Japan Stock Market and Long term Market Risk

[LH]

In terms of risk, I think a lot of new bogleheads see the efficient frontiers, with its risk as beta, see the 0 bond/100 stock beating everything, and think, I am young, I will go 100 percent stocks. That was me.

Now I think I realize, that long term, say 40 years, the stock market has a certain risk (not beta) of returning very poorly. The most recent, and current example of this on a large scale is Japan 1984 to 2007, or 1990 to 2007, both are pretty bad.




This is why I am going to buy some bonds, even though my retirement is 30 years away. There is no guarentee the nikkei index will go up much in the next 10-15 years either right? The same thing may happen with the US TSM, the world TSM, or whatever index one could imagine right? I do not think it likely, but it may. Bonds may outperform stocks over long periods of time in theory.

Please comment on this aspect of risk.

Also, I had another question with this recent "dow illusion" post about how in many foriegn currency the dow is actually pretty flat or even negative, and thus an illusion based on currency exchange rates. This makes me wonder about how that curve above looks like to a japanese investor, in yen? They have been in a deflationary period and such, so maybe it looks very different to them?

edit: dow illusion article http://usmarket.seekingalpha.com/article/23981

Thanks for info and comments,

LH

[grumel]

Does the Nikkei Reflect Dividends ?

I would love to have some more in dept Charts on this:

a) A stock chart that is adjusted for inflation
b) a Chart that compares stock returns with bond returns, would be interesting to see how long one had to invest 100% stocks to overcome 100% bond returns, and how long it did take for Stockholders to come out without loss in real money.

All with Dividends re-investment of course.

[bpp]

This makes me wonder about how that curve above looks like to a japanese investor, in yen?

That curve is in yen, so that is what a Japanese investor saw. But dividends are not included.

Grumel: Not a full answer to your question, but here are yearly returns for the TOPIX index (nearly a full-market index) with dividends included from 1990 to 2006, and also the yearly changes in the Japanese CPI over that period.

Code: Select all

Year	Topix	Inflation
1990	-39.44%	  3.10%
1991	-00.39%	  3.30%
1992	-23.02%	  1.60%
1993	  10.97%	  1.30%
1994	  09.11%	  0.70%
1995	  02.09%	-0.10%
1996	-06.06%	  0.10%
1997	-19.41%	  1.80%
1998	-06.57%	  0.60%
1999	  59.69%	-0.30%
2000	-24.96%	-0.70%
2001	-18.91%	-0.70%
2002	-17.50%	-0.90%
2003	  25.19%	-0.30%
2004	  11.34%	  0%
2005	  45.23%	-0.30%
2006	  03.02%	  0.30%

[Karl]

I love your chart.

I've long been using Japan as an example:

1) Of why not to invest everything in one nation (as some Japanese investors surely did).

2) That risk is real in stocks. Some people think stock risk means the market could fall 20% or even 50%, but then it just bounces right back. Yeah, right. Japan tells a different story.

Every time I ever used Japan I always got the same tired old argument that goes something like this: "Well, that can't happen to the US because we're magically different from every other nation on earth." I failed to find that argument compelling.

[nick22]

You mean Americans aren't magically different? LH brings up one of the main reason behind my diversification strategy; diversifying to avoid specific geographic, currency, or sector risk (even equity risk itself). I have no idea what future correlations will be and how to effectively optimize the volatility of my portfolio, but I will not let one country, currnecy, or geographic region unduly impact my portfolio return.

Own stocks, own bonds, own things in many geographic regions, all sectors, and you will be well protected from the risk of overconcentration killing your retirement plans. So sell your company stock, buy something in EM, Europe, and the Pacific region, buy 10-20% bonds, and 12% real estate and sleep better at night knowing that the whims of one market cannot sink your fortunes.

[LH]

Ah, that curve from yahoo.com IS in a "yen perspective"? Then that chart doesnt reflect what a US investor would have returned right? Just like our DOW charts do NOT reflect the returns a foriegn investor would see, aka the "dow illusion" posts????

This whole DOW illusion thing is even more confusing than I thought, if our US/english language based Nikkei index charts are showing things relevant to a japanese investor, and not directly relevant to a dollar based investor?

I am now more confused than I started out in this thread, heh, or perhaps just more knowledgeable of my lack of knowledge : )

So, if this curve is how it looks to a japanese investor, investing in that index in yen, where is the chart that shows what happens to a US investor, investing in that index? I have just realized in all this that, much like REIT PE vs REIT aFFO issue, I have no clue what I have been looking at in these foriegn curves over the past couple of months.

The whole Dow illusion, http://usmarket.seekingalpha.com/article/23981
issue fascinates me, though I do not see how it relates all that much to me, as a dollar investor investing in the US (I can see the macroeconomics of it yeah), but I can see the converse relating more directly to me as a dollar investor in Japan..... If I am starting to understand it right.....

Thanks for help,

LH

[Elysium]

I'm going to offer a counter argument to the issues raised in the original post.

LH wrote

Now I think I realize, that long term, say 40 years, the stock market has a certain risk (not beta) of returning very poorly. The most recent, and current example of this on a large scale is Japan 1984 to 2007, or 1990 to 2007, both are pretty bad.

Yes, but you don't have to do anything to protect against that risk, in fact, it is going to be almost impossible to protect against such an event, and in the rarest event that all the stock markets of the world return flat over 40 years bonds are not going to be of help either. Saving more, working more, maintaining good health, living frugally are all things that will be more helpful.

