100% equities but only 1/3 domestic

[jabbahop]

I started 15 years ago (a few years before retirement) with 25% international. I have not rebalanced into it so it is lower now.

It has a slightly higher expense ratio, far more non-qualified dividends, and a much higher cost basis so plan to use it as source for paying for Roth conversions starting this year.

 

[Diogenes]

Is there really a good reason that Intl stocks are uncorrelated to US stocks, or is that just how they happened to be in the recent past, with no guarantee it'll continue that way in the future? Trying to answer my own questions, I looked up the charts. The premise is wrong, they are not uncorrelated at all.

Much of the historical folklore about ex-US being uncorrelated with US stocks dates back from the mid 20th century. The 1970s were awful in the US market, but pretty good in aggregate abroad. Recall that the most strenuous period for the 4% SWR begins in 1965-1966. That's exactly the 30-year period during with ex-US allocation would have been most helpful.

The question facing us now, is a flavor of "Is it different this time", except that instead of the very recent past, we're looking over the past 30 years. We ask: did something fundamentally change in the late 20th century, to give US stocks an enormous advantage? If it did, then anything beyond token ex-US allocation is a mistake. But if it didn't, then market-cap allocation (that is, in proportion to US capitalization and ex-US) might be closer to the efficiency frontier.

 

[cooch96]

The underlying idea of the authors is that highly volatile uncorrelated assets give the highest returns in the long run. The book, Fortune's Formula, is a great story about how this works across multiple domains. The key assumption is that an asset class never goes to zero, because any percentage growth from zero is zero.

But we're also human, and for many of us watching big swings, as happen in the emerging markets, are too much to stick with. As a result the conventional advice is to use bonds, with their more stable prices. The tradeoff is the loss of those once every few years big gains. See the periodic table of investments at Periodic Table - Callan

Yes, we 100% equity types are comfortable with the high volatility during accumulation. What surprised me about this study was the suggestion of going 33% domestic with the balance in developed foreign stocks. And to keep this AA throughout retirement. It’s a challenging concept since foreign markets have underperformed the U.S. for a couple decades now and staying 100% equities in retirement is contrary to prevailing wisdom.

I read through her assumptions for returns during the late 19th century and throughout the 20th, with all of its black swan events.

Is there really a good reason that Intl stocks are uncorrelated to US stocks, or is that just how they happened to be in the recent past, with no guarantee it'll continue that way in the future? Trying to answer my own questions, I looked up the charts. The premise is wrong, they are not uncorrelated at all.

First one is since 1996 (start of VTIAX). The red Intl Stock Inx line clearly matches the shape of US Total Stock Inx more closely than the green VBTLX line, with both lagging. I did the 5 year charts separately. Total Bond doesn't look that correlated to Total US stock. Meanwhile, Total Intl stock looks very correlated, just lagging. Both dipped in 2022. If you were investing for protection against large dips in the market, Total Intl was not it. Total Bond dipped some too, but not to the same degree.

I looked at the referenced paper in the first post but lack the concentration to follow what they are saying. If you want better returns over 30 years, sure, Intl stocks have done a little better than bonds. But the OP claimed it wasn't as well correlated, which isn't true. The last 30 years have been strong for the most part. Will the next 30 be the same? Maybe, but if you want uncorrelated protection, I'm not seeing it. Looking at just returns, 100% US would have done far better.

Before jumping to this 2/3 Intl Stock portfolio I'd want to understand the paper a lot better and verify what they are saying and why it apparently doesn't follow the 30 and 5 year charts. Given the charts here, I'm not going to go through that exercise.

View attachment 53713
View attachment 53712View attachment 53711

A big emphasis of the study was to include as much data as possible, by going as far back as records allow. They even used black market reports to estimate inflation during black swan events. So as I understand it, it’s not the recent data that suggests correlation, it’s the old.

I reckon if we figure the world economies will deglobalize, then international markets will be less correlated to the US in the future. If we predict the next few decades will be like the previous few, then keeping our funds in the US market is the right move.

Vanguard’s Total World index is about 68% N. America and 10% emerging markets. Not too much like what this study recommended, but It’s done well for me the past few years; I reckon I’ll keep riding it to victory.

