4% is now considered too high

[chinaco]

Another FP Journal Article Dec 2010 on the topic.

From an international perspective, a 4 percent real withdrawal rate is surprisingly risky. Even with some overly optimistic assumptions, it would have only provided “safety” in 4 of the 17 countries. A fixed asset allocation split evenly between stocks and bonds would have failed at some point in all 17 countries.

Looks like the study only included developed nations (mainly Western European nations or their former colonies).

An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule?

 

[Midpack]

Another FP Journal Article Dec 2010 on the topic. Looks like the study only included developed nations (mainly Western European nations or their former colonies).
An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule?

That was an interesting read (THANKS!). I would never have guessed it would be quite that bad (less than 1.5% SWR) for Spain, Belgium, Italy, France or Germany. Japan was not a surprise but a SWR of 0.5% - OMG!

And that's based on 1900-2008, presumably safe has ratched down for everyone into the future...

 

[Gone4Good]

An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule?

Nice article, and one that gives me reason to cheer.

If you look at this chart, you'll notice that there are clusters of failures around the early / mid 1910s and around the 1940s by countries that didn't fair well in WWI or WWII. You may also notice that at a 4% withdrawal rate, most of the countries still exhibit a 2/3 or better success rate (although Italy stands out as being particularly atrocious). That's surprisingly good considering the ravages of war experienced by the European continent during the first half of the 20th century.

And most of us should feel pretty good about this chart, showing a 100% success rate for 12 out of the 17 countries using a withdrawal rate between 2.5% and 4% . . .

This is a great data set and one that would be fun to dig deeper into. My takeaway from a quick look at the numbers is that your withdrawal plan will likely fail if your country is completely destroyed by war. But even with that included, you still have a really good shot at surviving with a ~3% withdrawal rate.

 

[Gone4Good]

That was an interesting read (THANKS!). I would never have guessed it would be quite that bad (less than 1.5% SWR) for Spain, Belgium, Italy, France or Germany. Japan was not a surprise but a SWR of 0.5% - OMG!

We were posting at the same time, and I came away with a different interpretation (see above).

Remember that the chart you're looking at shows the WR needed to reach 100% success. Also remember that Japan had two nuclear bombs dropped on it, and most of its other cities firebombed. I think its reasonable to conclude that if your country's history includes a period where it was completely destroyed by war, 100% success is going to be very difficult to accomplish. But even with all of the years of war included in the data set (and the past couple of decades of stagnation), a 4% WR still survived 62.5% of the time in Japan. That seems remarkably good.

Given the data set, it would be really instructive to lower the "success" threshold down to the 90% level, or maybe trim 5% of the best and worst years, and see what the minimum WR is then.

 

[Midpack]

We were posting at the same time, and I came away with a different interpretation (see above).

Remember that the chart you're looking at shows the WR needed to reach 100% success. Also remember that Japan had two nuclear bombs dropped on it, and most of its other cities firebombed. I think its reasonable to conclude that if your country's history includes a period where it was completely destroyed by war, 100% success is going to be very difficult to accomplish. But even with all of the years of war included in the data set (and the past couple of decades of stagnation), a 4% WR still survived 62.5% of the time in Japan. That seems remarkably good.

Guess we did. We agree it's an interesting article.

To me, a 62.5% "success rate" would be an oxymoron. The "success" threshold starts at 80% for me, but YMMV.

And a detail, but your characterization that 12 of 17 is being very generous with Spain in that you'd have to have had a stock allocation of 30-50% to succeed at 2.5% - who has that kind of foresight?

 

[donheff]

An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule?

Thanks for all the replies about this study (which allowed me to skim rather than read it ). From my skim, I take it that the historical runs by country evaluated how a portfolio with stocks/bonds of each country in question would have done. It would certainly be remarkable if such portfolios survived in countries devastated by WWI, WWII, civil wars, etc. It would be more interesting to see how well a hypothetical diversified portfolio containing a broad mix of assets from all of those countries and the US did in those same periods. I suspect even a Spaniard with a "world-wide portfolio" might have survived. Another useful chart would look at a typical home country biased portfolio (e.g. Spanish investors with 70% Spain, 30% international; German with 70% German, 30% international). That would better reflect current portfolio approaches.

