"The long run is only longer than the short run, not safer"

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Random Walker
Posts: 1920
Joined: Fri Feb 23, 2007 8:21 pm

"The long run is only longer than the short run, not safer"

Post by Random Walker » Fri May 12, 2017 6:48 pm

I'm currently reading a book called Retirement Portfolios by Michael Zwecher. The author quotes a friend of his when he writes "The long run is only longer than the short run, not safer". I think that's a succinct and clever statement of an important concept. Stocks are risky no matter the time frame. Certainly the longer the holding period, the more likely a positive outcome, but also the wider dispersion of potential returns. At any point in time stocks have the potential to plummet. Moreover, we don't anchor on the starting value, we tend to focus on the current value of our portfolios or the highest value of our portfolios. So indeed, stocks are always risky. Perhaps especially worth noting for people within a decade of retirement given current valuations.

Dave

ryman554
Posts: 712
Joined: Sun Jan 12, 2014 9:44 pm

Re: "The long run is only longer than the short run, not safer"

Post by ryman554 » Sat May 13, 2017 12:30 am

Random Walker wrote:I'm currently reading a book called Retirement Portfolios by Michael Zwecher. The author quotes a friend of his when he writes "The long run is only longer than the short run, not safer". I think that's a succinct and clever statement of an important concept. Stocks are risky no matter the time frame. Certainly the longer the holding period, the more likely a positive outcome, but also the wider dispersion of potential returns. At any point in time stocks have the potential to plummet. Moreover, we don't anchor on the starting value, we tend to focus on the current value of our portfolios or the highest value of our portfolios. So indeed, stocks are always risky. Perhaps especially worth noting for people within a decade of retirement given current valuations.

Dave


?? No, that does not make sense to me at all... If it *is* a random walk, then the central limit theorem (mostly) applies and the long-run variance is much smaller than the variance of an individual time frame. Something like 1/sqrt(N) if I recall correctly.

How do we see this in action?

https://novelinvestor.com/best-time-sp-500/ Look at the probability of the total return as a function of time. The probability of a negative return is high (but < 50%) in 1 year increments, but virtually non-existent in 15 year increments.

I do agree that the tails get wider (as they must), but the likelihood of the tails drop. Admittedly, that is little solace for the folks (say, Japan, 1989, or world+dog 2008) that get caught out by one of the tails, but that's the chance we all take going in.

Your second statement: being anchored on the current value and not the starting value is valid: We do agree that each and every year you roll the dice and takes your chances. Which is why I track total return to "remind" myself that I'm still ahead of the game even with a 50% drop.

If that's the crux of the argument, then perhaps, but I fear the statement would put off folks from investing *at all*, when the "chanciness" of the returns with (historical) larger than 1 expected return imply that long-range investing is as close to a sure thing as you can get -- far from chancy.

Oh, and one more edit: which means, unless you oversave, you should plan on an infinite retirement -- and with that planning you're actually better off with the higher-returning, higher-variabilty option (so long as the variability relative to the expected growth trend is muted over your timeframe, as it is with stocks in SNP 500) than moving to "safer" investments.

Ari
Posts: 430
Joined: Sat May 23, 2015 6:59 am

Re: "The long run is only longer than the short run, not safer"

Post by Ari » Sat May 13, 2017 1:36 am

Random Walker wrote:Certainly the longer the holding period, the more likely a positive outcome, but also the wider dispersion of potential returns.

I must say that "likelihood of a positive outcome" is a lot more relevant to my definition of "risky" than is "dispersion of returns".
All in, all the time.

kmok
Posts: 218
Joined: Thu Apr 25, 2013 10:17 pm

Re: "The long run is only longer than the short run, not safer"

Post by kmok » Sat May 13, 2017 1:37 am

For stocks, it is true that the probability of semi-bad outcome decreases over time. However, the probability of very bad outcomes actually increases.

Another way to look at it is, if stocks are less risky over increasing time horizons, then put options would become cheaper with longer expiration date. However, they don't.

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: "The long run is only longer than the short run, not safer"

Post by NiceUnparticularMan » Sat May 13, 2017 6:18 am

ryman554 wrote:The probability of a negative return is high (but < 50%) in 1 year increments, but virtually non-existent in 15 year increments.


Yep. If stocks give you a random walk around an upward central trend, the probability of stock returns walking below zero is decreasing over time. A similar result applies to comparing assets: even if you had a perfectly steady asset with, say, half the expected return of stocks, the probability of stock returns walking below that asset's returns is decreasing over time.

That said, stocks are not necessarily just a random walk over time, and there is a non-trivial risk of "fat tails," "black swans," or whatever term you prefer for a scenario in which stocks perform completely outside their normal random-walkish framework. The risk of one of those showing up may be increasing with time.