This is why I am going to buy some bonds, even though my retirement is 30 years away. There is no guarentee the nikkei index will go up much in the next 10-15 years either right? The same thing may happen with the US TSM, the world TSM, or whatever index one could imagine right? I do not think it likely, but it may. Bonds may outperform stocks over long periods of time in theory.

Like I stated above this is a very remote possibility that you don't need to be worried about, and even if you are bonds are not the answer. Japan is just one country, United States is also just one country, so don't bet everything on large cap stocks of just one country. Instead spread your bets around major markets of the world, also buy some value stocks and small stocks, and you have set yourself up for best probable result.

Now, let's just assume what you said can actually happen, then what investment choices will help you in such a situation? I think an investor in accumulation still have to focus more on saving, then they can wait until some of this is actully happening before they do anything. This is something that people overlook. That you could take action after the fact to correct course and still have good outcome in the long run. Let's say 10 years from now all stock markets of the world has returned flat, you did not lose a lot of money, but saved a lot, your returns are flat. You still have 20 or 30 more years before you start drawing on your capital. You can decide what to do next based on the new information available then, should you continue holding high percentage in stocks, or should you now buy more of something else, like bonds.

The course corretion can be done with the help of new data. This is much easier to do and much more better than trying to protect against something that may never happen.

[bpp]

So, if this curve is how it looks to a japanese investor, investing in that index in yen, where is the chart that shows what happens to a US investor, investing in that index?

Here is my stab at it:



If I haven't made a mistake somewhere this should be, in US$, the value of $100 invested at the end of 1989 in the TOPIX index, with dividends reinvested, until the end of 2006.

[LH]

Dieharder:

good points : )

I was wondering how bonds actually did that situation... Part of my problem, hitting up against this whole dow illusion currency deal in terms of these curves, is that I am no longer sure exactly what I am looking at anymore....... Skipping that for now(edit: ah Someone posted in between about that above):


So you are saying, that a Japanese investor, who started buying say 10K of all bonds in 1984 versus 10k of all stocks in 1984, would NOT be ahead of the guy in stocks? Or say in 1989 the same thing was started and continued to now? That the bond guy would not be ahead, and maybe even far ahead? It seems highly counterintuitive to me that that is the case, heh doesnt mean its not right of course.

Is that what you are maintaining, thats what historical returns actually show that bonds would not have helped?

Interesting stuff if true and something which would change my consideration of the issue.

Thanks,

LH

[LH]

So, if this curve is how it looks to a japanese investor, investing in that index in yen, where is the chart that shows what happens to a US investor, investing in that index?

Here is my stab at it:



If I haven't made a mistake somewhere this should be, in US$, the value of $100 invested at the end of 1989 in the TOPIX index, with dividends reinvested, until the end of 2006.

thanks a lot, so 17 years of negative return in the stock market there. Though it looks markedly different from the curve at yahoo. Kinda scary how far off my assumption of what that graph meant for me as a US investor in the Japanese market.

Now with all the worlds markets I would posit, becoming more linked, and given the fact that everyone in late 80s thought Japan was going to completely dominate for years(I seem to remember century of Japan, and a time/other magazines cover and such), whats to prevent the world TSM from doing that for a long time? Whats the real risk of it? I would say its hard to know really, the sample size is too small. But something to consider.

I would think bonds would outperform stocks then, but maybe not, that is historically testable at least withing the limits of the current japanese experience, as well as the limits of the US great depression.

Again, my assumption about bonds then may be wrong.

[LH]

That curve is in yen, so that is what a Japanese investor saw. But dividends are not included.

I would never have guessed that. So I am ok to assume all foriegn index curves published on yahoo.com are from the perspective of someone investing in that index in that countries own currency? And that they are likely to really NOT correlate with my return since I would have invested in dollars? Stunning to me really.

How about FXI, the china fund? Is that based in dollars or in chinese money perspective?

In general, is there some easy way to tell the currency perspective?

To be paranoid perhaps, I would think when I look at a curve of sony corp, on yahoo.com that its IS in a dollar perspective right?

Thanks for the help its highly appreciated,

LH

[mikenz]

any quote will be in the currency of the exchange on which it is traded

[mikenz]

This is disturbing news:

Following last week's news that Origami Bank had folded, we are hearing that Sumo Bank has gone belly up, and Bonsai Bank plans to cut back some of its branches. Karaoke Bank is up for sale, and it is going for a song.

Meanwhile, shares in Kamikaze Bank have nose-dived, and 500 back-office staff at Karate Bank got the chop. Analysts report that there is something fishy going on at Sushi Bank, and staff there fear they may get a raw deal.

[BrianTH]

I very much agree that people overlook the Japan example when contemplating how long an asset crisis could last. In general, I think it is unfortunate that people often talk about things like maximum possible losses without at the same time talking about how long those losses might persist.

That said, if you are still deep in accumulation (meaning your future contributions greatly outweigh what you have contributed so far), then even this sort of scenario may not be bad for you. However, I think some bonds often does make sense to the extent you cannot be sure that your contributions will continue as planned throughout an asset crisis.

[Malachi]

According to Wikipedia, the Tokyo indexes are not market-cap indexes. They are float based indexes.

Stocks are weighted on the Nikkei 225 by giving an equal weighting based on a par value of 50 yen per share. -- Nikkei 225, Wikipedia

Is this true?

If it is then those indexes are not showing us what investors' returns are.

[bpp]

Is this true?

That is true of the Nikkei. Though I quoted the TOPIX, which is a cap-weighted index.