 

[Gone4Good]

To me, a 62.5% "success rate" would be an oxymoron. The "success" threshold starts at 80% for me, but YMMV.

Yes, but we're talking about a data set where some of those countries were completely destroyed, twice. I think we all recognize that our plans will never be robust enough to survive that kind of scenario. I don't think it is oxymoronic at all to say "I have a financial plan that survives 62% of the time in a period where my country is conquered twice by hostile enemy forces in the most destructive wars the world has ever known." I'm actually surprised its that good.

I don't know how much insight that analysis provides, though. Certainly planning for 100% success doesn't strike me as productive given those circumstances. Striving for 80% seems more reasonable, but we don't know what WR accomplishes that because the authors don't tell us. We know that 4% passes that hurdle in 9 out of 17 countries. 3% seems like a probable winner given the large number of countries where you get 100% success at that WR. And I think most of us would be happy with that outcome.

And a detail, but your characterization that 12 of 17 is being very generous with Spain in that you'd have to have had a stock allocation of 30-50% to succeed at 2.5% - who has that kind of foresight?

No foresight required. 50% seems like a pretty reasonable, and common, equity allocation.

 

[Danmar]

I am pleased that Canada fared pretty well in the study.

 

[haha]

Looks like the study only included developed nations (mainly Western European nations or their former colonies).

An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule?

About 5 years ago we discussed Triumph of the Optimists. There was of course disagreement, but some of us concluded that the only way to arrive at 4% ="Safe"WR was to ignore the majority of countries for which good long term data existed. In other worlds, use incomplete data and rely on the divinely granted special status of the USA. This assumes the usual passive indexes and rebalancing.

YouTube - Sounds of the Sixties - Bob Dylan "With God on Our Side"

 

[Gone4Good]

This study, of course, also begs the question . . . "How would broad international diversification have changed the results?"

 

[REWahoo]

REWahoo: How you understand I? Mist be student tough english/vietnamese dactionery, no??

I'm guessing it's the year I spent in pilot training ("aviator" training for you Navy types) with Tran Tao, fellow student pilot.

 

[Gumby]

-1

This is so far from being grammatical English that I can't even figure out the intent of the posting.

It strikes me that someone who uses leetspeak in Post #42 is hardly in a position to be a grammar nazi.

 

[NW-Bound]

we'll have regrets that we didn't enjoy it more while we were alive....
Consumption at this stage of our lives is for life's experiences - not "things" as some may be thinking.

In a talk with an older colleague, actually a former boss, I said that I'd like to do some travel now, when I still had my health. Repeating an old saying, I said that, on my deathbed I would not want to regret that I did not take this or that trip.

My older friend shook his head and said "Don't worry. Most people on their deathbed would simply wish for the pain to stop".

His wise words left me speechless, and stuck with me. It's sad, but true. In the end, it does not matter that much.

 

[bizlady]

You might want to do some "additional" research about the 4%. Or you can simply ask. I guarantee you that most members of this board do not misunderstood that 4% SWR.

I totally agree that in general, those on this board are better informed and prepared. My statement was meant to refer to the 4% SWR widely quoted in the media, which many outside of this board read and refer to. Sorry if I offended!

 

[clifp]

Yes, but we're talking about a data set where some of those countries were completely destroyed, twice. I think we all recognize that our plans will never be robust enough to survive that kind of scenario. I don't think it is oxymoronic at all to say "I have a financial plan that survives 62% of the time in a period where my country is conquered twice by hostile enemy forces in the most destructive wars the world has ever known." I'm actually surprised its that good.

.

The data does seem to suggest that being conquered creates havoc with a retirement portfolio. A pity the authors didn't include the SWR for the Koreas or Vietnam or Africa with its huge number of civil wars.

So I guess future retirement planning also needs to include the relative strength and national security assessment of your country. The good news is that foreign affairs and national defense policy seem to be a legitimate topic FIRE topic and not mere politics .

