Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 11, 2018 2:06 am

I very much like the idea behind the LS portfolio and I had come up for myself with a barbell portfolio based on a similar philosophy. The LS portfolio has been discussed on this forum and is described e.g. here:
http://www.nytimes.com/2011/12/24/your- ... =undefined

However, I have had a doubt for a while concerning using SCV, since it appears that SC, all over the world, are overpriced relative to LC, as you can see e.g. here:
http://mebfaber.com/2015/07/23/small-cap-cape-ratios/
Similarly, if you compare the P/E ratios of MSCI US and MSCI US small cap value, you will see that the latter are higher than the former at present (I am not sure how the long term averages compare):
https://www.msci.com/documents/10199/67 ... b53e8d0d9f
and
https://www.msci.com/documents/10199/25 ... 2d4fd82423

In other words, one of the ideas behind the LP portfolio is that value outperforms, however by choosing to invest in small caps, you now own the more expensive segment of the market.

Linked to this question is the one of whether you can compare the valuations of SC versus LC. And then use the comparison to time your exposure. (For example, Peter Lynch claimed to have done so successfully in one of his books, but a value manager I corresponded with thinks Lynch was just lucky.)

Would be interested in your take on this.
When everyone is thinking the same, no one is thinking at all

david1082b
Posts: 195
Joined: Fri Jun 09, 2017 12:35 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by david1082b » Sun Feb 11, 2018 7:47 am

Looking at the 1 year fordward estimated P/E given in the second PDF:
MSCI US Small Cap Value 17.45
MSCI USA 18.89
(for January 31st). The forward p/es will be lower now after the stock declines no doubt. It could be that one-year forward P/E is not always so useful, due to notorious over-estimation by analysts. But I think one-year trailing is much less useful than forward estimations. For an idea of how questionable one-year trailing is, look at P/Es in Spring of 2009, huge losses in the previous year saw monster P/Es right at the bottom of the market. The PDFs use trailing I imagine for the P/E that has small value higher than MSCI USA.

Also note the lower price to book for small value, 1.66 versus 3.48 for MSCI USA. A lot of people prefer the S&P indexes due to positive earnings being a requirement for inclusion, i.e. small cap 600, which may have a different forward p/e than the MSCI index.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 11, 2018 9:27 am

david1082b wrote:
Sun Feb 11, 2018 7:47 am
Looking at the 1 year fordward estimated P/E given in the second PDF:
MSCI US Small Cap Value 17.45
MSCI USA 18.89
(for January 31st). The forward p/es will be lower now after the stock declines no doubt. It could be that one-year forward P/E is not always so useful, due to notorious over-estimation by analysts. But I think one-year trailing is much less useful than forward estimations. For an idea of how questionable one-year trailing is, look at P/Es in Spring of 2009, huge losses in the previous year saw monster P/Es right at the bottom of the market. The PDFs use trailing I imagine for the P/E that has small value higher than MSCI USA.

Also note the lower price to book for small value, 1.66 versus 3.48 for MSCI USA. A lot of people prefer the S&P indexes due to positive earnings being a requirement for inclusion, i.e. small cap 600, which may have a different forward p/e than the MSCI index.
Ah ok I had looked at what the pdf calls just P/Es (so like you say it's probably the trailing P/Es), but I see your point about forward P/Es being lower for small caps, which must reflect the fact that they have a higher rate of increase of earnings.

Do you know whether small cap funds like DFA use trailing of forward P/Es? (unless they used price to book ratios, but I had seen papers showing that P/E and expecially P/CF seem to work better than price to book ratios)
When everyone is thinking the same, no one is thinking at all

dbr
Posts: 26128
Joined: Sun Mar 04, 2007 9:50 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by dbr » Sun Feb 11, 2018 10:21 am

I think trying to overlay market timing onto a general portfolio theoretical result can only result in a mess. That is true even if you really believe you can succeed at tactical asset allocation. If I were to choose to tilt to SCV Larry Portfolio style, which I don't, I would plan on the strategy for long term investing and ignore valuation.

1nv35t
Posts: 111
Joined: Wed Dec 13, 2017 3:37 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by 1nv35t » Sun Feb 11, 2018 10:42 am

Large cap index are top bounded. Mid and small can feed in/out of both the top and bottom. At the extreme top of LC a single stock can be weighted 10%, or more for single sectors. Such concentration is a risk such that LC can lag. That LC and SCV are at different stages in their cycles may tempt you to time/switch, but more broadly its suggested its better to just stay the course as often attempted timing proves to have been worse than not having timed.

Conceptually yearly rebalancing back to target weightings would have you taking some of SCV gains off the table during good times, in readiness to be redeployed again at a later date. That in itself is sufficient 'trading' (cost averaging). Full rotations into other areas where relative 'value' might be apparent, could work, but might not work.

halfnine
Posts: 826
Joined: Tue Dec 21, 2010 1:48 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by halfnine » Sun Feb 11, 2018 3:36 pm

Lauretta wrote:
Sun Feb 11, 2018 2:06 am
...I very much like the idea behind the LS portfolio and I had come up for myself with a barbell portfolio based on a similar philosophy....
I've never seen anyone show that the LS portfolio is pervasive. As an investor (like yourself) outside of the USA, I am still waiting to be shown that it works well across individual countries.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 11, 2018 3:47 pm

halfnine wrote:
Sun Feb 11, 2018 3:36 pm
Lauretta wrote:
Sun Feb 11, 2018 2:06 am
...I very much like the idea behind the LS portfolio and I had come up for myself with a barbell portfolio based on a similar philosophy....
I've never seen anyone show that the LS portfolio is pervasive. As an investor (like yourself) outside of the USA, I am still waiting to be shown that it works well across individual countries.
Thanks for your feedback. I would think for it to work the small value premium should be pervasive in different countries, which I think it is (as seems to be implied by the fact that the LS for US investors has 50% in international).
So for a EU investor having 50% in EU and the rest in non EU should have a symilar exposure to currency risk and to SCV premium.
At least that's my understanding; actually when I made my AA I didn't know of this portfolio, I just overweighed SCV and EM; whilst having a larger part in safe cash-like assets than most people here at my age. So I have a kind of LS portfolio I think... I still have some large because small seemed expensive and also I understand that it may be the case that SCV may lag behind for long periods of time.
When everyone is thinking the same, no one is thinking at all

halfnine
Posts: 826
Joined: Tue Dec 21, 2010 1:48 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by halfnine » Sun Feb 11, 2018 4:17 pm