Hedging such risks is inherently difficult due to their fundamental unpredictability. Moreover, these concepts are in potential conflict--for example, it might be better to have more in non-stocks when such a black swan event occurs, but if always being more in non-stocks has decreased your returns a lot by that point, you may still end up worse off. Ideally you would be looking at assets with similar expected return but different "black swan" exposure--hence international investing, REITs, factor investing if you believe in it, and such.

And if no equities of any sort anywhere are fulfilling expectations even over the long run . . . well, maybe nothing would work, because that's a pretty bad black swan event such that no financial assets might be safe.

User avatar
nisiprius
Advisory Board
Posts: 33783
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: "The long run is only longer than the short run, not safer"

Post by nisiprius » Sat May 13, 2017 7:29 am

Most claims that stocks become safer in the long run ignore compounding. The chart you see over and over again in which the bars are broad at one year and narrow at thirty years ignore compounding. If you include compounding, they explode. If you think the effect of compounding 1% in expenses is big, it's nothing compared to compounding the difference between 3.11% over twenty years (stock returns for the period ending in 1948) versus 17.88% (period ending in 1999).

The charts in which the bottom of the bar doesn't go below zero, seemingly showing that you're loss-proof, are usually charts that ignore inflation and taxes.

To me, the important question is, "what can I count on?" That is, what is the expected low-percentile return, maybe not the absolute-worst-case scenario, but the real-world reasonably-likely bad-case scenario. And the reason why debates go on endlessly is that the rough answer from the past is that the pessimistic end of the dispersion of returns is just about the same for stocks and for bonds. Again, looking only at random samples from the past, the chances of doing worse than bonds with stocks isn't very high, but the statistics also say that you have a serious chance of being in serious trouble if you count on stocks doing better. For example, if you reduce your savings rate and invest more aggressively to make up for it.

A lot of the happy-talk about the long run traces back to the title, and to summaries of summaries of summaries, of Jeremy Siegel's book, Stocks for the Long Run. It makes the subtle and limited claim that there is active mean reversion in stock market returns, that periods of unusually high returns are more likely to be followed by periods of unusually low returns than a random walk would predict, and vice versa. Therefore, the long-term returns are narrower and more stable than they would be if stocks followed a random walk. Nevertheless, even with mean reversion, they still become riskier in the long run and Siegel has been perfectly clear about that himself.

This is what Jeremy Siegel himself has said.
Let me show you the graph that is in my book, Stocks for the Long Run, that I think is so critical to that. We have it on my right, on your left. It is the standard deviation of stocks, bonds, and treasury bills after inflation, measured over the longest period we have ever done research on. 200 years. That’s the one year standard deviations, those are the two year average standard deviations, five, ten, 20, and 30. Now, many of you who have gone to my presentations have probably seen that slide before. Now, one thing I should make very clear, I never said that that means stocks are safer in the long run. This is the standard deviation of average annual returns. We know the standard deviation of the average goes down when you have more periods. Even if it’s random walk, it goes down. What I pointed out here is that the standard deviation for stocks goes down twice as much— twice as fast as random walk theory would predict. In other words, they are relatively safer in the long run than random walk theory would predict. Doesn’t mean they’re safe.
The other detail that seems to be widely ignored, and I don't know if that's because it's been discredited or whether it's just something people don't like to keep in mind, is the Pastor and Stambaugh paper which says that although mean reversion acts to reduce the long-term risk of stock investing, there are other factors that increase it, and that they outweigh mean reversion. One of the big ones is uncertainty about the value of the mean itself.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

Levett
Posts: 4134
Joined: Fri Feb 23, 2007 2:10 pm
Location: upper Midwest

Re: "The long run is only longer than the short run, not safer"

Post by Levett » Sat May 13, 2017 8:31 am

Zwecher's is a very good book.

Lev

User avatar
knpstr
Posts: 1735
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: "The long run is only longer than the short run, not safer"

Post by knpstr » Sat May 13, 2017 8:36 am

nisiprius wrote:Again, looking only at random samples from the past, the chances of doing worse than bonds with stocks isn't very high, but the statistics also say that you have a serious chance of being in serious trouble if you count on stocks doing better. For example, if you reduce your savings rate and invest more aggressively to make up for it.


So if the chances of doing worse than bonds isn't very high, what are the chances of doing better than bonds?
What do you mean by a "serious chance of being in serious trouble"?
What if you don't reduce your savings rate but rather maintain an otherwise equally aggressive savings rate and invest more "aggressively"?
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

dh
Posts: 214
Joined: Sun Mar 13, 2011 8:01 pm

Re: "The long run is only longer than the short run, not safer"

Post by dh » Sat May 13, 2017 8:44 am

Random Walker wrote:I'm currently reading a book called Retirement Portfolios by Michael Zwecher. The author quotes a friend of his when he writes "The long run is only longer than the short run, not safer". I think that's a succinct and clever statement of an important concept. Stocks are risky no matter the time frame. Certainly the longer the holding period, the more likely a positive outcome, but also the wider dispersion of potential returns. At any point in time stocks have the potential to plummet. Moreover, we don't anchor on the starting value, we tend to focus on the current value of our portfolios or the highest value of our portfolios. So indeed, stocks are always risky. Perhaps especially worth noting for people within a decade of retirement given current valuations.