 

[Spartacus]

95% rule

personally, I'm using Bob's 95% rule, 4% of the current year's diversified portfolio, only I apply 3.5% instead of 4, just to add another layer of "security".

I believe in the efficient market theory.
I believe only new news moves markets. And since new news is unpredictable, markets are therefore totally unpredictable.
Analyzed potential future events are already "priced in" the markets with their expected probabilities applied by the millions of investors.

But, with the rising tide of the beauty of capitalism, markets will go up over the long term.

 

[Gone4Good]

But, with the rising tide of the beauty of capitalism, markets will go up over the long term.

Fair point. When looking at historic returns, and how they might be applicable to today, it's worth considering some context. Is the world becoming more like the U.S. with stable governments, peaceful borders, and private enterprise? Or is it becoming more like Europe of the early 20th century with powerful and belligerent dictators/emperors making war on their neighbors?

While projecting the U.S. record forward is probably too optimistic, using the European record is probably too pessimistic.

 

[Focus]

personally, I'm using Bob's 95% rule, 4% of the current year's diversified portfolio, only I apply 3.5% instead of 4, just to add another layer of "security".

I'm also leaning toward a more conservative version of the 95% rule (but I still have a few years to change my mind). In my case:

In bad years, take the greater of 4% of current portfolio or 95% of the previous year's spending. In good years, take the lesser of 4% of current portfolio or 4% of original portfolio at retirement (adjusted for inflation).

But really, the percentages above represent what I'll be living on, rather than what I'll be taking from my savings, since my withdrawal rate will drop at 55 (when I start receiving a very small non-COLA pension from a previous company) and at 62 or 67 (when I start taking Social Security).

While projecting the U.S. record forward is probably too optimistic, using the European record is probably too pessimistic.

I agree. There might even be a book in that. It would need a catchy title. "Our Goldilocks Future"? "The Middle Path"? Forget I mentioned it...

 

[walkinwood]

Another variable withdrawal strategy that doesn't get a lot of attention is Guyton's Decision Rules
http://cornerstonewealthadvisors.com/files/08-06_WebsiteArticle.pdf

It is an expansion of an earlier paper by Guyton.
http://www.bobneiman.com/NWM_Pages/...tirees Safe Withdrawal Rate - John Guyton.pdf

I created a spreadsheet a while ago to calculate the withdrawal based on his rules. I use this as another check point to make sure I'm not going too far off track.
http://www.early-retirement.org/for...tions-using-guytons-decision-rules-29684.html

 

[tuixiu]

Isn't it the case that the vast majority of the failures are because of poor market performance early on?

Maybe I'm missing something (in fact it is likely) but I'd think it reasonable to start out at 4% and if the **** is hitting the fan in the first x years adjust down to 3.5% or 3%. Of course this would require being able to retire on 3% of your amount as well as 4%, or be able to work part time or turtle to a cheaper location.

My problem with 3% versus 4% isn't the amount left over, it is the extra years of work that might be required to increase your portfolio an additional 33%. That could be the difference of 5-10 years in your prime.

 

[FUEGO]

Isn't it the case that the vast majority of the failures are because of poor market performance early on?

Yes, I think this is true. I did some research on this exact issue a few years ago and posted the results here.

Basically, if your portfolio is suffering severely 5 or 10 years into your withdrawal period, you need to do something to slow the bleeding. Work part time, reduce your withdrawals, etc.

I personally plan on having a significant portion of my withdrawal be a variable amount based on portfolio value each year, and the rest of my withdrawal would be inflation indexed. Empirically, this will let me pull out around 4% starting out, and there is a 95% chance my withdrawals won't decrease below about 3% in the first 10 years or so. By year 10, I am virtually guaranteed to meet or exceed the 4% SWR.

In practice, this would look like the following: 2% withdrawal indexed to inflation each year plus a 2% withdrawal that is 2% of the actual portfolio balance each year. The former part of the withdrawal dampens the volatility of the latter part of the withdrawal (in good years and bad). Over time, odds are very good that my withdrawal will grow. If, say, I had a million bucks in my portfolio, and I needed $25000 as my absolute bare bones living expenses, then I could take $40,000 a year per my 2%/2% strategy and have a very good chance (95%+) my withdrawals would never dip below $28,000-30,000.

