Larry Swedroe: Disrupted Expectations

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Larry Swedroe: Disrupted Expectations

Post by Random Walker » Mon Aug 06, 2018 9:25 am

http://www.etf.com/sections/index-inves ... nopaging=1

Larry discusses the important topic of estimating expected returns. Valuations matter a lot, and it is important to account for them rather than blindly use historical averages; people planning for retirement using only historical averages may well be disappointed. He goes through the exercise of showing how his firm estimates nominal returns based on CAPE and accounting for inflation with the difference between the yields of 10 year Treasury’s and TIPS. Lower CAPE implies higher expected returns for International Developed and Emerging Markets than for S&P500, but the lower valuations also imply greater risk. Those of us who were hoping for greater returns in these asset classes based on valuations during the first half of this year have seen the risk but not the return.
Larry then looks at current bond yields, and shows expected returns for a typical 60/40 portfolio. They are way lower that what we’ve seen in the past. One can increase the expected return of their portfolio by increasing exposure to size, value, international, EM.
I’ve become a big fan of Monte Carlo analysis. The inputs are critical though for it to be worthwhile. Larry gives an example of this at the end of the essay.

Dave

User avatar
HomerJ
Posts: 11567
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: Disrupted Expectations

Post by HomerJ » Mon Aug 06, 2018 11:05 am

Random Walker wrote:
Mon Aug 06, 2018 9:25 am
Valuations matter a lot, and it is important to account for them rather than blindly use historical averages; people planning for retirement using only historical averages may well be disappointed.
I disagree. Valuations have proven to matter very little. We have seen multiple forecasts based on valuations in the past that turned out to be incorrect. They are not a good prediction tool. It is certainly not necessary to account for them. Many financially smart people have retired successfully without ever worrying about valuations.

I do agree that using historical averages is a bad idea. But I doubt very few people on this board are PLANNING around getting 10% return from stocks, and will see their plan utterly fail if we "only" get 5% instead.

For the first 20 years, you save a good amount, invest in a fairly aggressive stocks/bonds AA (since you have plenty of time) and you get what you get. You don't constantly tinker with your Asset Allocation based on "valuations" or any other metric.

When you get closer to retirement, you probably have a much better idea of what your expenses will be in retirement, and can more realistically see how close you are to achieving a 25x or 30x expenses goal.

I don't think anyone here then calculates "Well, I'm sure to get 10% from stocks for the next ten years, so I only need to save this much a year to make my goal"

Instead we say something like "Well, I'm on track to retire at 60, with my current savings rate, if we get 4% returns going forward. If we get less, I guess I'll have to work until 63 which still isn't too bad, or save more, or reduce my expenses in retirement. If we get more than 4%, I can consider retiring earlier. "

Saving more, working longer, reducing expenses in retirement (downsizing, less trips, etc.) are all options. You get what you get, and you adjust.

But if you start with low expectations already (not historical averages), you don't get upset when someone runs in and proclaims "We're only going to get 5% nominal going forward!"

You say "Really? 5%? Nice."
The J stands for Jay

User avatar
packer16
Posts: 1012
Joined: Sat Jan 04, 2014 2:28 pm

Re: Larry Swedroe: Disrupted Expectations

Post by packer16 » Mon Aug 06, 2018 11:47 am

I think how much valuation matters is dependent upon your investment horizon and your distribution of your contributions. The longer your investment horizon the less valuations matter as long as you are investing in market where as a shareholder you can reap the rewards of your investment like the US. If you compare the US to a growth stock, the growth rate is more important than the price over the long term. If the time horizon is less than 5 to 10 years, valuation can have a larger impact on terminal values than growth rates. As time goes on growth can overcome valuations.

As the timeframe over which you are investing increases, valuations become less important the continued growth. So lump sum investors should care about valuations if they need the money in less than 5 to 10 years. If you are DCAing, then the importance of growth of your initial contributions can offset valuations. Since you are investing over time, the cases of overvaluation are offset by the cases of undervaluation.

The opportunity cost of growth is why it is a bad idea to time the market based upon valuation.

Packer
Buy cheap and something good might happen

dh
Posts: 273
Joined: Sun Mar 13, 2011 8:01 pm

Re: Larry Swedroe: Disrupted Expectations

Post by dh » Mon Aug 06, 2018 12:02 pm

packer16 wrote:
Mon Aug 06, 2018 11:47 am
I think how much valuation matters is dependent upon your investment horizon and your distribution of your contributions.
I agree with Packer. When I hear friends say "valuations don't matter," I believe they are suggesting it doesn't matter today (don't go out an sell everything today, nor go 100% stocks today). However, if you look at a longer historical time period, it would be hard to argue that a PE10 of 7.4 in 1981 wasn't a good time to be in stocks for the next 10 to 15 years. It would be equally hard to argue that a PE10 of over 30 in the late 90s wasn't a "value stretched" period of time to invest where returns were likely to be lower. Obviously no one can predict the short run (and no one should day trade), but when you examine the data over a longer period of time (say 10 years) it is obvious that valuations do matter in terms of predicting future returns. I don't have the exact quote, but I recall Jack Bogle pulling back on his stocks for two reasons (he was getting older and the markets were getting expensive). If valuations matter to Jack (in terms of calculating a future return), then they matter to me (a much less wise individual). Best wishes to all Bogleheads!
:sharebeer

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Mon Aug 06, 2018 12:13 pm

I agree that valuations should have little if any effect on a young accumulator’s investment plan. I think a late accumulator close to retirement or in early retirement should think a lot about valuations. Lofty valuations don’t only imply a lower future mean expected return, they also imply that the whole future potential dispersion of returns shifts left: good outcomes less good and bad outcomes more bad. Bad sequence of returns in the 5 years on either side of retirement start date can be disastrous I think. This is also why I have focused so much on portfolio efficiency and diversifying across sources of return over the last few years. Narrowing the expected SD of a portfolio and lessening potential maximal drawdown while lowering expected return to a lesser extent can be very worthwhile.

Dave
Last edited by Random Walker on Mon Aug 06, 2018 12:21 pm, edited 1 time in total.

Park
Posts: 525
Joined: Sat Nov 06, 2010 4:56 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Park » Mon Aug 06, 2018 12:35 pm

packer16 wrote:
Mon Aug 06, 2018 11:47 am
I think how much valuation matters is dependent upon your investment horizon and your distribution of your contributions. The longer your investment horizon the less valuations matter as long as you are investing in market where as a shareholder you can reap the rewards of your investment like the US. If you compare the US to a growth stock, the growth rate is more important than the price over the long term. If the time horizon is less than 5 to 10 years, valuation can have a larger impact on terminal values than growth rates. As time goes on growth can overcome valuations.

As the timeframe over which you are investing increases, valuations become less important the continued growth. So lump sum investors should care about valuations if they need the money in less than 5 to 10 years. If you are DCAing, then the importance of growth of your initial contributions can offset valuations. Since you are investing over time, the cases of overvaluation are offset by the cases of undervaluation.

The opportunity cost of growth is why it is a bad idea to time the market based upon valuation.

Packer
The above is well worth reading.

A corollary to DCAing is how distributions are made from a portfolio. If 3-4% is withdrawn annually over a 25 year retirement, then growth is important. But valuations play a important role in the 10 year period surrounding the date you retire.