Dave


Hi, Dave.
I haven't read Zwecher's book, yet his idea seems consistent with an idea shared by Prof. Zvi Bodie. I am paraphrasing, but Bodie argues that not only are stocks risky today, they are risky at any point out in time. Logical, tautological, ...? While it doesn't influence my IPS, it makes sense when I read articles about the importance of the sequence of returns early in retirement when one is drawing down their portfolio rather than adding to.

soupcxan
Posts: 158
Joined: Tue Apr 14, 2009 8:54 am

Re: "The long run is only longer than the short run, not safer"

Post by soupcxan » Sat May 13, 2017 8:50 am

Anyone who thinks the risk of the market declines over time should go price some put options on S&P 500 ETFs. If your belief is correct, these securities should be cheaper the further out on the calendar you go.

If you still believe this, why don't you sell some long dated S&P puts yourself? The premium should be essentially free money to you because there is minimal risk as long as you make the time to expiration long enough, right?

dbr
Posts: 22870
Joined: Sun Mar 04, 2007 9:50 am

Re: "The long run is only longer than the short run, not safer"

Post by dbr » Sat May 13, 2017 8:55 am

ryman554 wrote:

?? No, that does not make sense to me at all... If it *is* a random walk, then the central limit theorem (mostly) applies and the long-run variance is much smaller than the variance of an individual time frame. Something like 1/sqrt(N) if I recall correctly.

That is not the measure of concern. The measure is the end point wealth, the dispersion of whoch increases by sqrt(N) with time.

How do we see this in action?

https://novelinvestor.com/best-time-sp-500/ Look at the probability of the total return as a function of time. The probability of a negative return is high (but < 50%) in 1 year increments, but virtually non-existent in 15 year increments.

I do agree that the tails get wider (as they must), but the likelihood of the tails drop. Admittedly, that is little solace for the folks (say, Japan, 1989, or world+dog 2008) that get caught out by one of the tails, but that's the chance we all take going in.

Correct, these last two statements are to the point sort of. Chances of a loss is also not really the measure. The measure would be the chance of falling short of an objective, that objective being hoped for significant gain. But, you are correct about the fact that if the average return exceeds what is needed to hit that end-point target, then the chances of failing decrease with time. On the other hand, if the average return is not enough to hit the target, the chances of failure increase with time. That is one reason bonds can be more risky than stocks, they don't return enough to meet objectives.

Your second statement: being anchored on the current value and not the starting value is valid: We do agree that each and every year you roll the dice and takes your chances. Which is why I track total return to "remind" myself that I'm still ahead of the game even with a 50% drop.

We all agree this is very important and often ignored. The short statement is that I would rather have $10M and lose half of it than have $1M and lose none of it.

If that's the crux of the argument, then perhaps, but I fear the statement would put off folks from investing *at all*, when the "chanciness" of the returns with (historical) larger than 1 expected return imply that long-range investing is as close to a sure thing as you can get -- far from chancy.

It is still difficult to get this picture generated in everyone's mind and then decide how to manage financial planning with so much uncertainty. A point to add is that the actual outcome among all this uncertainty will be a matter of history yet to be and not known at the start. It is not as if the uncertainty can be controlled short of investing less in stocks for more certainty and less return.


Oh, and one more edit: which means, unless you oversave, you should plan on an infinite retirement -- and with that planning you're actually better off with the higher-returning, higher-variabilty option (so long as the variability relative to the expected growth trend is muted over your timeframe, as it is with stocks in SNP 500) than moving to "safer" investments.

User avatar
whodidntante
Posts: 2010
Joined: Thu Jan 21, 2016 11:11 pm

Re: "The long run is only longer than the short run, not safer"

Post by whodidntante » Sat May 13, 2017 8:59 am

soupcxan wrote:Anyone who thinks the risk of the market declines over time should go price some put options on S&P 500 ETFs. If your belief is correct, these securities should be cheaper the further out on the calendar you go.

If you still believe this, why don't you sell some long dated S&P puts yourself? The premium should be essentially free money to you because there is minimal risk as long as you make the time to expiration long enough, right?


No, because options can be exercised at any time before expiration.

User avatar
knpstr
Posts: 1735
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: "The long run is only longer than the short run, not safer"

Post by knpstr » Sat May 13, 2017 9:10 am

soupcxan wrote:If you still believe this, why don't you sell some long dated S&P puts yourself? The premium should be essentially free money to you because there is minimal risk as long as you make the time to expiration long enough, right?