If it is then those indexes are not showing us what investors' returns are.

Well, there are Nikkei index funds.

[Malachi]

Is this true?

That is true of the Nikkei. Though I quoted the TOPIX, which is a cap-weighted index.

Ah, sure enough. Never mind.

[Adrian Nenu]

Nikkei 1989, NASDAQ 2000 and the US Crash of 1929 are events to keep in mind when evaluating how much risk investors can handle. These events had several things in common: outrageous valuations, lots of leverage, tons of media coverage and (except for NASDAQ 2000), very weak regulatory controls. The 1973-1974 bear market is a more realistic event for our times, given the presence and oversight of the SEC and banking regulatory agencies such as the Fed. Nevertheless, all these events show the importance of global equity diversification in portfolio management and the risk of concentrating assets in a country/region. Just something to keep in mind.

Adrian
anenu@tampabay.rr.com

[Elysium]

LH:

So you are saying, that a Japanese investor, who started buying say 10K of all bonds in 1984 versus 10k of all stocks in 1984, would NOT be ahead of the guy in stocks? Or say in 1989 the same thing was started and continued to now? That the bond guy would not be ahead, and maybe even far ahead? It seems highly counterintuitive to me that that is the case, heh doesnt mean its not right of course.

Is that what you are maintaining, thats what historical returns actually show that bonds would not have helped?

No. That is not what I am saying. What I am saying is that this is such a low probability event. That all the stock markets of the world will go the Nikkei route for 40 years. If it were to happen then yes having planned for it would help, and bonds certainly may help. But ... I am saying that you should not plan for such a low probability event, rather let it happen, look at new information after some of this has already happended, then take correction if you need to. It's not a question of whether a bond portfolio would have been ahead or not, it's a question of what will be the impact to your future plans and how well you can adapt to the situation. Sometimes you just have to let events unfold and then try to adapt rather than plan for it. I know this is an unconventional thought, but in managing risk you have to always look at probability too.

[bobcat2]

From Triumph of the Optimists I have the real returns for Japanese stocks and bonds for the ten years from EOY 1990- EOY 2000.

The real return on Japanese stocks from 1990-2000 was -7.1% per year. The real return on Japanese bonds from 1990-2000 was 5.4% per year. In other words from 1990-2000 Japanese bonds beat Japanese stocks by an average of 12.5% per year. In inflation adjusted terms a stock portfolio worth $100,000 in December of 1990 would have been worth less than $50,000 in December of 2000. In inflation adjusted terms a bond portfolio worth $100,000 in December of 1990 would have been worth over $164,000 in December of 2000.

During the three years surrounding this decade, i.e. 1990, 2001, and 2002 the Japanese stock market did much worse on average than it did on average during the 10 years from 1990-2000. However, I don't have bond returns for those years, but it is safe to say that bonds did much better than stocks during those three years. For example, the Japanese stock market was down nearly 40% by most measures in 1990.

Bob K

[bobcat2]

I agree that stocks having really bad returns over a 40 year period is a very low probability event. However, stocks having negative real returns over a 20 year horizon has a reasonable probability of occurring. Imagine inflation averaging 3% over 20 years while stocks return an average of 2% nominal over the 20 years. Our pre-retirement investing lives are so relatively short that such an outcome would be devastating to most retirement portfolios over that period. It would be very difficult for most investors between the ages of 30 thru 55 at the beginning of this bear market to recover from it. In real terms the end stock portfolio would be worth less than 82% of the beginning portfolio. If you are depending on your portfolio for a significant amount of your retirement income, you are headed for the Alpo retirement diet. On the other hand, a portfolio of 20 year TIPS purchased at the beginning of the 20 year period with a coupon of 2.5% would be worth in real terms approximately 164% of their beginning value at the end of the 20 years. This retirement diet won't be heavy on caviar, but there won't be any Alpo served.

I'm not advocating holding an all TIPS portfolio here. I am warning that a portfolio that is 80% or more stocks is risky in the LR.

Bob K

ps - There are many US investors for which holding 80%-100% stock portfolio is reasonably advice. If you have a separate emergency fund and a very small portfolio you have little to lose and a lot to potentially gain by investing heavily in stocks. Once you cross the $50,000 and $100,000 portfolio value thresholds, though, this advice ceases to be prudent. ('')

[Peppe]

In inflation adjusted terms a stock portfolio worth $100,000 in December of 1990 would have been worth less than $50,000 in December of 2000. In inflation adjusted terms a bond portfolio worth $100,000 in December of 1990 would have been worth over $164,000 in December of 2000.

The numbers sound bad, but those aren't realistic portfolios.

If you are in the accumulation stage the charts look ok... Which if you are 100% stocks you should still be in accumulation phase. It looks like if you added 10000 a year for 10 years you would have atleast positive returns today.

If you aren't accumulating then you should have bonds to ensure some returns to live on in down markets. In this case you will likely need a lot of bonds or diversify outside the Japanese markets.

I guess this shows why we diversify to international stocks in case the US experiences a long down period and why we go to more bonds later in life.

[LH]

LH:

So you are saying, that a Japanese investor, who started buying say 10K of all bonds in 1984 versus 10k of all stocks in 1984, would NOT be ahead of the guy in stocks? Or say in 1989 the same thing was started and continued to now? That the bond guy would not be ahead, and maybe even far ahead? It seems highly counterintuitive to me that that is the case, heh doesnt mean its not right of course.