User avatar
packer16
Posts: 1012
Joined: Sat Jan 04, 2014 2:28 pm

Re: Larry Swedroe: Disrupted Expectations

Post by packer16 » Mon Aug 06, 2018 1:46 pm

Interesting point Park. It appears the plus or minus 5 years around retirement can be a danger window for sequence of returns risk. For the probabilistic folks out there, does it make sense to minimize the risk (i.e. smooth our returns by taking risk off the table) & thus return outside of the sequence of risk tent of plus 5 to 10 years around your retirement date. If withdraws are over a long period of time, I would think you could take advantage of spreading the withdraw risk over time like you spread out the investment risk out. This would imply a 10 year of expenses in a minimal risk portfolio (like the type you are describing Dave) & the residual in a more risky portfolio.

Packer
Buy cheap and something good might happen

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Mon Aug 06, 2018 2:14 pm

Packer,
Yes, you definitely get what I’m thinking. I tend to believe that if one can secure maybe 15-20 years living expenses in bonds, then maybe feel free to take risk with the remainder. This is the liability matching portfolio concept I learned from reading Bernstein. He I think believes the LMP should be 25-30 years living expenses.
One reason I think sequence of returns is such a big issue for that retirement red zone is that I think most retirees tend to spend more early in retirement. While still healthy enough, they tend to travel and check off boxes on the bucket list. Deeper into retirement I think costs can decrease significantly. Of course health and associated costs of health care can jack all that up in any given individual circumstance.
In my own personal case, I’m 55 and have benefittted from the almost 10 year equity bull run. Valuations are high, future expected returns more modest, whole potential dispersion of returns shifted left, so I thought it worthwhile to take risk off the table, secure a good portion of the LMP, and go from there. The fail safe will be my current asset allocation, but if I generate enough of a LMP bond floor in absolute dollar terms in the intermediate future, I can see increasing the overall equity allocation deeper in the retirement years.

Dave

User avatar
HomerJ
Posts: 11567
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: Disrupted Expectations

Post by HomerJ » Mon Aug 06, 2018 3:05 pm

Random Walker wrote:
Mon Aug 06, 2018 2:14 pm
In my own personal case, I’m 55 and have benefittted from the almost 10 year equity bull run. Valuations are high, future expected returns more modest, whole potential dispersion of returns shifted left, so I thought it worthwhile to take risk off the table, secure a good portion of the LMP, and go from there.
Valuations were high in 2011, and Swedroe predicted 4.5% real return going forward. Instead, we've gotten like 12% real returns from the U.S. Total Stock Market Index.

How come you didn't listen to him then, and put most of your money in Total International Stock Market Index? That one has returned 4% real or so over the past 7 years. You must have ignored Swedroe's writings in the past, since you say you've benefited from this current bull market.

The fact of the matter is if you had listened to Swedroe in 2011, you would have far less money today.

Now, his calculations on CAPE may have been correct. But even the best proponents of CAPE say it only accounts for 40% of returns. There are a lot of other variables, and they apparently showed up in the past 7 years.

Swedroe sounds very confident today when he makes his predictions (but with only 40% of the variables accounted for), but he also sounded very confident in 2011. Anyone following his advice made less money and is less financially secure today than someone who ignored him.

That's an important data point as well. Why are you so sure he's right now?
The J stands for Jay

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Mon Aug 06, 2018 4:56 pm

HomerJ wrote:
Mon Aug 06, 2018 3:05 pm
Random Walker wrote:
Mon Aug 06, 2018 2:14 pm
In my own personal case, I’m 55 and have benefittted from the almost 10 year equity bull run. Valuations are high, future expected returns more modest, whole potential dispersion of returns shifted left, so I thought it worthwhile to take risk off the table, secure a good portion of the LMP, and go from there.
Valuations were high in 2011, and Swedroe predicted 4.5% real return going forward. Instead, we've gotten like 12% real returns from the U.S. Total Stock Market Index.

How come you didn't listen to him then, and put most of your money in Total International Stock Market Index? That one has returned 4% real or so over the past 7 years. You must have ignored Swedroe's writings in the past, since you say you've benefited from this current bull market.

The fact of the matter is if you had listened to Swedroe in 2011, you would have far less money today.

Now, his calculations on CAPE may have been correct. But even the best proponents of CAPE say it only accounts for 40% of returns. There are a lot of other variables, and they apparently showed up in the past 7 years.

Swedroe sounds very confident today when he makes his predictions (but with only 40% of the variables accounted for), but he also sounded very confident in 2011. Anyone following his advice made less money and is less financially secure today than someone who ignored him.

That's an important data point as well. Why are you so sure he's right now?
Fair questions. I don’t believe in market timing. But, coincident with moves in the market, we age, pass through stages of life, needs versus wants become more clear. So Larry’s need, willingness, ability to tolerate risk change over time. I don’t believe in timing the market in an in and out sort of fashion. But I do believe in sort of customizing the glide path towards the retirement portfolio. In this sense, target date funds don’t make a lot of sense to me. Changing allocation by 1-2% a year when the equity market has an SD of 15-20% doesn’t seem productive. The reason my allocation has changed is that over the decade my financial circumstance has changed. Because of market performance I have more, and I have more to lose!
I’m not at all sure Swedroe is right. In fact I don’t think he’s even made any kind of specific prediction. If he did, he wouldn’t be sure he’s right: he always mentions cloudy crystal balls and means with wide dispersions.
So no, I’ve never ignored Larry’s writings. I think I’ve interpreted them pretty correctly and done my best to adapt them to my changing circumstances. Those changing circumstances include age, career prospects, financial goals, past returns, current valuations, future expected returns, and my ongoing investing education.

Another thought. As you know behavioral finance has shown that the pain of a loss is twice as much as the happiness derived from an equal sized gain. I believe (and I think it’s been demonstrated) that that 2:1 ratio increases a lot as one’s net worth increases. This serves as a behavioral reason in addition to the economic human capital reason to become more conservative as one ages.

Last thought. In the last post, I did comment on maybe getting more aggressive with the AA later in retirement after the LMP is solidly established. Once again, this would not be market timing per se as much as opportunistically adjusting the AA according to personal circumstance. I am a strong advocate of having a plan and sticking to it; that is always the default choice. But makes sense to periodically readdress the plan as the assumptions underlying the plan change.

Dave

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Mon Aug 06, 2018 5:15 pm

Homer,
Another thought :-) more money is certainly better than less. But I think mature investing involves defining specific realistic goals and sticking to an AA that maximizes likelihood of achieving those goals. Sticking to the more aggressive plan has a higher likelihood of accumulating more, but also has a higher likelihood of not meeting goals. Maximizing terminal portfolio value is only one potential goal. Other potential goals include not running out of money, and reaching financial goals with the least stress possible.
Warren Buffett’s quote about the LTCM guys is so simply wise “they risked money they had and needed to make money they didn’t have and didn’t need”. I think it’s impossible to consider this issue without looking at valuations, expected returns, dispersion of returns, needs, wants. I really like the idea of taking definitive action in response to past returns/valuations when one approaches retirement. I think opportunistically shaping one’s own glide path is pretty wise. I think market timing with both in and out moves is foolish. William Bernstein discusses this nicely at the end of his life cycle investing book in the Investing For Adults series. I’m a huge fan of Monte Carlo Simulation. It can help tremendously in making these AA decisions. And the results can be somewhat surprising.