Where can I sell some 2047 european put options betting that the market won't be lower in 2047 than it is today?
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

soupcxan
Posts: 158
Joined: Tue Apr 14, 2009 8:54 am

Re: "The long run is only longer than the short run, not safer"

Post by soupcxan » Sat May 13, 2017 9:17 am

whodidntante wrote:
soupcxan wrote:Anyone who thinks the risk of the market declines over time should go price some put options on S&P 500 ETFs. If your belief is correct, these securities should be cheaper the further out on the calendar you go.

If you still believe this, why don't you sell some long dated S&P puts yourself? The premium should be essentially free money to you because there is minimal risk as long as you make the time to expiration long enough, right?


No, because options can be exercised at any time before expiration.


Ok then sell some EUROPEAN options.

Random Walker
Posts: 1920
Joined: Fri Feb 23, 2007 8:21 pm

Re: "The long run is only longer than the short run, not safer"

Post by Random Walker » Sat May 13, 2017 9:18 am

Dh,
I saw that Bodie was acknowledged in the book by the author.

Dave

soupcxan
Posts: 158
Joined: Tue Apr 14, 2009 8:54 am

Re: "The long run is only longer than the short run, not safer"

Post by soupcxan » Sat May 13, 2017 9:22 am

knpstr wrote:
soupcxan wrote:If you still believe this, why don't you sell some long dated S&P puts yourself? The premium should be essentially free money to you because there is minimal risk as long as you make the time to expiration long enough, right?


Where can I sell some 2047 european put options betting that the market won't be lower in 2047 than it is today?


I'm sure there's at least one insurance company that's interested in cheap insurance for its portfolio. I would start with the Japanese insurers.

When people start talking like "stocks are risk free as long as you hold them long enough" we've gotta be close to a bear market.

User avatar
knpstr
Posts: 1735
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: "The long run is only longer than the short run, not safer"

Post by knpstr » Sat May 13, 2017 9:37 am

soupcxan wrote:When people start talking like "stocks are risk free as long as you hold them long enough" we've gotta be close to a bear market.


I don't think anyone is saying it is "risk free" of course nothing is risk free.
But it is close to what Warren Buffett has been saying his whole career.
In a similar tone, John Templeton's prediction was Dow 1,000,000 by year 2100
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

dbr
Posts: 22870
Joined: Sun Mar 04, 2007 9:50 am

Re: "The long run is only longer than the short run, not safer"

Post by dbr » Sat May 13, 2017 9:41 am

knpstr wrote:
soupcxan wrote:When people start talking like "stocks are risk free as long as you hold them long enough" we've gotta be close to a bear market.


I don't think anyone is saying it is "risk free" of course nothing is risk free.
But it is close to what Warren Buffett has been saying his whole career.
In a similar tone, John Templeton's prediction was Dow 1,000,000 by year 2100


Buffet has correctly(?) pointed out that bonds are risky and stocks are safe because bonds do not return enough to meet people's reasonable objectives and stocks do, at least stocks in US companies historically and in his estimate for the future of concern to investors today.

User avatar
knpstr
Posts: 1735
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: "The long run is only longer than the short run, not safer"

Post by knpstr » Sat May 13, 2017 9:48 am

dbr wrote:
knpstr wrote:
soupcxan wrote:When people start talking like "stocks are risk free as long as you hold them long enough" we've gotta be close to a bear market.


I don't think anyone is saying it is "risk free" of course nothing is risk free.
But it is close to what Warren Buffett has been saying his whole career.
In a similar tone, John Templeton's prediction was Dow 1,000,000 by year 2100


Buffet has correctly(?) pointed out that bonds are risky and stocks are safe because bonds do not return enough to meet people's reasonable objectives and stocks do, at least stocks in US companies historically and in his estimate for the future of concern to investors today.


Yes.

Incidentally, Buffett has made the very bet this thread is talking about. He sold 15-year european puts in the S&P 500, FTSE, euro stoxx 50, nikkei 225; receiving premiums of $4.5 B and investing the proceeds

He said options get mispriced due to the false precision given by black-scholes
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: "The long run is only longer than the short run, not safer"

Post by NiceUnparticularMan » Sat May 13, 2017 9:58 am

To the extent real stock returns are bounded (I prefer that framing to claiming they mean-revert), it is because they are ultimately based on the real productivity of the assets owned by the respective companies, and in most scenarios there is only so little or so much return those assets can produce in the long run.

But there are in fact scenarios where stocks in a particular country could nonetheless "go to zero" and stay there. Generally in such scenarios you would not want to be in ANY financial assets based in that same country.