Is that what you are maintaining, thats what historical returns actually show that bonds would not have helped?

No. That is not what I am saying. What I am saying is that this is such a low probability event. That all the stock markets of the world will go the Nikkei route for 40 years. If it were to happen then yes having planned for it would help, and bonds certainly may help. But ... I am saying that you should not plan for such a low probability event, rather let it happen, look at new information after some of this has already happended, then take correction if you need to. It's not a question of whether a bond portfolio would have been ahead or not, it's a question of what will be the impact to your future plans and how well you can adapt to the situation. Sometimes you just have to let events unfold and then try to adapt rather than plan for it. I know this is an unconventional thought, but in managing risk you have to always look at probability too.

Yes, its all about probability and thinking about what risk really is.

I think that there is not enough information present to conclude what the risk of a 40 year downturn in global stock markets is, the available sample size in years is simply not big enough. If you discount 40 as very low, how about making that 40 a 30 year, or a 20 year, and think along the same lines, one gets a better appreciation of what real risk is, at least I do.

I think to completely discount a long worldwide stock downturn, where bonds DO beat stocks worldwide for many years, is unwise deciding ones stock/bond split. It is possibly not an insignficant risk.

Worldwide markets, to me, seem to be becoming more interelated and likely more correlated.

What is the available historical data set in years for a worldwide market that is highly interdependant, where my shoes(and maybe about everything I use) are not made in the US, but in china? Was that true 80 years ago? How much data are we working with here really? The expected returns of stocks and bonds are derived historical returns, with some ex post theory to justify them in my book. Theory cannot/did not predict what happened in Japan ex ante right?

If you took a poll of wall street 1989 (or 1984), threw up say 10 charts of projected performance of japan stock market, including the real one, what would the probability they would assign to the chart we all know to be true? I would posit, very very very low, ie do not worry about it level. They would probably all posit good returns of the japanes stock market, if you predicted the actual performance, they likely would have laughed at you right?

To me, the reason I am going into bonds, is partially to mitigate long term risk that bonds will outperform total world market stocks over a significant interval, not just the "sleep factor" to help me handle simple beta type risk.

The efficient frontier does not say what some people think it does(me 3 months ago), a 90/10 stock/bond split could outperform a 100/0 stock bond split long term. The probability of that happening, is not insignificant, especially given that the predicted outcomes are close.

A 50/50 split may beat a 100 percent stock split too over 30 years. In Japan, I think it likely has over a long period of time. Which makes the reason to go into bonds even more compelling to me.

Interesting stuff to talk about, no right answer per se. Thanks for your comments.

[LH]

Thanks for the figures on the bonds returns, much appreciated.

This discussion is very helpful to me, thanks.

[fire5soon]

Because of these risks and Japan-like history, I will not put more than 35% of my retirement money in US stocks. I'm as patriotic as the next guy, but the risks of high country concentration are very real IMHO.

[johndcraig]

Diversification and Black Swans

Some see Japan as simply a call for diversification. Others see it as an example of a black swan. Both views are correct to a point. For those who think diversification is the ultimate solution, I suggest that they read the Global Bubble article at GMO for another viewpoint.

http://www.gmo.com/america

Many refuse to give anything but lip service to black swans; they believe that history explains the downside limits. It may well be that the US will never duplicate the Japan situation, but that does not mean that something equally negative cannot happen here.

A switch of part of the portfolio from equities to fixed (ST or TIPS) will materially lessen the risk, but IMO diversification among risky assets, including international, will not. Some seem to believe that a switch from equities to ST fixed or TIPs will not protect against downside risk in a very material way. This IMO defies logic.

John

[bpp]

It may well be that the US will never duplicate the Japan situation, but that does not mean that something equally negative cannot happen here.

In fact it already has, in a sense: $100 invested in the Japanese market at the beginning of the 1970's would have a larger value now than if that $100 had been invested in the US market at the same time. This is true for either a Japanese or a US investor. Which shows, I guess, that boom followed by bust is better than bust followed by boom.

[BrianTH]

I think the probability of such an asset crisis happening everywhere in the world is obviously less than it happening just in the US, more or less by definition (you are basically looking at the probability of A & B, where A is the probability of the event happening in the US and B is the probability of it happening everywhere else, and unless A necessarily implies B, the probability of A & B both happening will be less than the probability of just A happening). Exactly how much less likely a global asset crisis would be than a US-limited asset crisis is a more interesting and difficult question, but it is probably worth noting that Japan was in fact very important on a regional and global basis, and although there was some contagion it was not overwhelming. So, I suspect that if necessary, the world could in fact weather many possible US-specific asset crisis scenarios.

By the way, unfortunately johndcraig appears to be making up a strawman to argue against (yet again, I might note). Personally, I do not know of anyone who would claim that "a switch from equities to ST fixed or TIPs will not protect against downside risk in a very material way." Indeed, to me that is just as extreme as claiming that "diversification among risky assets, including international, will not [materially lessen these risks]."

[johndcraig]

Brian

Personally, I do not know of anyone who would claim that "a switch from equities to ST fixed or TIPs will not protect against downside risk in a very material way." Indeed, to me that is just as extreme as claiming that "diversification among risky assets, including international, will not [materially lessen these risks]."

Believe me, I don’t want to go “there” again. Please just answer this question:

Consider the entire universe of possible material risks including type, degree, duration and probability. Given this entire universe, please compare the relative benefits of the following two approaches for reducing such risk.

1. A switch from equities to ST fixed or TIPs
2. Diversification among equities including international, stock asset classes, REITS, commodities, etc.