Dave

User avatar
HomerJ
Posts: 11567
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: Disrupted Expectations

Post by HomerJ » Mon Aug 06, 2018 5:24 pm

I agree 100% with you (and Swedroe) about "need, willingness, and ability to take risk" being very important when determining an AA. Swedroe's earlier writings about that topic were very good, in my opinion.

I just use bonds and CDs to temper my risk, instead of tilting to sectors based on valuations.
The J stands for Jay

david1082b
Posts: 371
Joined: Fri Jun 09, 2017 12:35 pm

Re: Larry Swedroe: Disrupted Expectations

Post by david1082b » Mon Aug 06, 2018 7:44 pm

dh wrote:
Mon Aug 06, 2018 12:02 pm
packer16 wrote:
Mon Aug 06, 2018 11:47 am
I think how much valuation matters is dependent upon your investment horizon and your distribution of your contributions.
I agree with Packer. When I hear friends say "valuations don't matter," I believe they are suggesting it doesn't matter today (don't go out an sell everything today, nor go 100% stocks today). However, if you look at a longer historical time period, it would be hard to argue that a PE10 of 7.4 in 1981 wasn't a good time to be in stocks for the next 10 to 15 years. It would be equally hard to argue that a PE10 of over 30 in the late 90s wasn't a "value stretched" period of time to invest where returns were likely to be lower. Obviously no one can predict the short run (and no one should day trade), but when you examine the data over a longer period of time (say 10 years) it is obvious that valuations do matter in terms of predicting future returns. I don't have the exact quote, but I recall Jack Bogle pulling back on his stocks for two reasons (he was getting older and the markets were getting expensive). If valuations matter to Jack (in terms of calculating a future return), then they matter to me (a much less wise individual). Best wishes to all Bogleheads!
:sharebeer
Jack Bogle apparently waited until around the year 2000 to reduced the stock % a lot, I think that is what gets said? By that time CAPE was over 40, not the 30 that was seen in the summer of 1997. If Jack had been in a Target Retirement fund he wouldn't have reduced his stock % in such a knee-jerk fashion at the age of 70 or so, it would have been on a gradual glide-path. Jack Bogle gets very consistent praise for reducing the stock % in the year 2000 but he just got lucky in my opinion since 1997 could have easily been the top, CAPE in 1997-98 was the same as the peak of 1929. Glide-paths exist because market timing is based more on luck than skill. Valuations only seemed to matter to Jack when they were around 50% higher than the previous record peak. How is this supposed to be seen as something to learn from? It seems so bizarre to praise someone for market timing instead of following the basic ideas of the Target Retirment funds that his own firm had started to put out in the mid-90s.

It's interesting to look at the CAPE valuation for June 1922, which was 7.55. 10 years later was the approximate low-point of the great depression, and stock returns had been very low for a theoretical S&P 500, 1.8% annualized adjusted for inflation and actually negative without adjusting for "inflation", so theoretical S&P 500 returns were only positive because of some deflation. In the 1990s CAPE first went over 30 in June 1997 and the next ten years saw 4.5% annualized adjusted for inflation https://dqydj.com/sp-500-return-calculator/ http://www.multpl.com/shiller-pe/table?f=m

stlutz
Posts: 4660
Joined: Fri Jan 02, 2009 1:08 am

Re: Larry Swedroe: Disrupted Expectations

Post by stlutz » Mon Aug 06, 2018 7:49 pm

Valuations matter a lot
Dave: What does "a lot" mean? That phrase gets bandied around here a lot but what percent of returns over the next 10-20 years will be determined by today's valuations, do you think? 10%? 80%? Getting to a more precise term than "a lot" might help move the discussion forward as opposed to just giving everyone a chance to repeat their usual talking points.

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Mon Aug 06, 2018 9:47 pm

stlutz wrote:
Mon Aug 06, 2018 7:49 pm
Valuations matter a lot
Dave: What does "a lot" mean? That phrase gets bandied around here a lot but what percent of returns over the next 10-20 years will be determined by today's valuations, do you think? 10%? 80%? Getting to a more precise term than "a lot" might help move the discussion forward as opposed to just giving everyone a chance to repeat their usual talking points.
Think I’ve read that the explanatory power of CAPE is about 40%. That’s really at the limits of my understanding. With the combination of my limited knowledge and unknowable future, I wouldn’t venture beyond my usual talking points until I know more. What I do think though is that likely too many people focus on a mean expected return rather than an appreciation for the whole potential future dispersion of returns shifting left. With less good good outcomes and more bad bad outcomes in that distribution, and an appreciation of a loss hurting at least twice as much as an equal size gain, I think people can make some pretty good AA decisions in light of their own personal circumstance and some back of the napkin rough numbers. I wouldn’t be talking about this stuff at ages 20-45. I’m coming from the perspective of age 55 and working as a salaried employee: my human capital could be at risk.

My back of the napkin numbers would use Larry’s roughly 5% nominal for US equities with an SD of 20%. Estimates based on CAPE assume no change in valuations, Bogle’s speculative component. If the market decides to view the world as more risky and valuations change (P/E ratios contract) things can get ugly quickly. What’s more likely, for P/E ratios to further increase or decrease? To increase a lot or decrease a lot?

Dave

stlutz
Posts: 4660
Joined: Fri Jan 02, 2009 1:08 am

Re: Larry Swedroe: Disrupted Expectations

Post by stlutz » Mon Aug 06, 2018 10:32 pm

Think I’ve read that the explanatory power of CAPE is about 40%. That’s really at the limits of my understanding. With the combination of my limited knowledge and unknowable future, I wouldn’t venture beyond my usual talking points until I know more.
I think we can venture further than that.

A common way to break down to the components of future returns is into these three categories:

a) The current yield
b) Future EPS growth
c) Future changes in how investors value stocks.

Obviously the factor we're considering here is determinative only of (a). Stocks yielding 1% will return 2% less than stocks yielding 3%--all other things being equal. Of course all other things never are equal.

Let's consider EPS growth. 0% real EPS growth over 20 years would likely produce vastly different returns than, say, 4% real EPS growth.

Then there are future changes in how investors value stocks. There are all sorts of factors that can go into this--expectations of future earnings, interest rates, inflation rates etc. etc.

So, let's consider a couple of scenarios. Historical real earnings growth over the past century has been a little under 2% (just to use round numbers). Suppose it was 1% over the next 20 years. That's a pretty big hit. Furthermore, if that was the case, investors would probably re-value stocks in the downward direction over that time, compounding the issue. In the scenario, would the starting yield really determine 40% of returns?

Or to take a more static scenario. Suppose future earnings growth is also a little under 2%. And that there are no valuation changes. Would the current yield really represent 40% of your returns? It would at 1982 type valuations. Not today's.

As I see it, future earnings growth is what matters "a lot". Not just on their own but in terms of how they will lead future investors to value equities. Current valuations matter more in the "somewhat" category.

The problem is that forecasting future earnings growth is hard. It's not impossible--I would be willing to guarantee that EPS growth will not average 10% per year for the next 20 years after all. But the problem with CAPE determinism is that it's based on the assumption that the future growth rate = the historical growth rate. That's not unreasonable, but it's still simply projecting a past trend forward into the future which we're always told is bad.

Poster Simple Gift has had a number of threads taking a more thoughtful approach to projecting future growth. They promoted some good discussion. So, the issue of what future growth will be can be considered in a more in depth fashion than it usually is.

