You can pick just about any topic and there will be disagreement, that's for sure 😉. We are starting to get off the topic of this thread but what's new 😋. I read the report recommended by @jkandell and while I understand it is a research report with other financial researchers as it's target audience, it was hard to follow their logic. Having a retirement portfolio composed of 100% stocks, regardless of what countries it comprises, would freak most people out 😯. There a few thinks in the report that I agree with, such as having a high % of your assess in equities (but not 100% and not in foreign markets) and they do suggest that right at the beginning of your retirement: The optimal strategy is all equity at every age except for a brief period immediately upon retirement. Investors allocate 27% to fixed income (all in bills) upon retirement at age 65, but that weight shrinks to 7% by age 68 and 0% by age 70.
Some of the other things said in the report are problematic for me. For example,
The relatively short history of US financial markets poses a small sample problem given this setting, so we model forward-looking returns by examining the history of asset class returns from a broad cross section of developed economies.
What? The US has the longest meaningful financial history then any other country on the planet.
Our comprehensive dataset has returns on domestic stocks, international stocks, government bonds, and government bills from 39 developed countries and spans more than 2,600 years of country-month return data.
2,600 years? Come-on, that is meaningless.
Our couples’ aversion to bonds persists across a wide range of model specifications and parameter choices. Our results are insensitive to the sample period (i.e., restricting to post-World War II), bootstrap design, risk aversion, strength of the bequest motive, retirement withdrawal strategy, retirement age, contribution rate, and household type (e.g., single versus couple). Our findings are also not driven by small countries, countries with smaller stock markets, or the inclusion of the US in the sample.
So what exactly ARE you results based on??
In Table II, we list each sample country and the corresponding data coverage. Five countries — Denmark, France, Germany, the UK, and the US — are included in the sample over the full 1890 to 2023 period.
This really makes no sense. Their using data from Japan, Germany, France starting from before WW II. These countries were devastated by WWII and some of them were dictatorships before WWII, hardly reliable data. They are also using block bootstrapping to beef-up some of the data.
And finely this:
Based on comparisons with the pooled sample, the average real returns in the US sample are higher for domestic stocks, bonds, and bills and lower for international stocks. But the US is not an extreme outlier relative to other countries for any of the four asset classes.
Yes I am cherry picking a little bit, but some of this is concerning.
@boomdaddy3 It's the words "recent study" in the reply: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4590406
@pizzaman i actually agree with many of the points you raise against their article.
But i don’t think it’s irrational to take as your standard the sample of all equities in all developed markets rather then just the US market. Even going back to 1928 like pralana does, that’s just one instance in one short time period on how equities act. When we look at the broader history of world equities we indeed see things like country defaults and runaway inflation we haven't had here. Or times when stocks go nowhere for decades. They are rare--but they occur. Should we ignore that? I choose to take the broader perspective that (1) there are some commonalities in how equities bonds and treasuries act across countries and centuries (eg equities have higher expected return than bonds but with greater deviation), (2) one needs to face the possibility that we too might face the rare cases that markets have faced at times in other countries. Japan having 30 years of negative returns, germany losing ww1 and 2, etc.
How much one needs to focus on risk depends on how much of a cushion one has at retirement. With enough of a cushion you don't need pralana or risk management at all. At the other end of spectrum, one needs to take risk management seriously.
@jkandell Thanks for your comments 🙂. Not saying there's anything wrong with investing in developed foreign markets, but for me if you are looking long term (20-30 years) nothing beats the US stock market. It's one thing to look at market history going back 100-150 years, it's another to actually being able to invest in a market during that time frame. For foreign markets, the earliest mutual funds available to retail investors (you and me) that invest in foreign markets (sans USA) are Vanguards VWIGX (1981 - up 442% to present) and VTRIX (1983 - up 50% to present). The S&P 500 during the same time period is up 2,838% 🤩. For me I don't look at all those fancy algorithms and equations and human capital projections and what-not, I look at my retirement funds and how much they are worth now and how much they have increased in value over time 🤓. I will say that foreign bonds in general do better then US bonds and have so for a while 😏.