Lauretta wrote:
Sun Feb 11, 2018 3:47 pm
halfnine wrote:
Sun Feb 11, 2018 3:36 pm
Lauretta wrote:
Sun Feb 11, 2018 2:06 am
...I very much like the idea behind the LS portfolio and I had come up for myself with a barbell portfolio based on a similar philosophy....
I've never seen anyone show that the LS portfolio is pervasive. As an investor (like yourself) outside of the USA, I am still waiting to be shown that it works well across individual countries.
Thanks for your feedback. I would think for it to work the small value premium should be pervasive in different countries, which I think it is (as seems to be implied by the fact that the LS for US investors has 50% in international).
So for a EU investor having 50% in EU and the rest in non EU should have a symilar exposure to currency risk and to SCV premium.
At least that's my understanding; actually when I made my AA I didn't know of this portfolio, I just overweighed SCV and EM; whilst having a larger part in safe cash-like assets than most people here at my age. So I have a kind of LS portfolio I think... I still have some large because small seemed expensive and also I understand that it may be the case that SCV may lag behind for long periods of time.
The small value premium is pervasive in different countries according to Larry and I have no reason to disagree. However, SCV is only 30% of the entire portfolio with the other 70% being US treasuries. And while the entire portfolio backtests well in the USA over recent history, I have yet to see anyone show that it either works or doesn't work in other countries with other base currencies. So, I have my doubts as to whether the LS portfolio is pervasive.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 11, 2018 4:31 pm

halfnine wrote:
Sun Feb 11, 2018 4:17 pm
Lauretta wrote:
Sun Feb 11, 2018 3:47 pm
halfnine wrote:
Sun Feb 11, 2018 3:36 pm
Lauretta wrote:
Sun Feb 11, 2018 2:06 am
...I very much like the idea behind the LS portfolio and I had come up for myself with a barbell portfolio based on a similar philosophy....
I've never seen anyone show that the LS portfolio is pervasive. As an investor (like yourself) outside of the USA, I am still waiting to be shown that it works well across individual countries.
Thanks for your feedback. I would think for it to work the small value premium should be pervasive in different countries, which I think it is (as seems to be implied by the fact that the LS for US investors has 50% in international).
So for a EU investor having 50% in EU and the rest in non EU should have a symilar exposure to currency risk and to SCV premium.
At least that's my understanding; actually when I made my AA I didn't know of this portfolio, I just overweighed SCV and EM; whilst having a larger part in safe cash-like assets than most people here at my age. So I have a kind of LS portfolio I think... I still have some large because small seemed expensive and also I understand that it may be the case that SCV may lag behind for long periods of time.
The small value premium is pervasive in different countries according to Larry and I have no reason to disagree. However, SCV is only 30% of the entire portfolio with the other 70% being US treasuries. And while the entire portfolio backtests well in the USA over recent history, I have yet to see anyone show that it either works or doesn't work in other countries with other base currencies. So, I have my doubts as to whether the LS portfolio is pervasive.
that's a very good point but the same problem applies to other portfolios (for which I haven't seen backtests for Eu investors). Like the 60/40 works well in the US but I haven't seen results for the eurozone. But assuming a 60/40 worked for the EU, then the LS should work too because all it does is say that you get a higher CAGR for stocks by owning SCV; so you can affort to have a greater proportion of safe assets. So you just increase the amount of safe assets decrease that in stocks.
So I built my portfolio starting from examples I saw in the US, and then adapting it by taking account of the currency in which I will spend my money etc. Btw I think in the Us you have several advantages (e.g. very secure gov. bonds - with non negative yields unlike here! - and for stocks you represent more than 50% of the global market cap so you can own relatively little international without deviating too much from the global portfolio and you don't have to think about currency hedging). But I did my best to take the reasoning behind the AAs for a US investor and apply it to my situation.
However I agree a backtest would be reassuring!
When everyone is thinking the same, no one is thinking at all

User avatar
nedsaid
Posts: 9379
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by nedsaid » Sun Feb 11, 2018 5:03 pm

Although I use Price to Earnings as my primary market barometer, there are a few things you have to take into consideration.

First of all, you have to know which set of ratios you are looking at. I primarily look at forward P/E's based upon future estimated earnings. The market looks forward and so should we. I also look at P/E ratios based on historical earnings, which will tend to be higher than forward P/E's, a big reason being that earnings tend to grow over time and that Wall Street analysts tend to be optimistic.

Second, high P/E ratios can be caused by earnings growing very fast. Depressed earnings that the market expects to be temporary can also cause high P/E ratios.

You also have to factor in P/E behavior with cyclical stocks. This is a paradox, but economically sensitive stocks tend to have low P/E's when they are expensive and high P/E's when they are cheap. Again, this is a function of the market looking ahead. When the economy is going very strong late in the economic cycle, the market realizes that the good times will end and this knowledge is taken into account in the price. When earnings are weak in an economic downturn, prices don't fall as much as you would think because the market knows that earnings and the economy will rebound at some point.

You also have to understand that the market values consistency of earnings growth above all else. Companies with steady rates of earnings growth will have higher P/E ratios than companies with volatile earnings.