Projecting how investors will want to allocate between stocks and other assets in the future is even harder. To some extent a contrarian approach can be helpful--if people hold a low percentage of portfolios in stocks now, there are more scenarios where that will be higher in the future than not. So, low equity allocations currently probably forecasts higher future returns, and vice versa.

Oftentimes CAPE is used as the metric for evaluating investors current preference for equities. I'm not sure why one would do that when this can be quantified directly as the Philosophical Economics guy did in this famous blog post: https://www.philosophicaleconomics.com/ ... t-returns/.

To me, CAPE seems to carry a lot of weight because smoothing earnings seems to make sense and because it's a easy statistic to calculate (or look up on the internet). Forecasting earnings growth or future investor preferences--not so easy. However, "easy to look up" and "mattering a lot" aren't the same thing.

In short, I would say that valuations matter and should be considered by investors, but I would consider them determinative of something more like 20% of future returns than 40%.

Park
Posts: 525
Joined: Sat Nov 06, 2010 4:56 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Park » Tue Aug 07, 2018 7:50 am

http://www.crossingwallstreet.com/archi ... arket.html

"the blog’s indicator is based on the percentage of household financial assets—stocks, bonds and cash—that is allocated to stocks...superior to seven other well-known valuation indicators

This metric has an R-square of 0.61. Here are the seven others:

• The Q ratio, with an R-squared of 46%...

• The price/sales ratio, with an R-squared of 44%...

• with an R-squared of 39%. This indicator, which is the ratio of the total value of equities in the U.S. to gross domestic product...

• CAPE, the cyclically adjusted price/earnings ratio, came next in the ranking, with an R-squared of 35%...

• Dividend yield...sports an R-squared of 26%.

• Traditional price/earnings ratio has an R-squared of 24%.

• Price/book ratio...has an R-squared of 21%.

According to various tests of statistical significance, each of these indicators’ track records is significant at the 95% confidence level...

the differences between the R-squareds of the top four or five indicators I studied probably aren’t statistically significant, I was told by Prof. Shiller. That means you’re overreaching if you argue that you should pay more attention to, say, the average household equity allocation than the price/sales ratio."

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Tue Aug 07, 2018 9:14 am

Stlutz,
Thanks for that explanation. Like I said, limits of my knowledge. But seems to me dividend yield is pretty consistently solid roughly 2% and earnings growth usually in the range of 2-3%.

Dividend yield + earnings growth + valuation changes= return

Seems to me changes in the P/E ratio would dominate returns over shorter periods of time. If the P/E changed from say 20 to 15, that would be a 25% decrease. That dominates the relatively small numbers and changes for dividend yield and earnings growth. Equities have historically had a SD of about 15-20%. Seems to me most of that SD is due to changes in P/E ratios.

Dave

User avatar
corn18
Posts: 914
Joined: Fri May 22, 2015 6:24 am

Re: Larry Swedroe: Disrupted Expectations

Post by corn18 » Tue Aug 07, 2018 10:06 am

For investors, valuation matters. For traders/speculators, it does not. Somehow, the CAPE always finds its way home.

Image

I like this chart. The bumpy bits are traders/speculators. The line is the investors.

Image

User avatar
packer16
Posts: 1012
Joined: Sat Jan 04, 2014 2:28 pm

Re: Larry Swedroe: Disrupted Expectations

Post by packer16 » Tue Aug 07, 2018 2:07 pm

Random Walker wrote:
Tue Aug 07, 2018 9:14 am
Stlutz,
Thanks for that explanation. Like I said, limits of my knowledge. But seems to me dividend yield is pretty consistently solid roughly 2% and earnings growth usually in the range of 2-3%.

Dividend yield + earnings growth + valuation changes= return

Seems to me changes in the P/E ratio would dominate returns over shorter periods of time. If the P/E changed from say 20 to 15, that would be a 25% decrease. That dominates the relatively small numbers and changes for dividend yield and earnings growth. Equities have historically had a SD of about 15-20%. Seems to me most of that SD is due to changes in P/E ratios.

Dave
The one thing we are seeing with the 2% real growth rates are buy-backs on the order of 2% of market cap per year. So if that is added in you get 6-7% a number close to Seigel's constant over time.

Packer
Buy cheap and something good might happen

larryswedroe
Posts: 15693
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Larry Swedroe: Disrupted Expectations

Post by larryswedroe » Tue Aug 07, 2018 4:14 pm

to be helpful here

Real g of earnings should grow roughly in line with real gnp. That's 2-3% and pretty stable over long periods. So that's reasonable assumption IMO.

The remainder is basically the change in speculative valuation, which is of course not predictable.

The fact is that historically the CAPE 10 has explained about 40% of the returns over long horizons, like 10 years. And clearly it is very important information because not only is 40% a significant figure, but as the CAPE rises the entire distribution of outcomes shifts to the left, with not only the mean falling but the worst cases get worse and the best cases get less good. There's not a single financial economist of any note that I've met who doesn't think valuations matter a great deal. Note it works at shorter horizons (like 6, or 8) it's just that the shorter the horizon the wider the dispersions become, but there is still valuable information.

The problem is that some, like Grantham and Hussman try to use valuations to time market, and IMO (with possible exceptions of very extreme valuations where TIPS yield more than CAPE 10, and even then maybe not because have to know when to get back in) should not be used to market time, only to help you determine your NEED to take risk. What return you need to achieve your financial goal. Now if it causes you to take more risk than ability or willingness then you should likely not do that, instead adjust your plan to save more or lower goals or plan on working longer, or whatever the Plan B calls for.

I hope that is helpful

If have questions feel free to email or PM
Best wishes
Larry

Random Walker
Posts: 2980
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Disrupted Expectations

Post by Random Walker » Tue Aug 07, 2018 5:49 pm

With bond yields at about 2% and expected nominal return on US equity at about 5%, seems that the question is how much risk is an investor willing to take for the additional 3%. For a young investor with not much to lose, a lot of human capital in reserve, and the benefits of compounding to look forward to, that 3% is huge. For someone in sight of retirement though, knowing that equity SD is large and the whole dispersion has shifted left, stretching for that 3% could well be like picking up nickels in front of a steam roller.

Dave

User avatar
whodidntante
Posts: 3917
Joined: Thu Jan 21, 2016 11:11 pm

Re: Larry Swedroe: Disrupted Expectations

Post by whodidntante » Tue Aug 07, 2018 5:54 pm

HomerJ wrote:
Mon Aug 06, 2018 11:05 am
I disagree. Valuations have proven to matter very little. We have seen multiple forecasts based on valuations in the past that turned out to be incorrect. They are not a good prediction tool. It is certainly not necessary to account for them.
It's all fun and games until your portfolio gets cut in half.

User avatar
HomerJ
Posts: 11567
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: Disrupted Expectations

Post by HomerJ » Tue Aug 07, 2018 11:05 pm

whodidntante wrote:
Tue Aug 07, 2018 5:54 pm
HomerJ wrote:
Mon Aug 06, 2018 11:05 am
I disagree. Valuations have proven to matter very little. We have seen multiple forecasts based on valuations in the past that turned out to be incorrect. They are not a good prediction tool. It is certainly not necessary to account for them.
It's all fun and games until your portfolio gets cut in half.
Buy and hold. Don't forget the hold part.

Valuations don't matter. Years to retirement matter. Since I'm 5 years away, I'm 50/50 stocks/bonds. It would be difficult for my portfolio to be cut in half at this point (although still possible). Note I would be 50/50 even if valuations were "normal" or "low". Because the risk of a crash is never zero.