 

[Diogenes]

staying 100% equities in retirement is contrary to prevailing wisdom. ...

I reckon if we figure the world economies will deglobalize, then international markets will be less correlated to the US in the future. If we predict the next few decades will be like the previous few, then keeping our funds in the US market is the right move.

100% equities is unusual, because most folks are trying to maximize their SWR. For somebody withdrawing say <1%/year, the results flip, and pedal-to-the-medal equity allocation works just fine. Where to allocate those equities is of course its own debate. But the idea of 100% equities can be readily embraced, if we're confident that we don't need an aggressive draw from our portfolios.

If the world does substantially "deglobalize", then that ought to favor US stocks. The US is as close as is possible to a self-sufficient market. That's not presently the case, but it was fairly recently. Point being, that the US doesn't need export-markets to sustain its own production of goods and services, because it has enough consumers right here, domestically. And it has resources (agriculture, minerals, fossil fuels) to support itself and to turn raw materials into manufactured goods. Most other nations are dependent on either having markets to which to export, or foreign sources of raw materials.

 

[cooch96]

If the world does substantially "deglobalize", then that ought to favor US stocks. The US is as close as is possible to a self-sufficient market. That's not presently the case, but it was fairly recently. Point being, that the US doesn't need export-markets to sustain its own production of goods and services, because it has enough consumers right here, domestically. And it has resources (agriculture, minerals, fossil fuels) to support itself and to turn raw materials into manufactured goods. Most other nations are dependent on either having markets to which to export, or foreign sources of raw materials.

Sure, but most still want diversification. The US might do best, but it might stagnate or have an extended depression. It’s happened before.

This study used as much data as it could and reports a balance of 1/3 domestic and 2/3 developed foreign markets is the least unsafe AA. I get the argument that 130 year old data from Europe isn’t a strong predictor for investing over the next 30, but history is about the best we have. So many of our other retirement simulators are based similarly.

The conclusion is admittedly difficult to buy, but I spot checked a couple data points. It seems like a legitimate academic work. I’m definitely going to keep an ear out to see if others can debunk it.

This study has challenged my current plan of 100% equities plus a 3year CD Ladder. I don’t think I can ever bring myself to go 2/3 foreign, but I might shrink that CD Ladder and set it back into developed markets overseas. That would get me close to 60% domestic and 35% developed foreign.

 

[sengsational]

Interesting discussion. Peter Zeihan has the world becoming "less flat", from an economic perspective because the decreasing security of shipping intermediate goods on the high seas as the US steps back from the Bretton Woods accord. He's kind of a doomer, but has some justification, given the demographic collapses are hard to argue against; things like population distributions just can't change fast. Given the solution has always been to "grow our way out of things", that's just less of an option for everyone. But the US, Australia, and, surprisingly, France are mentioned by Zeihan to be in the best position. He says China is going to be a train wreck, and countries that you might expect to do ok, like Germany, are not far behind.

 

[VanWinkle]

I tax loss harvested my 20% in non US equities in 2020 during the pandemic. I got a nice cap loss and reinvested in Total US at that time. I've never felt a need to get back into the foreign market and recency bias is now becoming long term bias.

 

[Koolau]

Essentially, Ms. Anarkulova suggests the least risky portfolio is 100% equities, but 2/3 in foreign markets, since foreign equities are less correlated to domestic stocks than bonds and have a higher real return.

I haven't read the article. What is the definition of RISKY? Thanks.

 

[indiajust]

I look on an exUS cap weight world fund as a sector fund that has both currency and trade flow exposure to alter the cap weighting. Australia is commodity mining. China is mfg and coal. Brazil is ag. Russia is oil/gas/gold... etc. Plus you add in tax variables. I do this inside schwab with their EAFE direct index account. easy peasy

 

[aja8888]

I haven't read the article. What is the definition of RISKY? Thanks.

Risky: "Dangerous, hazardous, precarious, perilous, risky mean bringing or involving the chance of loss or injury. dangerous applies to something that may cause harm or loss unless dealt with carefully."

Hope this helps you....

 

[cooch96]

I haven't read the article. What is the definition of RISKY? Thanks.

Running out of money. The study seemed to use Monte Carlo simulations similar to firecalc, but I didn’t backcheck the math.