And again, if stocks in the whole world "go to zero", you may just be out of luck entirely with financial assets.

azanon
Posts: 1429
Joined: Mon Nov 07, 2011 10:34 am
Location: Little Rock, AR
Contact:

Re: "The long run is only longer than the short run, not safer"

Post by azanon » Sat May 13, 2017 10:54 am

I agree with the OP. I like the question that Zvi Bodie wants others to ask themselves, which is what is the least amount of risk one can take and still meet your financial goals. I think that makes a lot of sense.

So, you might be able to save enough or work long enough that you can't really justify a portfolio that's more aggressive than, say, a maximally risk-adjusted return portfolio that'll have you ending up with something like 30% equities, give or take. If you're running a common 60/40 or even higher equities, at least be able to say you also need that level of expected return (and associated risk), because it's not optimal from a risk-adjusted basis. Or if you really want to hold more equities, but can't really justify it from a need basis, at least try to limit yourself to a global market portfolio weighting, which would be somewhere in the neighborhood of 45% equities.

Ari
Posts: 430
Joined: Sat May 23, 2015 6:59 am

Re: "The long run is only longer than the short run, not safer"

Post by Ari » Sat May 13, 2017 11:23 am

What's the actionable advice of this realization? Don't hold more stocks when you're young and less when you're closer to retirement?
All in, all the time.

MnD
Posts: 3028
Joined: Mon Jan 14, 2008 12:41 pm

Re: "The long run is only longer than the short run, not safer"

Post by MnD » Sat May 13, 2017 11:34 am

How did a "safe" bond-heavy portfolio work out for class of 1966 retirees?

Random Walker
Posts: 1920
Joined: Fri Feb 23, 2007 8:21 pm

Re: "The long run is only longer than the short run, not safer"

Post by Random Walker » Sat May 13, 2017 12:09 pm

Ari,
I would say the actionable advice is for an individual to critically evaluate how much risk they are taking compared to ability, willingness, need on an ongoing basis. Perhaps re-evaluate every couple to few years. I strongly believe we anchor on the highest value of our portfolios and the current value is always at risk.
With recent past returns, current valuations, and future expected equity returns, I think it's reasonable to believe the potential upside is modest and potential downsides really down.
For myself, I pretty much envision a one way glide path of decreasing % equity over time. As investors we can't control much. We can control when we make allocation changes though. Rather than blindly follow a pre defined glide path determined by age/date, I think it makes sense to rarely make one way changes in response to size of portfolio, current valuations, age, and increased clarity on needs versus wants as we age. As one gets more clear on needs versus wants, I think Zwecher's concept of a floor to the portfolio makes great sense. Bernstein talks a lot about this with liability matching portfolios.

Dave

Levett
Posts: 4134
Joined: Fri Feb 23, 2007 2:10 pm
Location: upper Midwest

Re: "The long run is only longer than the short run, not safer"

Post by Levett » Sat May 13, 2017 1:10 pm

Random Walker observed:

"I strongly believe we anchor on the highest value of our portfolios and the current value is always at risk."

Sure seems that way.

Any reader of Zwecher's book should look at his comments on "Murphy's Law of Timing Retirement." Now might be an especially good time to have a look. ;-)

Lev

User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

Re: "The long run is only longer than the short run, not safer"

Post by backpacker » Sat May 13, 2017 1:21 pm

kmok wrote:If stocks are less risky over increasing time horizons, then put options would become cheaper with longer expiration date. However, they don't.

soupcxan wrote:Anyone who thinks the risk of the market declines over time should go price some put options on S&P 500 ETFs. If your belief is correct, these securities should be cheaper the further out on the calendar you go.


The put option argument was first made by Zvi Bode in his paper "On the Risk of Stocks in the Long Run" (link). The argument is fallacious because it completely misunderstands how option pricing works.

Suppose you want me to sell you a 50 year put on stocks. The put pays out if stocks returns less than risk free bonds. Selling you that put has the same risk for me as buying the S&P 500 for the same period. So, if I'm going to sell you this put, the expected return of selling it had better be as high as buying stocks. That means that the cost of the put to you will be outrageous. I am going to charge you the entire expected outperformance of stocks against risk free bonds over 50 years.

In fact, the better I expect stocks to do and the less risky I think they are, the more I am going to value buying stocks. Since selling you a put has the same risk as buying stocks, that means that the better I think stocks will do and the less risk I think they have, the more I will charge you. This predicts that if the market thought that stocks were safer the the long run, it would charge more for long dated puts, not less.

So we get the complete opposite of what Bode was trying to show. The safer and better I think stocks are in the long run, the more I am going to charge for puts.
Last edited by backpacker on Sat May 13, 2017 2:39 pm, edited 5 times in total.