Therefore, I am speaking correctly when I say "valuations do not matter". Because I would be 50/50 at this point no matter what valuations were at.
The J stands for Jay

stlutz
Posts: 4660
Joined: Fri Jan 02, 2009 1:08 am

Re: Larry Swedroe: Disrupted Expectations

Post by stlutz » Tue Aug 07, 2018 11:27 pm

In my post I drew a subtle distinction that I should be more explicit about.

A while back we had a thread about the relationship between and inverted yield curve and a recession and I asked the question if inversion was a cause of a recession or if had simply correlated to it historically. I actually did find someone who argued that it was in fact a cause, which was interesting.

That is what I was interested in exploring here--is a "high" CAPE a cause of lower returns or has it simply correlated to that historically. An R-squared of ~.4 between CAPE and future returns doesn't actually "explain" anything--it's just an interesting pattern that warrants further investigation

I suggested that current valuation is probably about 20% "determinative" of future returns. Which puts me in agreement with precisely nobody in this thread, as I am suggesting that valuation is in fact relevant, but it matters a lot less than future earnings growth.

One other note: Changes in valuation actually have played a significant role in past returns. There is no reason not to expect that in the future. Historically, the trend has been toward higher valuations. That trend could continue, it could return to some unknown mean, or it might stay about where is is. This is hard to forecast. I have proposed that investor expectations of future growth play a big role in how they value stocks, which reinforces my view that future growth is the dominant factor in future returns.

In short, if you want a good estimate of future returns, you need a good growth estimate first and then look at current valuations second, in my view. Unfortunately all of the attention goes to the later and not to the former.

User avatar
FIREchief
Posts: 2626
Joined: Fri Aug 19, 2016 6:40 pm

Re: Larry Swedroe: Disrupted Expectations

Post by FIREchief » Tue Aug 07, 2018 11:39 pm

"Research on the expected equity premium, including Aswath Damodaran’s paper, “Equity Risk Premiums (ERP): Determinants, Estimation and Implications,” has found that the best predictor of future equity returns is current valuations—whether using measures such as the earnings yield (E/P) derived from the Shiller CAPE 10 (or, for that matter, the CAPE 7, 8 or 9) or the current E/P—not historical returns."

Well, which one is it? There is a significant difference between E/P derived from CAPE 10 and current E/P. The first is 3.0%, the second 4.4% (going up to 5.1% by the end of this year for believers in Standard and Poors' forecasts). If you really want to stretch it, the S&P forecast for the end of next year is 5.8%. I'm with HomerJ on this stuff.... :sharebeer
I am not a lawyer, accountant or financial advisor. Any advice or suggestions that I may provide shall be considered for entertainment purposes only.

stlutz
Posts: 4660
Joined: Fri Jan 02, 2009 1:08 am

Re: Larry Swedroe: Disrupted Expectations

Post by stlutz » Tue Aug 07, 2018 11:48 pm

...including Aswath Damodaran...
Interestingly, the ERP that he has calculated has consistently been more optimistic than the pessimists in the CAPE threads on Bogleheads.

He currently estimates the ERP above the long-term Treasury rate as 4.78%. So, his projected future nominal return is 4.78 + 3 (the 10 year T-note yield) or 7.78%. If I subtract 2% for inflation, I get about 5.75% real. Which is not bad!

AlphaLess
Posts: 556
Joined: Fri Sep 29, 2017 11:38 pm

Re: Larry Swedroe: Disrupted Expectations

Post by AlphaLess » Tue Aug 07, 2018 11:56 pm

HomerJ wrote:
Mon Aug 06, 2018 3:05 pm
There are a lot of other variables, and they apparently showed up in the past 7 years.
LOVE, absolute LOVE that comment!

I am going to frame that one:
"THER ARE A LOT OF OTHER VARIABLES, AND THEY APPARENTLY SHOWED UP IN THE PAST 7 YEARS!"

User avatar
randomizer
Posts: 1443
Joined: Sun Jul 06, 2014 3:46 pm

Re: Larry Swedroe: Disrupted Expectations

Post by randomizer » Sat Sep 15, 2018 5:49 pm

Great article. Larry may or may not be right but he is consistently logical, well researched, and open to new data. The core message that you shouldn't assume that past performance predicts future performance is correct. The implication that optimistic planning is dangerous is worth repeating. I am going to stay the course and keep saving.
87.5:12.5, EM tilt — HODL the course!

james22
Posts: 1204
Joined: Tue Aug 21, 2007 2:22 pm

Re: Larry Swedroe: Disrupted Expectations

Post by james22 » Sat Sep 15, 2018 9:44 pm

HomerJ wrote:
Tue Aug 07, 2018 11:05 pm
Valuations don't matter. Years to retirement matter. Since I'm 5 years away, I'm 50/50 stocks/bonds. It would be difficult for my portfolio to be cut in half at this point (although still possible). Note I would be 50/50 even if valuations were "normal" or "low". Because the risk of a crash is never zero.

Therefore, I am speaking correctly when I say "valuations do not matter". Because I would be 50/50 at this point no matter what valuations were at.
Valuations don't matter to you, Homer.

How many of us are you?
This whole episode is likely to end so badly that future children will learn about it in school and shake their heads in wonder at the rank stupidity of it all... Hussman

User avatar
HomerJ
Posts: 11567
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: Disrupted Expectations

Post by HomerJ » Sat Sep 15, 2018 10:45 pm

james22 wrote:
Sat Sep 15, 2018 9:44 pm
HomerJ wrote:
Tue Aug 07, 2018 11:05 pm
Valuations don't matter. Years to retirement matter. Since I'm 5 years away, I'm 50/50 stocks/bonds. It would be difficult for my portfolio to be cut in half at this point (although still possible). Note I would be 50/50 even if valuations were "normal" or "low". Because the risk of a crash is never zero.

Therefore, I am speaking correctly when I say "valuations do not matter". Because I would be 50/50 at this point no matter what valuations were at.
Valuations don't matter to you, Homer.

How many of us are you?
We can play that game if you want. As long you guys stick with "Valuations matter to me", and not "Valuations matter, and you're completely irrational if you ignore them" (A Swedroe quote), then I won't dispute you.

My opinion is when you are young, valuations matter very little. Expected returns matter very little. You save a good amount, and you are heavily invested in stocks, regardless of valuations because you're young and you have plenty of time ahead of you. You don't need to change your AA every couple of years, lowering or raising your stock allocation in your 30s based on what the experts say. Buy and hold, and stay the course. This is not an irrational plan.

In my opinion, when you're older, the fact you are close to retirement is far more important than valuations. At that point, you need to play defense, regardless of valuations. I'm already planning for low returns, because I feel it would be foolish to count on high returns to meet my goals, regardless of valuations. If valuations were low, I wouldn't plan around high returns... I feel it's wiser to plan for low returns in all situations, regardless of valuations. This way I'm covered if low returns show up, and pleasantly surprised if returns are higher.

But that's just my opinion. I could certainly be wrong, and you could be right.

