the short term by permanently giving up gains in the long term  .

asset allocation is really the only free lunch - more return with less risk

https://www.portfoliovisualizer.com/backtest-asset-class-allocation#analysisResults 

Above is just a quick example of simple backtest comparing 90/10 to ivey5 and swenson portfolios. 

[Interest Compound]

most time frames rebalancing to bonds hurt performance . generally taking money from the asset with greater long term growth potential and putting it in one of lesser growth potential cuts gains .

 except for those with volatility aversion i never saw the logic during the long term accumulation periods of mitigating temporary downturns in the short term by permanently giving up gains in the long term  .

asset allocation is really the only free lunch - more return with less risk

https://www.portfoliovisualizer.com/backtest-asset-class-allocation#analysisResults 

Above is just a quick example of simple backtest comparing 90/10 to ivey5 and swenson portfolios.

No.

Asset allocation is considered a free lunch, as you get increased risk-adjusted returns. Not "returns". As mathjak107 said, this makes perfect sense. Think of it this way, if you don't rebalance, your portfolio eventually has a higher and higher percentage of stocks. Portfolios with more stocks, have a higher expected return. In a Vanguard study, it was found that the more you rebalance, the less return you have. Which makes perfect sense, because you own less stocks the more often you rebalance.

[mathjak107]

rebalancing generally only works if you are a good market timer .for the rest of us mortals it can hurt if we rebalance from assets of greater gain potential to ones of lower like stocks to bonds .

it is okay to rebalance from assets with similar gain potential .

[DK]

Has anyone looked at the issues on failing? Like the sequence of returns risk, immediate downturn after retirement - has it always correlated to a high P/E ratio or something like that? Seems like being able to look at other factors instead of just asset allocation or withdrawal rate would give a higher rate of success too.

[mathjak107]

michael kitces looked in to the effect of high valuations at retirement .

https://www.kitces.com/blog/shiller-cape-market-valuation-terrible-for-market-timing-but-valuable-for-long-term-retirement-planning/

[DK]

10% SWR if you retired in the early 20s. Nice.

[kenaces]

most time frames rebalancing to bonds hurt performance . generally taking money from the asset with greater long term growth potential and putting it in one of lesser growth potential cuts gains .

 except for those with volatility aversion i never saw the logic during the long term accumulation periods of mitigating temporary downturns in the short term by permanently giving up gains in the long term  .

asset allocation is really the only free lunch - more return with less risk

https://www.portfoliovisualizer.com/backtest-asset-class-allocation#analysisResults 

Above is just a quick example of simple backtest comparing 90/10 to ivey5 and swenson portfolios.

No.

Asset allocation is considered a free lunch, as you get increased risk-adjusted returns. Not "returns". As mathjak107 said, this makes perfect sense. Think of it this way, if you don't rebalance, your portfolio eventually has a higher and higher percentage of stocks. Portfolios with more stocks, have a higher expected return. In a Vanguard study, it was found that the more you rebalance, the less return you have. Which makes perfect sense, because you own less stocks the more often you rebalance.

Yes, but risk adjusted returns matter and we have to compare apples to apples.

I just read a few Vanguard PDFs on rebalancing(which the advise as a risk management tool) but I couldn't find the study you are mentions.  Can you please share the link?

[mathjak107]

michael kitces looked in to the effects of rebalancing

https://www.kitces.com/blog/how-rebalancing-usually-reduces-long-term-returns-but-is-good-risk-management-anyway/

[Interest Compound]

most time frames rebalancing to bonds hurt performance . generally taking money from the asset with greater long term growth potential and putting it in one of lesser growth potential cuts gains .

 except for those with volatility aversion i never saw the logic during the long term accumulation periods of mitigating temporary downturns in the short term by permanently giving up gains in the long term  .

asset allocation is really the only free lunch - more return with less risk

https://www.portfoliovisualizer.com/backtest-asset-class-allocation#analysisResults 

Above is just a quick example of simple backtest comparing 90/10 to ivey5 and swenson portfolios.

No.

Asset allocation is considered a free lunch, as you get increased risk-adjusted returns. Not "returns". As mathjak107 said, this makes perfect sense. Think of it this way, if you don't rebalance, your portfolio eventually has a higher and higher percentage of stocks. Portfolios with more stocks, have a higher expected return. In a Vanguard study, it was found that the more you rebalance, the less return you have. Which makes perfect sense, because you own less stocks the more often you rebalance.

Yes, but risk adjusted returns matter and we have to compare apples to apples.

I just read a few Vanguard PDFs on rebalancing(which the advise as a risk management tool) but I couldn't find the study you are mentions.  Can you please share the link?

Source (Page 9): http://www.vanguard.com/pdf/ISGPORE.pdf

[k9]

Anyway, rebalancing cannot be a free lunch either. Consider a German investor in 1923. As his German bonds were worth less and less each day, he would have had to sell his stocks in order to buy more and more worthless bonds/cash. That's also true with an japanese investor who kept selling his bonds to buy stocks since 1990. Or a PP investor who kept selling stocks in order to buy gold in the 1980s. That failed because the trend lasted much longer than the investor's rebalancing period.

No magic here. Rebalancing only works when "back to the mean" works. And there is no guarantee markets will revert to the mean after you rebalanced. As a side note : I've always found it funny that some investors believe in rebalancing and not in momentum (or the opposite), yet both rely on the belief that the current trend will follow a given pattern (keep the same in the next few months, then revert to the mean).

Oh, I do think rebalancing has virtues (it's the only way you can keep your asset allocation constant over time, after all), but I think many investors have too much faith in it.

[mathjak107]

my portfolio has never really been rebalanced in a conventional sense  . it swaps funds out from time to time for better choices that fit the bigger picture better and it strives to maintain a certain beta volatility and gain potential . .