The rate of earnings growth is another factor. The tendency is for faster earnings growth to cause higher P/E ratios and for lower earnings growth to cause lower P/E ratios.

Quality of a company is a factor in P/E ratios too. The market rewards consistent earnings growth and strong balance sheets. One way to see how balance sheets affect P/E ratios is that cash rich companies command higher P/E ratios than companies that have little extra cash. So you want to see companies that have consistent earnings growth faster than economic growth with lots of assets and lower levels of debt on the balance sheet. A company with a strong balance sheet can weather bad times a lot better than a company with a weak balance sheet with lots of debt.

Growth stocks tend to have higher P/E ratios than Value stocks. Nedsaid's definition of a growth stock is a company that grows its earnings faster than the economy. Right now, Gross Domestic Product is growing about 3% and we have about 2% inflation. So on a nominal basis, the economy is growing at 5% a year or 3% after taking into account inflation. A true growth stock will have a higher than a 5% growth rate. A slow growth stock will grow earnings but at an equal or slower pace than the economy itself.

Stock nirvana or stock heaven is a quality company that grows earnings at about 8% a year or so, does it consistently, and has a strong balance sheet. This is the definition of quality. These type of companies tend to be in the "sweet spot" of the market and tend to have among the best returns. Trouble is, it isn't easy for a company to stay here, hence these companies also tend to be pretty well run.

Fast growing companies are smaller and have earnings growth rates of 10% a year, 15% a year, or even more. Microsoft in the 1980's and the 1990's was like this. A lot of people who worked for Microsoft are now millionaires. The company had 12-15 years of fast growth and it was a money machine. Growth rates like this tend not to last too long as at 15% you eventually take over the world. Eventually things slow down or something goes wrong. These companies often become the quality stocks everyone wants or just fall out of the sky like a shooting star.

Value stocks can be either quality companies having temporary problems or a less than perfect company for which the market has very low expectations. Value stocks tend to have more debt on their balance sheets and more volatile earnings. Hence lower expectations and lower P/E ratios.

So you not only need to know more about how the metric works but also the different kinds of stocks that make up the stock market.

Edit: One thing that I forgot is that the level of interest rates have an effect on P/E ratios too. Lower interest rates tend towards higher P/E ratios and higher interest rates tend towards lower P/E ratios. I use the word "tend" as there are many other factors that determine stock prices. For example, forward P/E ratios in early 2000 got to be about 32 as opposed to 20 today even though interest rates then where much higher than today. In early 2000, you had a lot of market euphoria and that overrode the interest rate effect.
A fool and his money are good for business.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sat Feb 17, 2018 9:41 am

david1082b wrote:
Sun Feb 11, 2018 7:47 am
Looking at the 1 year fordward estimated P/E given in the second PDF:
MSCI US Small Cap Value 17.45
MSCI USA 18.89
(for January 31st). The forward p/es will be lower now after the stock declines no doubt. It could be that one-year forward P/E is not always so useful, due to notorious over-estimation by analysts. But I think one-year trailing is much less useful than forward estimations. For an idea of how questionable one-year trailing is, look at P/Es in Spring of 2009, huge losses in the previous year saw monster P/Es right at the bottom of the market. The PDFs use trailing I imagine for the P/E that has small value higher than MSCI USA.

Also note the lower price to book for small value, 1.66 versus 3.48 for MSCI USA. A lot of people prefer the S&P indexes due to positive earnings being a requirement for inclusion, i.e. small cap 600, which may have a different forward p/e than the MSCI index.
Hi David, I hadn't thought about trailing PE versus future PE prior to your message so thanks for bringing it up; now I've just run across this article where Mr Bogle says he seems to prefer trailing, so I think (like in so many things in investing) there doesn't seem to be consensus on this. Here's what he says:
I'm a stern taskmaster. I use last year's actual earnings. And I use reported earnings. I'm using last year's reported earnings. Wall Street, of course, is using this year's projected earnings. So they come up with a PE of 16. So if you ask me, I'm not super confident at 26. If you ask anyone from Wall Street, they would say they are super confident at 16. The numbers are the numbers
http://money.cnn.com/2017/03/08/investi ... index.html
When everyone is thinking the same, no one is thinking at all

User avatar
nisiprius
Advisory Board
Posts: 35433
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's portfolio: does the fact that SC are expensive cause a problem?

Post by nisiprius » Sat Feb 17, 2018 9:45 am

Is it "a fact" that small caps are expensive? I continually read statements in the forum and elsewhere that one asset class or another is "expensive" or "cheap" and I never understand how anybody knows.
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
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sat Feb 17, 2018 10:22 am

nisiprius wrote:
Sat Feb 17, 2018 9:45 am
Is it "a fact" that small caps are expensive? I continually read statements in the forum and elsewhere that one asset class or another is "expensive" or "cheap" and I never understand how anybody knows.
Well on the general question of valuations, you may read Mr Swedroe's latest article on Int'l, where he quotes academic evidence showing that there's a pretty high correlation between current valuations and long term returns.
So when valuations are e.g. significantly below historical averages one can speak of a cheap market and expect higher returns and vice versa.
You may also find of interest the video of the panel of experts in 2017 Bogleheads meeting, in which both Mr Clements and Mr Bernstein are in favours of EM because they are cheap.
You may also find of interest the Intelligent Investor by Ben Graham, where, in the context of discussing whether an individual firm is cheap or expensive, he says we cannot do that with scientific accuracy but we can still form a judgement, and say of a firm that it is expensive a bit like we can say of a person that they are fat (without being able to say the exact weight they should be in order not to be overweight).
On a more general note, I know there are people who think that the smart attitude to have is to state that the price is always right, since the markets know better, as they are efficient, reflect all available information etc. This is a nice theory, but the reality is that there is substatial empirical evidence showing that this is not the case, so perhaps simply invoking that theory is not smart after all (as a scientist I like to examine the evidence rather than accept blindly a philosophy or a hypothesis).
In this context, you may find this presentation given at Imperial College, London, showing the results of empirical studies of the Endogenous Dynamics of Markets, of interest:
http://wwwf.imperial.ac.uk/~amijatov/Ev ... uchaud.pdf
When everyone is thinking the same, no one is thinking at all