In addition, please opine on whether there would be a material reduction of such risks if action was limited to: 1 and not 2; 2 and not 1.

John

[BrianTH]

Believe me, I don’t want to go “there” again.

Then why did you go there again above, and why are you going there again now?

Please just answer this question:

Consider the entire universe of possible material risks including type, degree, duration and probability. Given this entire universe, please compare the relative benefits of the following two approaches for reducing such risk.

1. A switch from equities to ST fixed or TIPs
2. Diversification among equities including international, stock asset classes, REITS, commodities, etc.

As I have explained before, I do not believe a basis for such a comparison exists, in part because it is not possible to aggregate all the different relevant risks within a single measure. I think the only thing one can say in response to a question such as yours is that both strategies may or may not have benefits depending on the scenario in question.

By the way, in my view your first strategy is actually just a species of the second, meaning I see diversifying one's holdings between common stocks and bonds as just one of many diversification strategies worth considering.

In addition, please opine on whether there would be a material reduction of such risks if action was limited to: 1 and not 2; 2 and not 1.

Yes and yes, with the caveat that I am using my own definition of "a material reduction of such risks", since you have refused to provide such a definition.

By the way, as I noted above, I think to claim that either strategy could do nothing useful to moderate risks would be an extreme position, so a "no" in response to either part of your question would strike me as an extreme claim. Again, to my knowledge no one has taken such a position, besides of course yourself.

[test]

The Nikkei since 1955 is up over 15,000% not including dividends.

[Elysium]

bobcat2 wrote:

Our pre-retirement investing lives are so relatively short that such an outcome would be devastating to most retirement portfolios over that period. It would be very difficult for most investors between the ages of 30 thru 55 at the beginning of this bear market to recover from it. In real terms the end stock portfolio would be worth less than 82% of the beginning portfolio. If you are depending on your portfolio for a significant amount of your retirement income, you are headed for the Alpo retirement diet. On the other hand, a portfolio of 20 year TIPS purchased at the beginning of the 20 year period with a coupon of 2.5% would be worth in real terms approximately 164% of their beginning value at the end of the 20 years. This retirement diet won't be heavy on caviar, but there won't be any Alpo served.

I would argue that the pre-retirement investing period is not so short. A worker in his/her 20's have 40 years of accumulation, in their 30's they have 30 years of accumulation, even in their 40's they have 20 years of accumulation. This is relatively long. More important point though is the fact that someone is in accumulation gives them all sorts of choices to make and adapt to the situation. I think the case presented above is a highly exaggerated one, a highly diversified portfolio of stocks and bonds returning negative real returns over a 20 year period is highly unlikely. Some market will be performing well. And most people have at least 20% in bonds here. We are not talking 100% stock portfolios. I would argue even 100% stock portfolio is ok, but my risk tolerance is not that high.

What if it were to happen though? Using various scenarios for a worker in his/her 30's investing a modest $7500K starting with $50K initial balance and earning zero real over 30 years will end up with $300K. This is the worst case. If they save max allowed $14K and also get a 3% company match then this figure will be $500K. This does not even include yearly raise in salary and the increased contributions. They will not be eating Alpo by any means. $300K with Social Security can get you decent living if you also have a paid off house in 30 years. You could let out part of your house for rent for additional income. This is the worst case. Remember you are doing better than most people still, since you are a good saver.

[Karl]

The Nikkei since 1955 is up over 15,000% not including dividends.

Which is great if you invested in 1955. I couldn't seeing how I wasn't born till 1973 and didn't start investing till 1992.

Japan's market had an amazing run from the end of WWII till 1989 and it's been hell since then.

People love to look at long term averages as in the S&P has produced more than 10% a year since 1926. Yeah, great if you have an 80 year investment horizon. The problem is that averages hide the reality of extreme volitility where you have some periods -- sometime very long -- of stunning performance and have other periods -- also stunningly long -- of rotten performance.

Imagine if a grandfather in Japan had put a lump sum into his nation's market in 1989 as a gift expecting it to grow to fully fund his newborn grandchild's college education. The kid is now ready for college and the account is worth less than when the kid was born! Sure indexes don't count dividends, but then they also don't count real world expenses like mutual fund expense ratios -- so they pretty much cancel each other out here. We're talking about investing over an entire generation and still losing money.

If Japan's market were to earn 10% a year going forward, for example, it would take till 2015 before it gets back to its 1989 peak -- a full quarter century of going nowhere. People talk about how the market always wins in the long run. Problem: you're dead in the long run.

[Elysium]

I wanted to add one more comment to this statment by bobcat2.

On the other hand, a portfolio of 20 year TIPS purchased at the beginning of the 20 year period with a coupon of 2.5% would be worth in real terms approximately 164% of their beginning value at the end of the 20 years.

That may be true, but someone in accumulation has only a small amount to buy every period, usually bi-weekly through pay checks. So a snapshot of that taken 20 years prior would be a small amount, and it won't buy you much anyway regardless of returns. If we look at it is a single purchase made at the peak of a market and left alone it is a different story as opposed to a series of purchases made throughout our investment lives.

[nick22]

Consider the entire universe of possible material risks including type, degree, duration and probability. Given this entire universe, please compare the relative benefits of the following two approaches for reducing such risk.

1. A switch from equities to ST fixed or TIPs
2. Diversification among equities including international, stock asset classes, REITS, commodities, etc.