My way isn't the only way. Your way may indeed be better. I'd be happy to hear how you adjust your investments based on various valuations levels. Just please don't tell me I'm irrational to save a lot and plan around low returns and ignore valuations.
The J stands for Jay

User avatar
willthrill81
Posts: 5361
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Disrupted Expectations

Post by willthrill81 » Sat Sep 15, 2018 10:50 pm

The statement that "valuations have explained 40% of ten year stock returns" is somewhat misleading. I can point to other (completely nonsensical) variables that have had a far stronger relationship with stock returns. That 40% estimate primarily relies on data leading up to the late 1980s when Shiller proposed CAPE as an explanatory tool for stock returns. Since Shiller proposed CAPE, it's explanatory power has been significantly less than 40% (I don't recall exactly how much), implying that it may not be as useful of a predictor as originally thought. It might even be that CAPE is little more than a 'data mined' artifact. It does have some intuitiveness going for it, but so do countless other variables that are not so often used in this fashion.

I agree with Homer in that (1) investors should have conservative expectations for stock returns regardless of any other factors in play and, consequently, (2) valuations can be ignored with no deleterious effect on the investor.
“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
willthrill81
Posts: 5361
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Disrupted Expectations

Post by willthrill81 » Sat Sep 15, 2018 10:57 pm

HomerJ wrote:
Sat Sep 15, 2018 10:45 pm
If valuations were low, I wouldn't plan around high returns... I feel it's wiser to plan for low returns in all situations, regardless of valuations. This way I'm covered if low returns show up, and pleasantly surprised if returns are higher.
:thumbsup

There are many markets that have had relatively low valuations for years that still experienced low returns for a long time. And the U.S. has had relatively high valuations for nearly 27 years and very good returns.

Even if there is something to valuations, there's also clearly a lot going on that has nothing to do with valuations.
“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

randomguy
Posts: 6171
Joined: Wed Sep 17, 2014 9:00 am

Re: Larry Swedroe: Disrupted Expectations

Post by randomguy » Sat Sep 15, 2018 11:03 pm

packer16 wrote:
Mon Aug 06, 2018 1:46 pm
Interesting point Park. It appears the plus or minus 5 years around retirement can be a danger window for sequence of returns risk. For the probabilistic folks out there, does it make sense to minimize the risk (i.e. smooth our returns by taking risk off the table) & thus return outside of the sequence of risk tent of plus 5 to 10 years around your retirement date. If withdraws are over a long period of time, I would think you could take advantage of spreading the withdraw risk over time like you spread out the investment risk out. This would imply a 10 year of expenses in a minimal risk portfolio (like the type you are describing Dave) & the residual in a more risky portfolio.

Packer
It doesn't help much from what I have read. If in 1966, you took you 25x and split it into 10 years in bonds a bond ten and 15 years in a portfolio (say 50/50), you ended up in about the same spot as just holding a 50/50 portfolio. You avoided the volatility of 73-74 but volatility wasn't the portfolio killer. It was the 0% real returns from both bonds and stocks for 15 years. There are tons of papers talking about bond tents, rising equity glide paths, buckets and so on. Some of the improve the sleep well at night test but I am unaware of any that post higher SWR by more than about 5%.

Now there are other cases where this scheme works better. I imagine the great depression worked best when you held as much bonds as possible in 1929-1932:) And 2000-9 was another period where holding more bonds paid off. There are some crazy inflation periods of the late 80s and the WWI years where I am guessing holding bonds got you slaughtered but I am not sure if stocks helped. I tend not place a lot of faith in data back that far.

james22
Posts: 1204
Joined: Tue Aug 21, 2007 2:22 pm

Re: Larry Swedroe: Disrupted Expectations

Post by james22 » Sun Sep 16, 2018 12:52 am

HomerJ wrote:
Sat Sep 15, 2018 10:45 pm
Just please don't tell me I'm irrational to save a lot and plan around low returns and ignore valuations.
Not irrational at all (as a satisficer) to invest as you do, Homer.

But just maybe a little irrational to continually argue on the Theory board against ways that might better returns for those maximizers who invest in other ways?
This whole episode is likely to end so badly that future children will learn about it in school and shake their heads in wonder at the rank stupidity of it all... Hussman

randomguy
Posts: 6171
Joined: Wed Sep 17, 2014 9:00 am

Re: Larry Swedroe: Disrupted Expectations

Post by randomguy » Sun Sep 16, 2018 1:29 am

The question I have is who believes in these estimates enough to increase their holdings of developed international and EM given they are getting almost 50% higher returns? As much as I like them, I sure am not increasing past the 30% I like to hang around.

typical.investor
Posts: 141
Joined: Mon Jun 11, 2018 3:17 am

Re: Larry Swedroe: Disrupted Expectations

Post by typical.investor » Sun Sep 16, 2018 2:18 am

HomerJ wrote:
Sat Sep 15, 2018 10:45 pm
james22 wrote:
Sat Sep 15, 2018 9:44 pm
HomerJ wrote:
Tue Aug 07, 2018 11:05 pm
Valuations don't matter. Years to retirement matter. Since I'm 5 years away, I'm 50/50 stocks/bonds. It would be difficult for my portfolio to be cut in half at this point (although still possible). Note I would be 50/50 even if valuations were "normal" or "low". Because the risk of a crash is never zero.

Therefore, I am speaking correctly when I say "valuations do not matter". Because I would be 50/50 at this point no matter what valuations were at.
Valuations don't matter to you, Homer.

How many of us are you?
We can play that game if you want. As long you guys stick with "Valuations matter to me", and not "Valuations matter, and you're completely irrational if you ignore them" (A Swedroe quote), then I won't dispute you.

My opinion is when you are young, valuations matter very little. Expected returns matter very little. You save a good amount, and you are heavily invested in stocks, regardless of valuations because you're young and you have plenty of time ahead of you. You don't need to change your AA every couple of years, lowering or raising your stock allocation in your 30s based on what the experts say. Buy and hold, and stay the course. This is not an irrational plan.
No, it's not an irrational plan. Investors with less wealth, more debt (more leverage) and are more sensitive to human capital risk tend to favor growth stocks. Investors who are not those things are better in a position to take on the higher risk of value stocks and earn their premium.
HomerJ wrote:
Sat Sep 15, 2018 10:45 pm
In my opinion, when you're older, the fact you are close to retirement is far more important than valuations. At that point, you need to play defense, regardless of valuations. I'm already planning for low returns, because I feel it would be foolish to count on high returns to meet my goals, regardless of valuations. If valuations were low, I wouldn't plan around high returns... I feel it's wiser to plan for low returns in all situations, regardless of valuations. This way I'm covered if low returns show up, and pleasantly surprised if returns are higher.
Opinion aside, there is research showing people do move to value as their ability to withstand risk climbs:
This paper documents strong empirical patterns in the holdings of value and growth stocks by
households. The average value investor is substantially older, and has higher financial wealth,
higher real estate wealth, lower leverage, lower income risk, and lower human capital than the
average growth investor.
Our paper is the first to document portfolio evidence in favor
of rational theories of the value premium.
We provide evidence that households actively manage their holdings of growth and value
stocks. At yearly frequencies, households dynamically rebalance their exposure to the value factor
in response to passive variation in the portfolio tilt.
https://www.econstor.eu/bitstream/10419 ... 578911.pdf

So I conclude that how much someone pays attention to valuations is in part a rational calculation about risk. Just because valuations are high, it doesn't necessarily mean it's the best strategy for a young investor who has their employment at risk in a recession to pile into value stocks just because expected returns are higher. Having to sell at a low obviously would put the ability to ultimately benefit from any value premium at risk.