Random Walker
Posts: 2477
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Random Walker » Sat Feb 17, 2018 12:11 pm

Hi Lauretta,
My thoughts are pretty simple ones. If you believe that size and value are risk premia, then the expected return should always be positive relative to large cap or total market. The gap in valuations and additional expected return may be higher or lower at different times, but the expected return premia should always be positive. Since I believe timing factors is very hard or nearly impossible, I think it makes best sense to just create a plan and stick to it through thick and thin.
Moreover, since size is weakly correlated with market, value uncorrelated with market, and size and value uncorrelated with each other, there should always be the potential for divesification benefit.

Dave

User avatar
siamond
Posts: 3819
Joined: Mon May 28, 2012 5:50 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by siamond » Sat Feb 17, 2018 1:30 pm

The idea of a 'barbell' portfolio came from Nassim Taleb, Larry Swedroe just anchored himself to this fundamental idea of combining a very 'safe' vehicle (e.g. 70% IT bonds) with a smaller portion of highly risky vehicles (e.g. SCV, Emerging, Int'l Small - say 10% of each).

For this to work, it is reliant on the fact that the highly risky vehicles display occasional big spurts of growth. But then, by nature, such occurrences are very random, and one might have to wait more than a decade to witness one of those. This is Taleb's point, find a way to leverage extreme randomness. The idea seems attractive from an intellectual standpoint and has no reason to be US-specific, but sheesh, this is a hell of a bet, and requires to stick to the concept in an extremely steadfast manner (and certainly not to change one's mind depending on the valuation of the day or year). One way to make the bet more effective, as suggested by Taleb, is to diversify the 'risky' vehicles, as it would seem rather foolish to bet on a single asset class to concentrate all the risk on, notably when 70% is anchored on slow-moving treasuries... Let's also not forget that IT bonds have very fundamental risks on their own right (low returns in any case; and for non-TIPS, inflation). Yeah, historically, the SCV/IT-Bonds combination worked remarkably well (in the US, at least), but would you bet your retirement on the fact that this will happen again? Maybe not.

To illustrate the highly variable trajectory of such 'risky' asset classes, you might want to look at this blog article, with updated charts and stats about such risk asset classes, up and including 2017 returns.

Personally, I would not have the strength to use a barbell portfolio. This clearly would require nerves made of titanium. I much prefer to anchor most of my portfolio on a Total-Market approach (US and Int'l), and then use some tilts towards 'more risky' asset classes. Which is a very different strategy.

Random Walker
Posts: 2477
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Random Walker » Sat Feb 17, 2018 2:22 pm

This clearly would require nerves made of titanium.
Really? The large dose of safe bonds should narrow the SD. One could create a TSM/TBM portfolio with same expected return but it would have a larger equity allocation, exposure to only one equity risk factor as opposed to three, and highly likely greater volatility and bigger maximum drawdown.
Larry’s new version of Reducing The Risk Of Black Swans should be out in a month or so. This is basically his thesis.

Dave

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sat Feb 17, 2018 2:29 pm

siamond wrote:
Sat Feb 17, 2018 1:30 pm
The idea of a 'barbell' portfolio came from Nassim Taleb, Larry Swedroe just anchored himself to this fundamental idea of combining a very 'safe' vehicle (e.g. 70% IT bonds) with a smaller portion of highly risky vehicles (e.g. SCV, Emerging, Int'l Small - say 10% of each).

For this to work, it is reliant on the fact that the highly risky vehicles display occasional big spurts of growth. But then, by nature, such occurrences are very random, and one might have to wait more than a decade to witness one of those. This is Taleb's point, find a way to leverage extreme randomness. The idea seems attractive from an intellectual standpoint and has no reason to be US-specific, but sheesh, this is a hell of a bet, and requires to stick to the concept in an extremely steadfast manner (and certainly not to change one's mind depending on the valuation of the day or year). One way to make the bet more effective, as suggested by Taleb, is to diversify the 'risky' vehicles, as it would seem rather foolish to bet on a single asset class to concentrate all the risk on, notably when 70% is anchored on slow-moving treasuries... Let's also not forget that IT bonds have very fundamental risks on their own right (low returns in any case; and for non-TIPS, inflation). Yeah, historically, the SCV/IT-Bonds combination worked remarkably well (in the US, at least), but would you bet your retirement on the fact that this will happen again? Maybe not.

To illustrate the highly variable trajectory of such 'risky' asset classes, you might want to look at this blog article, with updated charts and stats about such risk asset classes, up and including 2017 returns.