Johndcraig,
What the heck does that first sentence mean? If you want to minimize the risk that "unexpected" returns from one type of concentrated investment (equity, country, or currency) unduly impact your portfolio performance, then you need to employ parts of strategy 1 and 2. But defining the entire universe of possible material risks is a ridiculous and vague task.

Like I said before, if you want to spread your bets around then have some total domestic equity, some total int., some EM, some RE, some bonds, some TIPS and you should sleep better.[/b]

[johndcraig]

OK Brian

Then why did you go there again above, and why are you going there again now?

I admit I was stupid to think you would ever give a responsive answer to this most important point. You could of course define "material" in any way that you choose. Of course the message that you send is that one approach is as good as the other for covering the universe of possible risks. I assume no one will actually take you serious, so there is no need to press the point further.

John

[BrianTH]

Of course the message that you send is that one approach is as good as the other for covering the universe of possible risks. I assume no one will actually take you serious, so there is no need to press the point further.

John, with all due respect, I think that if you stopped trying to put words in my mouth, it would be easier for you to realize that I AM giving you responsive answers. For example, if you actually paid attention to what I wrote above, you would know that I would not in fact say "one approach is as good as the other" (a message that you attributed to me, but which I never said). That is because to make such a claim would still require having a basis for making such a comparison, and I do not think such a basis exists. So, once again, your attempt to put words in my mouth has gotten my views wrong, as has happened so often in our discussions.

The real problem, of course, is that in responding to your questions I am often refusing to accept your premises. In other words, I am not giving you the responses that you want or expect to hear, but that is what happens in a dialogue with real people--they don't answer according to your script.

[johndcraig]

Brian; no basis for comparison?

Here is the question I posed.

Consider the entire universe of possible material risks including type, degree, duration and probability. Given this entire universe, please compare the relative benefits of the following two approaches for reducing such risk.

1. A switch from equities to ST fixed or TIPs
2. Diversification among equities including international, stock asset classes, REITS, commodities, etc.

Here are two of your comments.

I do not believe a basis for such a comparison exists, in part because it is not possible to aggregate all the different relevant risks within a single measure. I think the only thing one can say in response to a question such as yours is that both strategies may or may not have benefits depending on the scenario in question.

I would not in fact say "one approach is as good as the other" (a message that you attributed to me, but which I never said). That is because to make such a claim would still require having a basis for making such a comparison, and I do not think such a basis exists.

So there is no basis by which to make a comparison; is there therefore no basis for selecting one approach over the other when seeking risk reduction? You say that a comparison depends on the scenario in question, but this question is aimed at the universe of risks generally faced by all investors. Regarding a specific case, you previously said that Larry’s switch to 20% equities materially reduced risk, but now it seems that you may be saying there is no basis for believing that he couldn’t have reduced risk to the same degree or even more by going to a diversified 80% equity portfolio.

I’m not trying to put words in your mouth; I am simply trying to understand why you do not believe you can compare the two approaches. If I have misstated your views, please explain.

John

[BrianTH]

So there is no basis by which to make a comparison; is there therefore no basis for selecting one approach over the other when seeking risk reduction?

No, there is not. Fortunately, there is also no need to select one approach over the other, since you can use both (indeed, as I see it, they are really just variations on the same approach).

You say that a comparison depends on the scenario in question, but this question is aimed at the universe of risks generally faced by all investors. Regarding a specific case, you previously said that Larry’s switch to 20% equities materially reduced risk, but now it seems that you may be saying there is no basis for believing that he couldn’t have reduced risk to the same degree or even more by going to a diversified 80% equity portfolio.

John, I honestly do not get why you keep trying to put words in my mouth (which is what you do every time you start a phrase with something like "you may be saying ..." and then finish it with something I never said). It seems to me that virtually every time you have done so, you have gotten my views wrong, and then I have to correct you before we can continue the discussion, and then we get nowhere because you repeat this habit of putting words in my mouth.

Specifically, in this case, you claim: "you may be saying there is no basis for believing that he couldn’t have reduced risk to the same degree or even more by going to a diversified 80% equity portfolio."

Of course, that is not what I would say (so far as I can tease out the meaning of your double negative). If someone claimed that Larry could "reduce risk to the same degree by going to a more diversified equity portfolio" (of course, I believe his equity portfolio is already highly diversified, but lets pretend it isn't), I'd again suggest there was no basis for that comparison.

So, it doesn't matter to me whether you want to claim diversifying with equities is better for managing risk than diversifying with bonds, or vice-versa--either way I am going to suggest you have no valid basis for such a comparison.

I’m not trying to put words in your mouth; I am simply trying to understand why you do not believe you can compare the two approaches.

As I have explained before, to make such a comparison you would need to be able to reduce this "universe of risks" to one single, simple, quantifiable measure. I don't think that is possible, and hence there is no such basis for comparison.

Consider, for example, your phrase above (not mine, I would emphasize): "reduced risk to the same degree or even more." It is quite clear in this phrase you are assuming risk is subject to a single measure, such that you can determine the "degree" to which different risk management strategies have reduced risk by this measure. Now, drop the assumption that risk is measurable in that way, and this phrase no longer has meaning.

To help you see this, consider two really different things. For example, consider sandpaper and a refrigerator. The sandpaper is used to make something smooth, and the refrigerator is used to make the same thing cold. So, which has affected this thing to a greater "degree"?

The question is meaningless, of course. Why? Because making something smooth and making it cold are two different things, and there is no way to compare the two by the same measure.