The advantage of this method (the hybrid method I call it) is two fold: you don't annually increasing amounts blindly as your portfolio craters and you get to enjoy spending more in the vast majority of scenarios where your portfolio appreciates over the decades. I think the hybrid method more closely approximates what you will want to do anyway: conserve money when you get poorer and spend more when you get richer.

This is all assuming that the past is prologue for the future.

 

[Gone4Good]

Isn't it the case that the vast majority of the failures are because of poor market performance early on?

That is what the data shows. But it's worth remembering that "early on" is based on a 30 year retirement. What does "early on" mean for someone retiring in their 50's . . . the first couple of decades?

 

[verygoodthings]

+1
the study on which the "4% SWR" is based assumed a 30-year retirement and an equity allocation of 75%. I think also the 4%WR assumed that the portfolio would be totally consumed at the end of the 30 years and had a 5% probability that you would outlive your money. I'm going by memory here, so I'm open to correction.

If your retirement will differ from those parameters, an adjustment to the WR is probably in order.

Ok the above is incorrect. The 4% Safe withdrawal rate was intended as a level of withdrawal that would ensure that the portfolio would NOT deplete in a 30 year time period. The aftcast on this shows that it is possible, or has been possible in the past, that the portfolio could deplete. It is not likely though, and it is much more likely you would have a large estate left behind.

The best book written on this SWR topic is Unveiling the retirement myth by Otar.

Regarding the 5% chance of "outliving your money", that would be probably in the range of a 8% withdrawal rate, based on the initial year, and indexing to inflation.

Something that doesn't get mentioned much is the variance people take in their withdrawal rate. If you retire on 2M asset base, use 4% WR (80k initial year), and the portfolio falls 40% the first year, it is unlikely you will still feel find pulling 80k + inflation the very next year. However, this is actually what these SWF studies are suggesting. They are using the number of 4% as the very first year, and regardless of asset movement, continuing to withdraw the same amount (not percentage!) plus inflation in all following years. The understanding here is that your assets fluctuate, but your spending should not.

The 4% withdrawal rate depends on many things for portfolio survival, the biggest of which in my opinion is management and trading fees. If you have a million dollar portfolio, pay 1.5% in management and trading fees, and decide to withdraw using this 4% rule, you are actually withdrawing 5.5% as per the studies conducted. This is kind of my primary reason for shying away from any asset management. (that being said, I am considering a flat fee DFA advisor).

I'd love to talk more about this topic with anyone interested. If anyone hasn't already, really check out that book. I actually read it first from the free download the offer gave, and then decided to buy the paper copy ($50!, way expensive for a financial book), but worth every penny to me. The book was written and targeted to retirement advisors if I recall correctly.

 

[verygoodthings]

That is what the data shows. But it's worth remembering that "early on" is based on a 30 year retirement. What does "early on" mean for someone retiring in their 50's . . . the first couple of decades?

Early on as described by the people that have done these studies is the first several years. The importance of market activity is actually most important in the first year, second most important the second year, etc. By the time you get to your late retirement years, the market really doesn't matter as much. The reason for this is that you are withdrawing the funds from what could be a damaged portfolio. If you retire with 1M asset, and you decide to use a 6% WR (admittedly high, but this is probably a situation MANY people will find themselves in), the first couple years make all the difference. If the portfolio goes to 650k year 2, and you make another 62k (4% initial amount plus inflation) withdrawal that year, you can see that effectively that is around a 9.5% withdrawal rate! Hopefully this makes sense of this for some people. The opposite is true is you have a banner year 1, portfolio goes to 1.2M, you make a 62k withdrawal. Effictively, for that isolated year, you've only withdrew 5.1%.

 

[Danmar]

Verygoodthings: This topic has been discussed extensively on this forum. Otar's book as well. I think you might be a little high on the SWR for a 95% success probability. Use the search function if you want to read previous threads on a particular topic. You will learn a lot here, I did.