Personally, I would not have the strength to use a barbell portfolio. This clearly would require nerves made of titanium. I much prefer to anchor most of my portfolio on a Total-Market approach (US and Int'l), and then use some tilts towards 'more risky' asset classes. Which is a very different strategy.
Thanks for your input; yes of course I knew about the idea in Taleb; that's one of the reason I implemented a symilar portfolio; but the main reason is that I felt very comfortable with it: it's interesting because I feel exactly the opposite as you when you say the barbell strategy
clearly would require nerves made of titanium
For me this strategy is much more reassuring than say a 60/40 portfolio because stocks are never garanteed from a black swan event, so having a larger proportion in a safe asset (for me in Europe it's cash at the moment) seems a lot safer. Also since I started investing a year ago; SCV in EU have returned around 30% in my fund; and LC around 10% so it feels that LC are a bit pointless. Perhaps I was just lucky and started to invest at the right time for SCV in EU.
Finally I don't see why you say that a total market portfolio with a tilt towards SCV is a very different strategy. It seems to me that it depends on how much you tilt. In my understanding there is no discrete discontinuity between the two strategies, there seems to be more like a continuum going between the extreme of say a 60/40 and a LS portfolio.
When everyone is thinking the same, no one is thinking at all

drk
Posts: 331
Joined: Mon Jul 24, 2017 10:33 pm
Location: Seattle

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by drk » Sat Feb 17, 2018 2:56 pm

Random Walker wrote:
Sat Feb 17, 2018 2:22 pm
This clearly would require nerves made of titanium.
Really? The large dose of safe bonds should narrow the SD. One could create a TSM/TBM portfolio with same expected return but it would have a larger equity allocation, exposure to only one equity risk factor as opposed to three, and highly likely greater volatility and bigger maximum drawdown.
Larry’s new version of Reducing The Risk Of Black Swans should be out in a month or so. This is basically his thesis.

Dave
I suspect that the titanium nerves come into play when trying to stick to that portfolio in the face of a big bull market (e.g., the last nine years).

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sat Feb 17, 2018 3:12 pm

drk wrote:
Sat Feb 17, 2018 2:56 pm
Random Walker wrote:
Sat Feb 17, 2018 2:22 pm
This clearly would require nerves made of titanium.
Really? The large dose of safe bonds should narrow the SD. One could create a TSM/TBM portfolio with same expected return but it would have a larger equity allocation, exposure to only one equity risk factor as opposed to three, and highly likely greater volatility and bigger maximum drawdown.
Larry’s new version of Reducing The Risk Of Black Swans should be out in a month or so. This is basically his thesis.

Dave
I suspect that the titanium nerves come into play when trying to stick to that portfolio in the face of a big bull market (e.g., the last nine years).
good point even though to be fair Mr Swedroe uses SCV funds from DFA; e.g. for Int'l I think it would be DISVX (not sure as I don't live in the US) instead of VSS which is a SC not SCV.
Still like you say in the last bull market the LS portfolio would have lagged behind.
When everyone is thinking the same, no one is thinking at all

drk
Posts: 331
Joined: Mon Jul 24, 2017 10:33 pm
Location: Seattle

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by drk » Sat Feb 17, 2018 3:21 pm

Lauretta wrote:
Sat Feb 17, 2018 3:12 pm
good point even though to be fair Mr Swedroe uses SCV funds from DFA; e.g. for Int'l I think it would be DISVX (not sure as I don't live in the US) instead of VSS which is a SC not SCV.
Still like you say in the last bull market the LS portfolio would have lagged behind.
Touché. For comparison's sake, I swapped in some DFA funds: DFSVX for US small-cap value, DISVX for international small-cap value, and DEMSX for emerging markets small-cap (DFEVX seemed too big). It's better but still not very pretty.

jmk
Posts: 322
Joined: Tue Nov 01, 2011 7:48 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by jmk » Sat Feb 17, 2018 3:36 pm

The underlying point LP is to use higher expected return funds in order to increase safe portion to control tail risk. So if sc doesn’t suit you you could use more emerging or a smart beta fund.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sat Feb 17, 2018 3:41 pm

jmk wrote:
Sat Feb 17, 2018 3:36 pm
The underlying point LP is to use higher expected return funds in order to increase safe portion to control tail risk. So if sc doesn’t suit you you could use more emerging or a smart beta fund.
thank you, good point, you are right of course, and I am indeed a bit more confortable with EM.
When everyone is thinking the same, no one is thinking at all

User avatar
siamond
Posts: 3819
Joined: Mon May 28, 2012 5:50 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by siamond » Sat Feb 17, 2018 7:18 pm

Lauretta wrote:
Sat Feb 17, 2018 2:29 pm
Thanks for your input; yes of course I knew about the idea in Taleb; that's one of the reason I implemented a symilar portfolio; but the main reason is that I felt very comfortable with it: it's interesting because I feel exactly the opposite as you when you say the barbell strategy
clearly would require nerves made of titanium
For me this strategy is much more reassuring than say a 60/40 portfolio because stocks are never guaranteed from a black swan event, so having a larger proportion in a safe asset (for me in Europe it's cash at the moment) seems a lot safer. Also since I started investing a year ago; SCV in EU have returned around 30% in my fund; and LC around 10% so it feels that LC are a bit pointless. Perhaps I was just lucky and started to invest at the right time for SCV in EU.
It's really a matter of perspective. If you look at past performance, yes, a barbell portfolio in the US provided excellent returns with a very smooth ride (i.e. low volatility). What I find extremely unnerving is the reliance on those rare spurts of growth of one of the risky assets. Black swan events are by their very nature extremely unpredictable. If you look carefully at the telltale charts I posted in the blog article, you'll see that, in the past, we had many cases where SCV did NOT provide any premium for more than a decade (and EM and Int'l Small had similar behaviors). When looking at that in retrospect, yes, it worked, very cool, but can you imagine staying the course for 10 or 15 years, with all the pressure of financial writers clamoring "the SCV premium is dead for good this time, blah", all that before it finally (maybe) pans out again? I don't think I could stand that. Simply look at what you just stated, pondering about last's year performance (one single year!), about valuations of the day, etc? This is your brain being very active and in truth, taking in way too much noise that would need to be forcibly set aside to stay the course with a barbell portfolio... Not an easy feat to do.
Lauretta wrote:
Sat Feb 17, 2018 2:29 pm
Finally I don't see why you say that a total market portfolio with a tilt towards SCV is a very different strategy. It seems to me that it depends on how much you tilt. In my understanding there is no discrete discontinuity between the two strategies, there seems to be more like a continuum going between the extreme of say a 60/40 and a LS portfolio.
Well, yes, you're right, there is a mathematical continuity of sorts, and it all depends on your exact numbers, but those are still very far apart, and definitely displayed quite different behaviors in the past. And there is definitely a great comfort factor of primarily relying on the whole market, instead of the risky/moody subsets. If you are Excel-inclined, don't stop at the literature from Swedroe (which, quite frankly, isn't terribly well researched), download the Simba backtesting spreadsheet, and play around for a while. The more you research and learn by yourself, the more you'll truly trust whatever strategy you'll settle on, and the better chance you have of staying the course.