Now, in this case you have thrown a term around all the bad things that can happen to someone's investments and described them all as "risks". But in my view, it is no more the case that reducing risks of one kind can necessarily be compared to reducing risks of another kind as that one can compare making something smooth or alternatively cold.

Hopefully that helps. And please, if it isn't clear, just ask me--don't put words in my mouth.

[johndcraig]

Brian

The risks faced by investors are clearly varied. Most investors face a similar universe of risks, particularly if the major risks are being considered. I agree that different investments might better deal with different types of risks. However, every investor is required to somehow consider the entire universe of risks, and it is of little help to say every risk requires a different investment solution. I am attempting to deal with this typical case where an investor is trying to determine a single approach to best deal with the myriad of risks out there.

I suspect that if a poll were taken 90% of those polled and 100% of the authors/advisors would agree that the most effective way to reduce this universe of risks is to switch out of equities and into conservative fixed as described above. That is not to say that most do not generally believe in equity diversification benefits; it is simply to say that they have no doubt that an equity fixed shift is the better way to avoid risk.

You state “So, it doesn't matter to me whether you want to claim diversifying with equities is better for managing risk than diversifying with bonds, or vice-versa--either way I am going to suggest you have no valid basis for such a comparison.” Does this mean you disagree with the approach suggested in the above two paragraphs?

Regarding the Larry case you say,

Specifically, in this case, you claim: "you may be saying there is no basis for believing that he couldn’t have reduced risk to the same degree or even more by going to a diversified 80% equity portfolio."

Of course, that is not what I would say … If someone claimed that Larry could "reduce risk to the same degree by going to a more diversified equity portfolio" … I'd again suggest there was no basis for that comparison.

I am still really confused regarding what you are saying. If there is no basis for comparison, then isn't it reasonable to assume that diversification might reduce risk to the same degree or even more (or even less for that matter)? Isn’t any outcome possible if there is no basis for comparison?

John

[HockeyMike35]

I am attempting to deal with this typical case where an investor is trying to determine a single approach to best deal with the myriad of risks out there.

I think this false assumption is what is causing a great deal of your confusion with this topic. I do not think it is advisable for anyone to limit themselves to a single approach. It makes more sense to use all the tools available. Why would you suggest people only use one method rather then using complimenting methods to deal with different types of risk?

Good Luck,

Mike

[BrianTH]

I am attempting to deal with this typical case where an investor is trying to determine a single approach to best deal with the myriad of risks out there.

I don't think this should be, or is in fact, the "typical case". There is just no reason to limit oneself to a single approach to managing risk, and in my experience few if any people do.

You state “So, it doesn't matter to me whether you want to claim diversifying with equities is better for managing risk than diversifying with bonds, or vice-versa--either way I am going to suggest you have no valid basis for such a comparison.” Does this mean you disagree with the approach suggested in the above two paragraphs?

I'm not sure what you mean with this question. Insofar as the first paragraph suggests an investor should pick a single approach, as noted I disagree. The second paragraph was your guess at what a poll would say. As an aside, I strongly suspect that if you gave them an option in this poll where they could say something like "do both", that option would be the clear winner. Again, though, if you are asking whether I think one should pick a single approach to managing risk, then the answer is no.

I am still really confused regarding what you are saying. If there is no basis for comparison, then isn't it reasonable to assume that diversification might reduce risk to the same degree or even more (or even less for that matter)? Isn’t any outcome possible if there is no basis for comparison?

Again, if there is no basis for comparison, then the phrase "might reduce risk to the same degree or even more" is simply inapplicable. To extend my analogy, that would be like saying, "if there is no basis for comparing what the sandpaper did to what the refrigerator did, then isn't it reasonable to assume the sandpaper made the thing as cold as the refrigerator, and the refrigerator made the thing as smooth as the sandpaper?"

[johndcraig]

PLEASE

Now you are trying to put words into my mouth. I was never suggesting that an investor should pick one approach or the other; it is absolutely fine if an investor chooses to do both. The poll I was suggesting deals with what the individuals believe to be the single best approach; nothing at all about an exclusive approach.

So please respond to the earlier post considering the matter as I originally intended.

Regarding the Larry situation. Larry decided to go with 20% equities. I’m sure if asked he would say this was done because he believes the substantial equity reduction was the best way to reduce the universal risks as discussed. He has said as much in earlier posts. I have no clue what your sandpaper refrigerator answer means; can you try explaining this in specific investment terms that Larry might consider?

John

[bob90245]

You state “So, it doesn't matter to me whether you want to claim diversifying with equities is better for managing risk than diversifying with bonds, or vice-versa--either way I am going to suggest you have no valid basis for such a comparison.” Does this mean you disagree with the approach suggested in the above two paragraphs?

I'm not sure what you mean with this question. Insofar as the first paragraph suggests an investor should pick a single approach, as noted I disagree. The second paragraph was your guess at what a poll would say. As an aside, I strongly suspect that if you gave them an option in this poll where they could say something like "do both", that option would be the clear winner. Again, though, if you are asking whether I think one should pick a single approach to managing risk, then the answer is no.

First let me say, that I've been following the exchanges between John and Brian with great interest. So I hope the following comment does not detract, but rather illustrate a point.

I think the "do both" option is the one I would choose, as well. Although, I do lean toward the equity-fixed split as the main driver to reduce portfolio risk. It works well during bear market declines.

I would consider equity diversification as the secondary driver to reduce portfolio risk. A widely diversified equity portfolio worked better than only US Large stocks during the 2000-2002 bear market. However in the 1973-1974 bear market, most every equity asset class fell in synchronous fashion.