Random Walker
Posts: 1920
Joined: Fri Feb 23, 2007 8:21 pm

Re: "The long run is only longer than the short run, not safer"

Post by Random Walker » Sat May 13, 2017 1:24 pm

Levett,
Totally agree! For those who haven't read it, Murphys Law of Retirement is the following. The stock market goes through an extended bull, people's portfolios swell, and they consider or do retire. Problem is that at this point, after huge returns and generous valuations, future expected returns are much lower and the new retiree is most likely to experience poor returns early in retirement. So many retirees are extra susceptible to bad sequence of returns at worst time naturally because they enter retirement after bull markets. That's why may be especially good for preretirees to consider taking risk off table presently.

Dave

garlandwhizzer
Posts: 1559
Joined: Fri Aug 06, 2010 3:42 pm

Re: "The long run is only longer than the short run, not safer"

Post by garlandwhizzer » Sat May 13, 2017 1:35 pm

In the really long run, we're all dead. Although it is true that in the long run stocks have a wider dispersion of outcomes that bonds, that doesn't keep me from making rational assumptions about portfolio construction. It is entirely easonable to expect stocks to outperform bonds in the long run. If you need considerable real growth of portfolio over the coming decades you may well have to take risk to get there starting from where we are now. The quest to reduce risk and volatility results in an emotionally comfortable lack of volatility now and for a some years into the future, but those who overdo it may well run out of money in their extended lifespans. If you run out of money at that point there is no alternative to poverty, let alone emotional comfort. That is a very real risk that needs to be considered by those with heavily tilted bond portfolios which don't expose them to emotional distress in the short run but may well be disastrous in the long run.

Garland Whizzer

User avatar
nedsaid
Posts: 8262
Joined: Fri Nov 23, 2012 12:33 pm

Re: "The long run is only longer than the short run, not safer"

Post by nedsaid » Sat May 13, 2017 4:20 pm

garlandwhizzer wrote:In the really long run, we're all dead. Although it is true that in the long run stocks have a wider dispersion of outcomes that bonds, that doesn't keep me from making rational assumptions about portfolio construction. It is entirely easonable to expect stocks to outperform bonds in the long run. If you need considerable real growth of portfolio over the coming decades you may well have to take risk to get there starting from where we are now. The quest to reduce risk and volatility results in an emotionally comfortable lack of volatility now and for a some years into the future, but those who overdo it may well run out of money in their extended lifespans. If you run out of money at that point there is no alternative to poverty, let alone emotional comfort. That is a very real risk that needs to be considered by those with heavily tilted bond portfolios which don't expose them to emotional distress in the short run but may well be disastrous in the long run.

Garland Whizzer


As Alan Greenspan might say, this is a conundrum. I would characterize it as a dilemma or a quandary. What makes it even more difficult is that there is not an either/or answer. It is not a yes or no. It is more like turning a dial and deciding how much will do the trick. It is picking the right point on a continuum. It is more difficult in that the future is unknown. Nirvana would be retiring in 1982, just before gigantic stock and bond bull markets, it would be hard to screw that up. Heck would be retiring in about 1968, just before an era not too kind on either stocks or bonds.

The Nedsaid solution is throwing enough asset classes in the portfolio hoping that something in there will work if things go badly wrong. That and crossing your fingers. It is truly a difficult thing to be 100% prepared for. Hint: Be nice to your kids and other family members.
A fool and his money are good for business.

soupcxan
Posts: 158
Joined: Tue Apr 14, 2009 8:54 am

Re: "The long run is only longer than the short run, not safer"

Post by soupcxan » Sun May 14, 2017 5:29 pm

backpacker wrote:Selling you that put has the same risk for me as buying the S&P 500 for the same period.


Sorry, what?

Selling a 30 year European put on the S&P 500 with a strike price of 2390 has a completely different risk/payoff profile compared to a long-only purchaser of the S&P 500 at 2390. They don't have the same risk at all.

User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

Re: "The long run is only longer than the short run, not safer"

Post by backpacker » Sun May 14, 2017 5:52 pm

soupcxan wrote:
backpacker wrote:Selling you that put has the same risk for me as buying the S&P 500 for the same period.


Sorry, what?

Selling a 30 year European put on the S&P 500 with a strike price of 2390 has a completely different risk/payoff profile compared to a long-only purchaser of the S&P 500 at 2390. They don't have the same risk at all.


Think about it in reverse if that helps. The combination of owning stocks and a put at the risk free rate for 30 years (or whatever) has the same risk as owning risk free bonds over the same period. The put guarantees that you can't do worse than risk free bonds. But then, if the put cost less than the (discounted) difference in expected return between stock and risk free bonds, your portfolio would generated higher returns than risk free bonds without taking any risk. In an efficient market, that's impossible. So the cost of the put will be the entire difference in (discounted) expected returns.