PS. I just noticed you're Italian. Hm, then there is another consideration. If I am not mistaken, inflation has been quite a problem in Italy in the past few decades. Remember, inflation is THE primary risk for government bonds. Unless you use a good chunk of TIPS (well, the Italian equivalent?), but that is no magic bullet, and come with its share of uncertainties. Bonds are NOT safe, the truth is they present different risks than stocks.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sat Feb 17, 2018 7:32 pm

siamond wrote:
Sat Feb 17, 2018 7:18 pm
If you are Excel-inclined, don't stop at the literature from Swedroe (which, quite frankly, isn't terribly well researched), download the Simba backtesting spreadsheet, and play around for a while. The more you research and learn by yourself, the more you'll truly trust whatever strategy you'll settle on, and the better chance you have of staying the course.

PS. I just noticed you're Italian. Hm, then there is another consideration. If I am not mistaken, inflation has been quite a problem in Italy in the past few decades. Remember, inflation is THE primary risk for government bonds. Unless you use a good chunk of TIPS (well, the Italian equivalent?), but that is no magic bullet, and come with its share of uncertainties. Bonds are NOT safe, the truth is they present different risks than stocks.
Thank you for your feedback, ok I'll download the spreadsheet tomorrow. Yes you are right about inflation, they used to devalue the lira a lot, but that was before we joined the euro; now inflation is quite low in the Eurozone but in future it might be interesting for me to look at inflation linked bonds. At the moment they have negative yields though...
When everyone is thinking the same, no one is thinking at all

User avatar
siamond
Posts: 3819
Joined: Mon May 28, 2012 5:50 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by siamond » Sun Feb 18, 2018 11:45 am

Lauretta, let me ask you something. Do you actually invest with Italy-only funds (or Europe-only) funds? Or do you actually use US funds (while dealing with local exchange rates and inflation)? Could you please share the specifics of your asset allocation? I would be curious to research a bit more the case of an European investor like you, but I need to better understand your choices...

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 18, 2018 12:13 pm

siamond wrote:
Sun Feb 18, 2018 11:45 am
Lauretta, let me ask you something. Do you actually invest with Italy-only funds (or Europe-only) funds? Or do you actually use US funds (while dealing with local exchange rates and inflation)? Could you please share the specifics of your asset allocation? I would be curious to research a bit more the case of an European investor like you, but I need to better understand your choices...
Hi Siamond,
Thank you for your interest. I use funds and ETFs domiciled in Europe because for tax reasons it's easier/more advantageus and also btw following the Mifid II regulation many US funds are no longer available to us EU investors.

I have more than a third of my assets in rental real estate. However, as far as my liquid assets are concerned, my AA is roughly as I detailed it here (I started investing seriously in stocks at the end of 2016 and expecially during 2017 following some sale of real estate) :
viewtopic.php?f=1&t=222724
I may add further to that post that in Japan I have roughly 50% LC and 50% SCV; I have a somewhat larger proportion of EM (nearly 20% of the stock allocation) and for the US I have only LC because as I mention in this post I understood that SC are expensive particularly in the US. Since setting up that portfolio the stocks part have done significantly better than Msci World in euro terms so I can't complain; though of course one year means nothing.
Would be interested in any feedback/advice/suggestion.
When everyone is thinking the same, no one is thinking at all

User avatar
siamond
Posts: 3819
Joined: Mon May 28, 2012 5:50 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by siamond » Sun Feb 18, 2018 12:53 pm

Ok, got it. Not truly a barbell portfolio, but close. So there is no Italy-only investment on the stock side. I think most of what you described could be mapped to 1970+ data series for a custom version of Simba. Adding some math for the local inflation and currency exchange rate. A bit of work, but entirely feasible. This could be an interesting mini-project.

Now what about the bonds? Are they Italian government bonds? Or more diversified?
Small cap value Euro 12,5%
MSCI Europe 7,5%
Emerging Markets 7,5%
Japan + MSCI Pacific 7,5%
US 7,5 %
Vanguard World Value 7,5%
Gold 4%
Cash+bonds 46%

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 18, 2018 1:05 pm

siamond wrote:
Sun Feb 18, 2018 12:53 pm

Now what about the bonds? Are they Italian government bonds? Or more diversified?
Actually at the time I wrote the post I had made up my mind (sort of) about the stock allocation but didn't know about bonds yet - that's why I wrote cash+bonds. But in practice I have only cash and cash-like investments at the moment which garantee the invested capital, because Eurozone bonds have negative yields.
Once QE is relaxed and interest rates become positive, I think the ideal would be to have a global bond Etf currency hedged to euro to diversify credit risk (at the moment the cost of hedging (due to interest rate differential) would make even US bonds unattractive) and perhaps some eurozone bonds inflation linked.
But to back-test the portfolio I don't know whether using a global bond currency hedged fund is possible?
When everyone is thinking the same, no one is thinking at all

lazyday
Posts: 3247
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by lazyday » Sun Feb 18, 2018 1:09 pm

Lauretta, on expensive smallcap stocks:

If you haven’t seen it yet, check out https://interactive.researchaffiliates. ... -beta.html which attempts to estimate 5 year excess returns from various long only strategies including Small Cap. The page also gives estimates for 5 yr returns from various long/short factors within largecap and smallcap.