Here's chart for comparison:



Source: fundadvice.com

[johndcraig]

Bob

As stated above, I was asking about the single best approach and was not suggesting an exclusive approach; I personally use both, and would never suggest otherwise. You have stated that “I do lean toward the equity-fixed split as the main driver to reduce portfolio risk. It works well during bear market declines." My guess is that Brain is in a very small minority that thinks there is no basis for making this comparison.

John

[HockeyMike35]

A switch of part of the portfolio from equities to fixed (ST or TIPS) will materially lessen the risk, but IMO diversification among risky assets, including international, will not.

As stated above, I was asking about the single best approach and was not suggesting an exclusive approach; I personally use both, and would never suggest otherwise.

John,

You keep changing your argument. I think the conversation could be more helpful if you would stick to a point. If you have changed your mind based on the conversation please say so. Changing your message and claiming that is what you said all along is confusing and counter productive. You have repeatedly said that divisification between different equities does not reduce risk. Now you calim the that you use this method. Please explain.

Good Luck,

Mike

[BrianTH]

PLEASE

Now you are trying to put words into my mouth. I was never suggesting that an investor should pick one approach or the other; it is absolutely fine if an investor chooses to do both. The poll I was suggesting deals with what the individuals believe to be the single best approach; nothing at all about an exclusive approach.

So please respond to the earlier post considering the matter as I originally intended.

I honestly do not know what you are asking me. For the reasons I have given, I don't think the poll you designed is a good one, and I do not think it would give people an opportunity to express their true views on the subject.

By the way, how is asking for the "single best approach" not an "exclusive" question? By forcing them to choose a "single" approach to call the best, you are preventing them from responding (as I would) that the best approach would actually be to use both of the approaches you listed.

Regarding the Larry situation. Larry decided to go with 20% equities. I’m sure if asked he would say this was done because he believes the substantial equity reduction was the best way to reduce the universal risks as discussed. He has said as much in earlier posts.

I don't think that is right, and I again think you are better off letting people speak for themselves.

I have no clue what your sandpaper refrigerator answer means; can you try explaining this in specific investment terms that Larry might consider?

First, I think you should try to understand my analogies. I think your assumption that all investment risks can be captured by a single measure is so strong that in an investment context, you just cannot imagine anything else. That is why I think it would help for you to consider an analogy from outside investing.

But I agree the sandpaper/refrigerator analogy is a little odd (that was intentional, but maybe not so helpful). So, as an alternative, I think it would be useful for you to indulge me by either answering this question, or explaining why it is not a good question:

Which is the best tool, a hammer or a screwdriver?

And although you may disagree, I really have tried to respond to your questions in the best way I can. So I think it would be fair for you to do the same and try to respond to the question above.

Anyway, with investing: here is a simple example which you might even like. TIPS help manage the risk of unexpected inflation. TBonds help manage the risk of unexpected deflation.

Now, I chose two kinds of bonds because I know you like bonds, and are willing to admit that bonds manage risk, but so far are unwilling to admit that anything else can manage risk. So, here is another question that I would like you to try:

Which is the single best approach for managing risk: holding TIPS or holding TBonds?

[BrianTH]

I think the "do both" option is the one I would choose, as well. Although, I do lean toward the equity-fixed split as the main driver to reduce portfolio risk. It works well during bear market declines.

I would consider equity diversification as the secondary driver to reduce portfolio risk. A widely diversified equity portfolio worked better than only US Large stocks during the 2000-2002 bear market. However in the 1973-1974 bear market, most every equity asset class fell in synchronous fashion.

I guess we are starting to conduct an actual poll, which is infinitely preferable to guessing at responses. Maybe if we develop a good set of options, we can actually put it in an "official" forum poll.

Anyway, along those lines: what do you feel about diversification by country (the nominal subject of this thread, in fact). You could cut this across both equity and bonds if you like (and any other asset class you might have in mind), and just view it as a dimension of your portfolio along which you could diversify.

And to be clear, I obviously do not like John's "pick the single best" question, so please feel free not to answer in that fashion. I'm just curious about how important you think country diversification might be, including in light of the Japan example.

[HockeyMike35]

I would also vote to use both methods. I think country diversification can provide a material difference. It all depends on the type of risk that presents itself.

Good Luck,

Mike

[johndcraig]

Another try

By forcing them to choose a "single" approach to call the best, you are preventing them from responding (as I would) that the best approach would actually be to use more than one of the approaches you listed.

OK, you would suggest a combination of approaches; we agree because so would I. Does it matter that the equity piece is 90% or 10%? Why or why not. How relevant and important is this piece? Can you please get off this exclusivity point and respond to the original question as originally intended?

I have no problem with your suggestion that, when S&D is used, different investments might better cover different risks. I have said so before. Continuing to dwell on this agreed point simply continues to obfuscate the issues. I won’t disagree with any diversification mix of equities or fixed that you suggest, but that isn’t the point. There is a universe of risks out there and a switch from equities to fixed basically covers any such risks that result in market declines. As Bob points out, avoiding market declines by diversification among equities is more questionable, though it might work to some extent if you guess right about which asset classes to hold; shifting equities to fixed covers all asset classes. That is the reason that all of the authors suggest that the equity-fixed split explains (94%) of market returns, and that is why I am certain that they all agree that shifting equity to fixed is clearly the single best method for avoiding universal risk.

John

Mike, you have ignored the word “materially”