HEDGEFUNDIE
Posts: 631
Joined: Sun Oct 22, 2017 2:06 pm

Re: Larry Swedroe: Disrupted Expectations

Post by HEDGEFUNDIE » Sun Sep 16, 2018 2:51 am

typical.investor wrote:
Sun Sep 16, 2018 2:18 am
HomerJ wrote:
Sat Sep 15, 2018 10:45 pm
james22 wrote:
Sat Sep 15, 2018 9:44 pm
HomerJ wrote:
Tue Aug 07, 2018 11:05 pm
Valuations don't matter. Years to retirement matter. Since I'm 5 years away, I'm 50/50 stocks/bonds. It would be difficult for my portfolio to be cut in half at this point (although still possible). Note I would be 50/50 even if valuations were "normal" or "low". Because the risk of a crash is never zero.

Therefore, I am speaking correctly when I say "valuations do not matter". Because I would be 50/50 at this point no matter what valuations were at.
Valuations don't matter to you, Homer.

How many of us are you?
We can play that game if you want. As long you guys stick with "Valuations matter to me", and not "Valuations matter, and you're completely irrational if you ignore them" (A Swedroe quote), then I won't dispute you.

My opinion is when you are young, valuations matter very little. Expected returns matter very little. You save a good amount, and you are heavily invested in stocks, regardless of valuations because you're young and you have plenty of time ahead of you. You don't need to change your AA every couple of years, lowering or raising your stock allocation in your 30s based on what the experts say. Buy and hold, and stay the course. This is not an irrational plan.
No, it's not an irrational plan. Investors with less wealth, more debt (more leverage) and are more sensitive to human capital risk tend to favor growth stocks. Investors who are not those things are better in a position to take on the higher risk of value stocks and earn their premium.
HomerJ wrote:
Sat Sep 15, 2018 10:45 pm
In my opinion, when you're older, the fact you are close to retirement is far more important than valuations. At that point, you need to play defense, regardless of valuations. I'm already planning for low returns, because I feel it would be foolish to count on high returns to meet my goals, regardless of valuations. If valuations were low, I wouldn't plan around high returns... I feel it's wiser to plan for low returns in all situations, regardless of valuations. This way I'm covered if low returns show up, and pleasantly surprised if returns are higher.
Opinion aside, there is research showing people do move to value as their ability to withstand risk climbs:
This paper documents strong empirical patterns in the holdings of value and growth stocks by
households. The average value investor is substantially older, and has higher financial wealth,
higher real estate wealth, lower leverage, lower income risk, and lower human capital than the
average growth investor.
Our paper is the first to document portfolio evidence in favor
of rational theories of the value premium.
We provide evidence that households actively manage their holdings of growth and value
stocks. At yearly frequencies, households dynamically rebalance their exposure to the value factor
in response to passive variation in the portfolio tilt.
https://www.econstor.eu/bitstream/10419 ... 578911.pdf
Could this be due to a preference by older people for stable dividend-paying income stocks, rather than an explicit strategy to capture the value premium?

typical.investor
Posts: 141
Joined: Mon Jun 11, 2018 3:17 am

Re: Larry Swedroe: Disrupted Expectations

Post by typical.investor » Sun Sep 16, 2018 3:01 am

HEDGEFUNDIE wrote:
Sun Sep 16, 2018 2:51 am
Could this be due to a preference by older people for stable dividend-paying income stocks, rather than an explicit strategy to capture the value premium?
Based on the data in that research, and how value exposure rises over time, I would conclude it's not a sudden desire for dividend paying income stocks in retirement.

User avatar
nedsaid
Posts: 10216
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe: Disrupted Expectations

Post by nedsaid » Sun Sep 16, 2018 6:02 am

packer16 wrote:
Mon Aug 06, 2018 11:47 am
I think how much valuation matters is dependent upon your investment horizon and your distribution of your contributions. The longer your investment horizon the less valuations matter as long as you are investing in market where as a shareholder you can reap the rewards of your investment like the US. If you compare the US to a growth stock, the growth rate is more important than the price over the long term. If the time horizon is less than 5 to 10 years, valuation can have a larger impact on terminal values than growth rates. As time goes on growth can overcome valuations.

As the timeframe over which you are investing increases, valuations become less important the continued growth. So lump sum investors should care about valuations if they need the money in less than 5 to 10 years. If you are DCAing, then the importance of growth of your initial contributions can offset valuations. Since you are investing over time, the cases of overvaluation are offset by the cases of undervaluation.

The opportunity cost of growth is why it is a bad idea to time the market based upon valuation.

Packer
I think of the Nifty Fifty stocks back in the 1960's. These were Blue Chip stocks from that era, the idea was that you would buy them and put them away. The so-called one decision stocks, all you had to do was decide to buy them. Problem was that Price/Earnings ratios on some of these stocks got to be over 50. That is pretty extreme for even the bluest of blue chips, that is the best companies out there to invest in. Well, what happened? The US Stock Market was essentially flat from 1968 through about 1984 with the 50% down 1973-74 bear market in between.
So valuations do matter. Your returns from stocks during those years were from dividends.

Those with long time horizons who held those Nifty Fifty stocks did well if they held on. They just had to be very, very, very patient. Eventually the growth in earnings overcame the high valuations.
Problem is, you had to wait twenty years or more.

We saw this again in the late 1990's when forward market P/E's got to about 32 and P/E's based upon the last year's earnings got to be about 45. There were stocks priced to infinity. Despite really good earnings growth during the 2000's, the market was essentially flat from early 2000 through 2012. Again, whatever returns you got were from dividends. Instead of one 50% bear market experienced during the 1968-1984 period, the 2000-2012 period saw two 50% down markets.

Of course, we all think of the roaring twenties with a booming economy and a speculative stock market. There was the awful crash in 1929 and stocks there essentially flat from about 1929 through about 1948 or perhaps even later. Again, those dividends were really important and helped cushion the losses. Again, if you were really, really, really patient; you would have been rewarded during the 1950's and 1960's.

I have given three examples of where stock returns were very muted for many years after speculative excesses. Not talking to Packer here, but it just astonishes me when certain folks here say that valuations don't matter. They certainly do. The weight of historical evidence is just overwhelming. Higher valuations lead to lower future returns, particularly when you see market manias.

We all know that markets can rise during times of high valuations and that markets can fall during times of low expectations. But as Larry has pointed out, valuations affect the distribution of returns. If valuations are high, you can still have good years in the stock market but the distribution of returns will shift, you will tend to get more negative surprises from the stock market. If valuations are low, you can still have bad years in the stock market but the distribution of returns will shift, you will tend to get more positive surprises in the stock market.

Pretty much, P/E ratios are a measure of investor enthusiasm for stocks. In many cases, the investor enthusiasm is warranted. However, if expectations get to be too high, it will take many years for the earnings growth to overcome high P/E ratios. Indeed, P/E ratios can contract even as earnings grow as we saw during the 2000's.

So Packer is right here, given enough time, growth will win out no matter how high P/E ratios are. Problem is, you might have to wait a long time. He is also right that valuations are more important to those with shorter time horizons. An investor is his or her twenties need not give all this a single thought, just get started and keep investing. Such an investor should welcome bear markets as this gives opportunity to buy in at low prices. A 50% down bear market could be just devastating to a near retiree or recent retiree if their portfolio is still stock heavy.
A fool and his money are good for business.