The observation then is that if the market really thinks that stocks are safer in the long run, it will think that stocks are more valuable over longer terms. But then by further application of the above reasoning, puts will be more expensive over longer terms, not less. Bode gets the whole thing backwards.

The paper "On the Risk of Stocks in the Long Run: A Resolution to the Debate?" has a helpful discussion of this. I'm basically just repeating what they say, at least as I understand it.

NibbanaBanana
Posts: 194
Joined: Sun Jan 22, 2017 10:34 pm

Re: "The long run is only longer than the short run, not safer"

Post by NibbanaBanana » Sun May 14, 2017 6:30 pm

"We simply do not know what the future holds, and we must accept the self-evident fact that historic stock market returns have absolutely nothing in common with actuarial tables." - John Bogle

soupcxan
Posts: 158
Joined: Tue Apr 14, 2009 8:54 am

Re: "The long run is only longer than the short run, not safer"

Post by soupcxan » Sun May 14, 2017 6:44 pm

backpacker wrote:
soupcxan wrote:
backpacker wrote:Selling you that put has the same risk for me as buying the S&P 500 for the same period.


Sorry, what?

Selling a 30 year European put on the S&P 500 with a strike price of 2390 has a completely different risk/payoff profile compared to a long-only purchaser of the S&P 500 at 2390. They don't have the same risk at all.


Think about it in reverse if that helps. The combination of owning stocks and a put at the risk free rate for 30 years (or whatever) has the same risk as owning risk free bonds over the same period. The put guarantees that you can't do worse than risk free bonds. But then, if the put cost less than the (discounted) difference in expected return between stock and risk free bonds, your portfolio would generated higher returns than risk free bonds without taking any risk. In an efficient market, that's impossible. So the cost of the put will be the entire difference in (discounted) expected returns.

The observation then is that if the market really thinks that stocks are safer in the long run, it will think that stocks are more valuable over longer terms. But then by further application of the above reasoning, puts will be more expensive over longer terms, not less. Bode gets the whole thing backwards.

The paper "On the Risk of Stocks in the Long Run: A Resolution to the Debate?" has a helpful discussion of this. I'm basically just repeating what they say, at least as I understand it.


Interesting, but I still go back to what Vanguard said in a whitepaper:

Code: Select all

The debate over time diversification has been long running and remains unresolved. However, there is little empirical evidence to support the claim that time moderates the risks inherent in risky assets. In actuality, a longer investment horizon increases the magnitude of potential outcomes, both negative and positive.


http://www.vanguard.com/pdf/icrtd.pdf?2210045172

gilgamesh
Posts: 835
Joined: Sun Jan 10, 2016 9:29 am

Re: "The long run is only longer than the short run, not safer"

Post by gilgamesh » Sun May 14, 2017 7:08 pm

Delete

User avatar
willthrill81
Posts: 2209
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: "The long run is only longer than the short run, not safer"

Post by willthrill81 » Sun May 14, 2017 9:17 pm

dbr wrote:
ryman554 wrote:?? No, that does not make sense to me at all... If it *is* a random walk, then the central limit theorem (mostly) applies and the long-run variance is much smaller than the variance of an individual time frame. Something like 1/sqrt(N) if I recall correctly.

That is not the measure of concern. The measure is the end point wealth, the dispersion of whoch increases by sqrt(N) with time.

How do we see this in action?

https://novelinvestor.com/best-time-sp-500/ Look at the probability of the total return as a function of time. The probability of a negative return is high (but < 50%) in 1 year increments, but virtually non-existent in 15 year increments.

I do agree that the tails get wider (as they must), but the likelihood of the tails drop. Admittedly, that is little solace for the folks (say, Japan, 1989, or world+dog 2008) that get caught out by one of the tails, but that's the chance we all take going in.

Correct, these last two statements are to the point sort of. Chances of a loss is also not really the measure. The measure would be the chance of falling short of an objective, that objective being hoped for significant gain. But, you are correct about the fact that if the average return exceeds what is needed to hit that end-point target, then the chances of failing decrease with time. On the other hand, if the average return is not enough to hit the target, the chances of failure increase with time. That is one reason bonds can be more risky than stocks, they don't return enough to meet objectives.


+1

There is absolutely no disputing that the longer your holding period for stocks, the greater the likelihood that the net return will be positive. This is where the statement that "stocks are just as risky over the long-term" is, at best, potentially misleading, particularly to those not savvy about the issue. The historic likelihood of losing money in any given year with stocks has been about 28%, whereas the same likelihood over a 30 year period has been zero.

I really don't think anyone on this forum really likes the volatility of stocks, though many of us view it as necessary in order to get stocks' historic returns. If we could achieve the historic returns of stocks without their volatility, we'd jump on it. But as things stand, most of us have some need for the returns stocks have provided in order to achieve our goals.

dbr wrote:
ryman554 wrote:Your second statement: being anchored on the current value and not the starting value is valid: We do agree that each and every year you roll the dice and takes your chances. Which is why I track total return to "remind" myself that I'm still ahead of the game even with a 50% drop.