For example, select Strategies near the top left, and scroll down to select Small Cap near the bottom left. On the bottom right select “Model” and click the button to the right of i (information) to make Model take most of the screen. Each dot represents a historical relative valuation* of Small Cap to the broad market on the X axis, and the following 5 year excess return over the broad market on the Y axis. The highlighted dot shows our recent relative valuation on the fitted curve, which predicts 5 yr returns.

Near the top of the screen you can select US, Developed (excluding US), and EM. By this analysis, US Small does not seem expensive, but Developed and EM sure do.

The data seems to be updated every 3 months, with perhaps a 1-2 month delay. Recently it was updated with end of year data.

* Valuation of Small divided by valuation of the broad market. Look for Methodology links which explain in some detail. I don’t recall if it’s simply P/B or something else.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 18, 2018 1:54 pm

lazyday wrote:
Sun Feb 18, 2018 1:09 pm
Lauretta, on expensive smallcap stocks:

If you haven’t seen it yet, check out https://interactive.researchaffiliates. ... -beta.html which attempts to estimate 5 year excess returns from various long only strategies including Small Cap. The page also gives estimates for 5 yr returns from various long/short factors within largecap and smallcap.

For example, select Strategies near the top left, and scroll down to select Small Cap near the bottom left. On the bottom right select “Model” and click the button to the right of i (information) to make Model take most of the screen. Each dot represents a historical relative valuation* of Small Cap to the broad market on the X axis, and the following 5 year excess return over the broad market on the Y axis. The highlighted dot shows our recent relative valuation on the fitted curve, which predicts 5 yr returns.

Near the top of the screen you can select US, Developed (excluding US), and EM. By this analysis, US Small does not seem expensive, but Developed and EM sure do.

The data seems to be updated every 3 months, with perhaps a 1-2 month delay. Recently it was updated with end of year data.

* Valuation of Small divided by valuation of the broad market. Look for Methodology links which explain in some detail. I don’t recall if it’s simply P/B or something else.
thank you! :happy That's a great site! And thanks for taking the time to explain how to use it.
I am going to explore it but apparently SC don't seem particularly expensive in the US using e.g. aggregate valuations (which from the Methodology seem to be the geometrical mean of 4 valuation methods (P/B, P/S, P/E and P/Div)).
On the other hand SV seem more expensive than the market in Int'l and EM.
I must play around with it more to understand it better (I am quite excited about it :happy ) but at first sight, from the 'Performance history' window, it looks like RA way of computing relative valuation makes small cap look less expensive than the method described here which uses only P/E (though trends are similar of course):
http://mebfaber.com/2013/11/26/small-ca ... expensive/
Also one take away is probably that using CAPE, as in the link I originally posted:
http://mebfaber.com/2015/07/23/small-cap-cape-ratios/
is probably not very useful for small caps as they have larger earnings growth, so that if you divide by the earnings over the last 10 years they look more expensive than they are.
Anyway thanks again for sharing this link.
When everyone is thinking the same, no one is thinking at all

lazyday
Posts: 3247
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by lazyday » Sun Feb 18, 2018 2:20 pm

Glad you like it :)

You can also see a chart of US Small PE10 over time, on a different RA site: https://interactive.researchaffiliates. ... e=Equities

If that link doesn't put you there, then select an equity on the left side, such as US Small. On the right side, select Charts and then "CAPE Ratio Time Series". If you put your mouse cursor over the top line in the chart, the other ones turn grey so it's less difficult to read.

User avatar
Lauretta
Posts: 538
Joined: Wed Jul 05, 2017 6:27 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Lauretta » Sun Feb 18, 2018 3:01 pm

lazyday wrote:
Sun Feb 18, 2018 2:20 pm
Glad you like it :)

You can also see a chart of US Small PE10 over time, on a different RA site: https://interactive.researchaffiliates. ... e=Equities

If that link doesn't put you there, then select an equity on the left side, such as US Small. On the right side, select Charts and then "CAPE Ratio Time Series". If you put your mouse cursor over the top line in the chart, the other ones turn grey so it's less difficult to read.
Thanks again for the link :happy I have a few takeways from these charts:
- So yes it looks indeed like it's normal that the CAPE for small caps should be much higher than that of the market - this must be due to their higher earning growth.
- EM seem to be the the most promising class by value metrics, as confirmed by GMO's predictions discussed in a separate thread. (And I also saw a video (2017 Boglehead conference) in which Mr Bernstein said that he usually advices to underweight them but he thinks that now because of their valuations their expected long term returns should be relatively high).
- Concerning the US my understanding is that predictions based on valuations make it the least attractive market. However some people believe that higher valuations are justified by the tech sector, which according to them deserves to be much more expensive because of vertical growth (a friend who used to manage a big Fidelity fund recently wrote to me about 'vertical growth' in the US, but I learnt only yesterday what that means after seeing a video of Peter Thiel :D ) and the fact that FANGs are more or less monopolys etc.
Interesting questions. Anyway thanks again for providing the links and for taking the time to explain how to use the site :happy
When everyone is thinking the same, no one is thinking at all

Nahtanoj
Posts: 5
Joined: Tue Apr 04, 2017 7:01 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Nahtanoj » Mon Feb 19, 2018 12:36 am

The Research Affiliates site is interesting, but it does not present an encouraging picture for US small caps. It's true that the linked page shows expected returns for US small caps as being close to those of the US market benchmark. But the page also indicates expected real returns of approximately 0% per annum for US small caps - with an SD of annual returns of approximately 18%.

In other words, US small cap may not be any more overvalued than the US market as a whole, but the US market as a whole does not seem priced to produce an attractive return, according to the RA model.