User avatar
nisiprius
Advisory Board
Posts: 36466
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: Larry Swedroe: Disrupted Expectations

Post by nisiprius » Sun Sep 16, 2018 7:25 am

So many things in investing remind me of the remarks in a paper by Cowles and Jones, 1937, "Some A Posteriori Probabilities in Stock Market Action." This paper showed the phenomena now called "momentum" and "mean reversion" (under the names "inertia" and "reversals"). It gave convincing evidence for their reality. The authors then noted that "This evidence of structure in stock prices suggests alluring possibilities in the way of forecasting." But, after several pages trying to work out a market timing formula based on it, they concluded that "This type of forecasting could not be employed by speculators with any assurance of consistent or large profits."

I believe it is possible to discover valid statistical patterns in past data, and yet (the part people find really hard to believe) not be able to "employ" them for forecasting "with any assurance of consistent or large profits."

Since we should be planning prudently, we should be assuming low-end performance by the stock market anyway. This is, for example, is what Fidelity's wildly over-detailed "Fidelity Retirement Income Planner" did (when a rep walked me through it years ago). The default projections are all based on 10th-percentile (low end) stock market performance. We can't avoid being curious to know whether or not we will really experience 10th-percentile performance--or average performance, or above-average performance--but such predictions shouldn't enter into our planning.
A. E. Housman wrote:Therefore, since the world has still
Much good, but much less good than ill,
And while the sun and moon endure
Luck's a chance, but trouble's sure,
I'd face it as a wise man would,
And train for ill and not for good.
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.

grok87
Posts: 8298
Joined: Tue Feb 27, 2007 9:00 pm

Re: Larry Swedroe: Disrupted Expectations

Post by grok87 » Sun Sep 16, 2018 7:46 am

nisiprius wrote:
Sun Sep 16, 2018 7:25 am


I believe it is possible to discover valid statistical patterns in past data, and yet (the part people find really hard to believe) not be able to "employ" them for forecasting "with any assurance of consistent or large profits."

yep

article cites average momentum premium from 2010 to present at around 4% based on fama french data.

here is the data from french website
2010 6.01
2011 7.26
2012 1.57
2013 7.9
2014 1.6
2015 20.66
2016 -20.34
2017 5.04

arithmetic average is 3.71% which would tie to the "around 4%" but i prefer CAGR (geometric average) which is 3.1%.

but here's the real zinger. anyone care to guess what the average momentum premium from 2009 to present was?

2009 -82.68
2010 6.01
2011 7.26
2012 1.57
2013 7.9
2014 1.6
2015 20.66
2016 -20.34
2017 5.04

i get a CAGR of -15.4% for those 9 years.

:shock:
Keep calm and Boglehead on. KCBO.

larryswedroe
Posts: 15693
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Larry Swedroe: Disrupted Expectations

Post by larryswedroe » Sun Sep 16, 2018 8:12 am

grok
Just to point out that the data applies to CS momentum and only in stocks--TS MOM would have shown strong positive returns in both 08 and 09 I would think
a) long only momentum does not experience the type of crash you saw in 2008, and of course long only mutual funds, using momentum screens would not have seen that impact. Crashes occur in reversals where the stocks that had done the worst turn around, and CS momentum is short those stocks which tend to be high beta stocks.
b) scaling momentum would have dramatically lowered that loss in 2008
c) it's why IMO if going to run gain exposure to CS momentum it should be in a multi-factor fund that also gains that exposure across asset classes, providing diversification (and if target vol then you are scaling momentum also)
Best wishes
Larry
Last edited by larryswedroe on Sun Sep 16, 2018 9:09 am, edited 1 time in total.

norciom
Posts: 45
Joined: Wed Sep 28, 2016 4:46 pm

Re: Larry Swedroe: Disrupted Expectations

Post by norciom » Sun Sep 16, 2018 8:46 am

I am curios if I am the odd one out or this would apply to most people.

In 10 years I might be single or with my partner, in the current country or in my home country, higher or lower income, or a complete lifestyle change.
Given how much life can change in 10 years, a shift in the stock market probability is the thing that matters "a lot"?

I will give you that, if nothing in your life changes in the next 10 years, sure, CAPE is the one variable you can pay attention to.
Else the valuation uncertainty multiplied by normal human uncertainty turns all planning in a "how many angels can dance on the head of a pin?" discussion

User avatar
nisiprius
Advisory Board
Posts: 36466
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: Larry Swedroe: Disrupted Expectations

Post by nisiprius » Sun Sep 16, 2018 12:26 pm

I agree with you, Norclom.

It's all "assume a spherical cow." The arguments are "my spherical cow is better than your spherical cow."

Factor mavens: spherical cows should have tilted axes.

Index investors: cheap spherical cows are better than expensive spherical cows.

Dividend investors: spherical Holsteins give more milk.

Target date glide slope designers: if you simply lift a spherical calf every day, by the time it is grown you will be able to lift a 1,700-pound spherical bull.

Load fund advisors: We can already lift a load of spherical bull.

ETF investors: Spherical pigs are better than spherical bulls.

Active ETF investors: My spherical pig has lipstick on it.
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.

heyyou
Posts: 3095
Joined: Tue Feb 20, 2007 4:58 pm

Re: Larry Swedroe: Disrupted Expectations

Post by heyyou » Sun Sep 16, 2018 1:15 pm

Schiller has noted that those who had good returns in the accumulation years, should expect lower returns in the decumulation years. That level of forecasting accuracy is adequate for me.

McClung, writing in 2015, wrapped his 30 year data beyond 1985 by adding in 1926 and its following years' data. He then wrote that retirees could choose to not worry about their future periods of low returns, since historical periods of low returns were included in the data.

Thinking about the solutions instead of the problem: Keeping your necessary retirement expenses at a reasonable level, and only spending a calculated % of each recent annual portfolio value will compensate for future periods of low returns when they do occur.

User avatar
willthrill81
Posts: 5361
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Disrupted Expectations

Post by willthrill81 » Sun Sep 16, 2018 1:25 pm

heyyou wrote:
Sun Sep 16, 2018 1:15 pm
Schiller has noted that those who had good returns in the accumulation years, should expect lower returns in the decumulation years. That level of forecasting accuracy is adequate for me.
So is the inverse true as well, that those who had poor returns in the accumulation years should expect higher returns in the decumulation phase?
“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

columbia
Posts: 835
Joined: Tue Aug 27, 2013 5:30 am

Re: Larry Swedroe: Disrupted Expectations

Post by columbia » Mon Sep 17, 2018 4:18 am

So how is this actionable?

Be prepared to save more and spend less (if retired)? That seems obvious, but my real question is how this might effect the equity risk premium.

User avatar
packer16
Posts: 1012
Joined: Sat Jan 04, 2014 2:28 pm

Re: Larry Swedroe: Disrupted Expectations

Post by packer16 » Mon Sep 17, 2018 8:34 am

I wish folks would give error bands around their expected return estimates then we would know when there is an issue.

Packer
Buy cheap and something good might happen

User avatar
market timer
Posts: 5937
Joined: Tue Aug 21, 2007 1:42 am

Re: Larry Swedroe: Disrupted Expectations

Post by market timer » Mon Sep 17, 2018 9:34 am

nisiprius wrote:
Sun Sep 16, 2018 12:26 pm
Dividend investors: spherical Holsteins give more milk.

Target date glide slope designers: if you simply lift a spherical calf every day, by the time it is grown you will be able to lift a 1,700-pound spherical bull.
LOL! I'm partial to the NEET investors: Why buy the spherical cow when you can get the milk for free?

Post Reply