@pizzaman I think we've just about taken this issue till its end. But, just to be clear, they're not basing their assumptions only on looking at all these markets 100-150 years or investing in things that are not available to retail investors. If you look at the data they use and when they start (table 5 and figure 2 in their other article "stocks for the long run?" ), you'll see some countries like USA, France, Germany go back to 1840s; but others start at 1930s, , 1960s, 1980s (e.g. Israel) depending on when their economies became larger enough to be considered "developed" by membership in the OECD.
OK, don't want to be accused of beating a dead horse 💣, so my last comment is; the research report takes in all this information and then runs it through all their equations and algorithms and comes up with an asset allocation plan that they think will provide the best results for the future. That is a prediction, and we know how wells that works 😣. Start from the beginning of time and go to the present, and use their asset allocation during that time frame and see how well it would have done. Compare that result to the S&P 500 or total US stock market during that same time frame 🤑. Their asset allocation would not even come close to the performance of the US stock market. Will their asset allocation work well into the future? Maybe, but I won't bet my retirement funds on it 😌.
@pizzaman That's not actually what they're saying in their two articles. They are not predicting future returns, which as you note cannot simply be extrapolated from the past. They are noting that the shape of the past distribution changes when you include the various other stocks markets 1840-2019, with a higher chance of "ruin" at the 1st and 5th percentiles , even at 30 year periods, than if you only look at US history. They also note that bonds tend to be correlated with stocks at the 30 year span, and are not as good a diversifier as people might think. Historically, diversifying half your stocks with international had a diversification effect compared to just investing domestically. (As you note, it would have hurt the American-only investor.) Not saying you need to trust these insights, of course.
Needless to say, though useful I think users should be cautious about taking Pralana's monte carlo percentiles too seriously, given they don't reflect the fat tails we see in history, only include US, and only since 1928.
Good points!! I think we can consider this discussion closed. Back to the title of this thread 😋.
Here is an interesting article about Roth conversions: https://rethinking65.com/opening-a-roth-account-may-be-a-good-move-no-matter-your-age/
From the article:
But Campbell, a retired tech worker who is still active and healthy, frequently hiking trails around her Arizona home, is concerned about how her health could change in the coming years and whether she might face the expense of long-term care.
“I think it’s important at our age not to put the burden on our children or grandchildren,” Campbell said. So, at the recommendation of her financial adviser, she recently decided to change her investment mix and is converting a portion of her retirement savings into a Roth IRA.
“Chances are I won’t need that money until later, and it can just grow,” she said.
Advisers like Andrea Clark, who works with Campbell, say contributing or converting funds into a Roth account can yield significant tax- and estate-planning benefits for older workers and retirees.
Clark, owner and founder of the Table Financial Planning, based in Fountain Hills, Arizona, said a Roth account offers the flexibility of having a tax-free bucket of funds that they can dip into for large or unplanned expenses.
Financial professionals say this is a good moment for people seeking to convert existing retirement accounts from pretax to Roth, because of historically low income tax rates that are set to readjust higher after 2025 unless Congress intervenes. In addition, Roth accounts can save heirs from getting hit with big tax bills from inherited IRAs.
He and others in the field say there are several good arguments for having a bucket of tax-free money available in retirement.
Jeremy Eppley, founder of Silverstone Financial in Owings Mills, Maryland, said he believes that the high national debt will prompt the government to raise taxes in the future. “It’s unlikely that, long term, taxes will remain this low,” he said.
Retirees or people approaching retirement who have accumulated large balances in accounts with pretax contributions could find themselves in an unexpectedly high tax bracket once they need to begin taking required minimum distributions at the age of 72 (age 73 for those who were 72 or younger in 2024), especially if they also have other taxable income streams, such as a pension.