Of course, this is just the output of a model, and the ultimate returns will be different (and furthermore, the characteristics of the whole portfolio are more important than those of just one standalone asset class). But still - a US investor today can get a pre-tax 0% per annum real return in five-year US Treasury inflation-protected securities, with a lot lower standard deviation of annual returns than 18%.

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by in_reality » Mon Feb 19, 2018 5:45 am

Nahtanoj wrote:
Mon Feb 19, 2018 12:36 am
The Research Affiliates site is interesting, but it does not present an encouraging picture for US small caps. It's true that the linked page shows expected returns for US small caps as being close to those of the US market benchmark. But the page also indicates expected real returns of approximately 0% per annum for US small caps - with an SD of annual returns of approximately 18%.

In other words, US small cap may not be any more overvalued than the US market as a whole, but the US market as a whole does not seem priced to produce an attractive return, according to the RA model.

Of course, this is just the output of a model, and the ultimate returns will be different (and furthermore, the characteristics of the whole portfolio are more important than those of just one standalone asset class). But still - a US investor today can get a pre-tax 0% per annum real return in five-year US Treasury inflation-protected securities, with a lot lower standard deviation of annual returns than 18%.
Do understand that we don't know what valuations will be in the future. You were looking at the valuation dependent model which assumes CAPE reversion to the average.

If you choose the Growth and Yield model, the US doesn't look so bad:

3.1% US Large
3.3% US Small
3.9% EM
4.2% EAFE

Personally I think there will be at least some CAPE reversion, but can't predict when or how much. RAFI also had a paper up claiming that looking at real yields along with PE is more informative short term, but I don't think it's entirely useful.
Shiller’s cyclically adjusted P/E ratio (henceforth P/E) is a potent predictor of long-horizon capital market returns all over the world. It is, however, much less successful in predicting short-term returns. The authors describe the empirical relationships among valuation, inflation, and real interest rates, and demonstrate that conditioning P/E on current inflation and real yields substantially improves its accuracy in forecasting returns for periods from one month to one year. This result suggests that valuation, always an effective tool for long-term investors, can also become useful for assessing short-term market prospects. In addition, the modeling technique the authors developed in the course of this research shows there are more powerful ways to inte­grate macroeconomic measures with stock valuation methods than the linear combinations that predominate in quantitative studies.
https://www.researchaffiliates.com/en_u ... tions.html

asif408
Posts: 1233
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by asif408 » Mon Feb 19, 2018 6:43 am

Random Walker wrote:
Sat Feb 17, 2018 12:11 pm
If you believe that size and value are risk premia, then the expected return should always be positive relative to large cap or total market. The gap in valuations and additional expected return may be higher or lower at different times, but the expected return premia should always be positive.
I agree with Dave's post above in that you believe it is pervasive or you don't. My own personal opinion is that it is not, and no strategy is pervasive, and almost any can have great periods but that none will persist and will eventually revert back to an average. From what I've seen described of the history of the small value premium, I'm not aware of anyone who could have easily implemented the strategy during the long time frames cited. Before small value funds you would have had to buy the individual stocks, which have larger spreads that large stocks, so you would have lost (possibly all) of the premium to spreads and trasaction costs, which were much higher before the 1980s. I do agree that historically they did outperform in the past, but I attribute a large part of that to its lack of easy availability. Most strategies that worked in the past have a large element of this.

So I am basically strategy agnostic and believe in reversion to a mean more strongly than anything else. No strategy is always going to outperform and no strategy is always going to underperform, it depends on "timing".

lazyday
Posts: 3247
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by lazyday » Mon Feb 19, 2018 7:02 am

Random Walker wrote:
Sat Feb 17, 2018 12:11 pm
If you believe that size and value are risk premia, then the expected return should always be positive relative to large cap or total market. The gap in valuations and additional expected return may be higher or lower at different times, but the expected return premia should always be positive. ....
Should, but maybe not always true? I'm pretty sure that by the dividend discount model on January 1 2000, the expected equity risk premium of the S&P 500 over 30 year TIPS was negative. I don't see why other risk premia can't also be negative. Sometimes you're paid to take risk, occasionally you pay to take risk.

Random Walker
Posts: 2477
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by Random Walker » Mon Feb 19, 2018 4:07 pm

lazyday wrote:
Mon Feb 19, 2018 7:02 am
Random Walker wrote:
Sat Feb 17, 2018 12:11 pm
If you believe that size and value are risk premia, then the expected return should always be positive relative to large cap or total market. The gap in valuations and additional expected return may be higher or lower at different times, but the expected return premia should always be positive. ....
Should, but maybe not always true? I'm pretty sure that by the dividend discount model on January 1 2000, the expected equity risk premium of the S&P 500 over 30 year TIPS was negative. I don't see why other risk premia can't also be negative. Sometimes you're paid to take risk, occasionally you pay to take risk.
Not sure I totally agree, but certainly get your point! Wouldn’t DDM still yield a positive result? Expected return = dividend yield + earnings growth? It’s the RTM of valuations (Bogle’s Speculative component) that really made expected returns so terrible. Small and Value are certainly risk factors, so certainly expect risk to show up periodically (and potentially long stretches too)

Dave

lazyday
Posts: 3247
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe's portfolio: does the fact that SC are expensive cause a problem?

Post by lazyday » Mon Feb 19, 2018 4:57 pm

Random Walker wrote:
Mon Feb 19, 2018 4:07 pm
Wouldn’t DDM still yield a positive result? Expected return = dividend yield + earnings growth?
Yes, but if S&P 500 yield+growth is 3% and 30 yr TIPS yield 4%, then the premium over TIPS is negative.

I guess it depends on whether you are looking at an equity premium compared to a risk free or lower risk asset, or just the total premium from owning equities.

Post Reply