We all agree this is very important and often ignored. The short statement is that I would rather have $10M and lose half of it than have $1M and lose none of it.

If that's the crux of the argument, then perhaps, but I fear the statement would put off folks from investing *at all*, when the "chanciness" of the returns with (historical) larger than 1 expected return imply that long-range investing is as close to a sure thing as you can get -- far from chancy.

It is still difficult to get this picture generated in everyone's mind and then decide how to manage financial planning with so much uncertainty. A point to add is that the actual outcome among all this uncertainty will be a matter of history yet to be and not known at the start. It is not as if the uncertainty can be controlled short of investing less in stocks for more certainty and less return.


Oh, and one more edit: which means, unless you oversave, you should plan on an infinite retirement -- and with that planning you're actually better off with the higher-returning, higher-variabilty option (so long as the variability relative to the expected growth trend is muted over your timeframe, as it is with stocks in SNP 500) than moving to "safer" investments.


+1

A recent study found that Millennials around age 20 overwhelmingly prefer investing their 401k assets into a guaranteed 1% fund rather than stocks. Most Bogleheads would (rightly IMHO) say that's far too conservative for virtually any young person out there. Proclaiming the volatility of stocks, while appropriate to some extent, is likely contributing to non-investors' fear that stocks are no better than a slot machine; many believe exactly that. Such thinking is likely a contributing factor to the 'retirement crisis' so many would-be retirees are dealing with right now and will be for the rest of their lives.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
nisiprius
Advisory Board
Posts: 33783
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: "The long run is only longer than the short run, not safer"

Post by nisiprius » Sun May 14, 2017 9:38 pm

willthrill81 wrote:...Such thinking is likely a contributing factor to the 'retirement crisis' so many would-be retirees are dealing with right now and will be for the rest of their lives...
Unlikely, since nowadays most 401(k) plans default savers into target-date funds with healthy stock allocations. The retirement crisis is probably due to employers that had defined-benefit plans shifting to defined-contribution plans, and workers then putting much less into the defined-contribution plans than employers used to put into the defined-benefit plans.

It's this:

Image

Did you see this thread? What makes it especially interesting is that the young poster is in "an actuarial career," so surely numerate, and well-informed on matters like life expectancy:
Why save so much for when you're old?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
willthrill81
Posts: 2209
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: "The long run is only longer than the short run, not safer"

Post by willthrill81 » Sun May 14, 2017 9:46 pm

nisiprius wrote:
willthrill81 wrote:...Such thinking is likely a contributing factor to the 'retirement crisis' so many would-be retirees are dealing with right now and will be for the rest of their lives...
Unlikely, since nowadays most 401(k) plans default savers into target-date funds with healthy stock allocations.


Considering that only half of Americans own any stocks at all, I find that a bit hard to believe. But even if it is, would-be retirees of today aren't helped by the current trend toward 401k plans defaulting to target-date funds, especially since those funds are getting a bit thin on stocks for those in their 50s and 60s.

nisiprius wrote:The retirement crisis is probably due to employers that had defined-benefit plans shifting to defined-contribution plans, and workers then putting much less into the defined-contribution plans than employers used to put into the defined-benefit plans.


I agree that that's likely another big contributing factor. Americans' savings rates have been in the toilet for many years.

nisiprius wrote:It's this:

Image

Did you see this thread? What makes it especially interesting is that the young poster is in "an actuarial career," so surely numerate, and well-informed on matters like life expectancy:
Why save so much for when you're old?


That thread and situation actually support my thesis. In both circumstances, people either don't want to save or don't want their savings to be subject to short-term declines in value. A short-term, present-oriented mindset exists in both instances.

As another example, my in-laws saved well but were afraid to put their principal at risk at all. They only put their money in FDIC insured investments, and now my MiL is living in close to poverty as a result. People love to talk about the 'sleep test', but sleeping well now may come at a high cost later in life.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Abe
Posts: 1607
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: "The long run is only longer than the short run, not safer"

Post by Abe » Fri May 19, 2017 10:49 am

Sometimes I go here to get a better perspective on things.
Annual Returns on Stock, T.Bonds and T.Bills: 1928 - Current

http://pages.stern.nyu.edu/~adamodar/Ne ... retSP.html

The compounded value of $100 invested in stocks from 1928 until 2016 was $328,645. The compounded value of $100 invested in 10 year treasury bonds from 1928 until 2016 was $7,110. Of course, the trade off to get the higher return with stocks was the higher volatility, but the thing that gets me is an investor in stocks could lose 90% of his investment in 2017 and still be better off than an investor who invested in 10 year bonds.
Slow and steady wins the race.

Post Reply