Article that highest risk is not highest reward

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
DaleMaley
Posts: 1582
Joined: Thu Mar 01, 2007 8:04 pm
Location: Fairbury, Illinois
Contact:

Article that highest risk is not highest reward

Post by DaleMaley » Thu Apr 26, 2012 5:34 am

Less risk offers more reward, study finds in InvestmentNews magazine.

http://www.investmentnews.com/article/2 ... /304229996

Studies find that highest risk stocks do not give highest rewards. One of the studies cited only covers time period of 1990 to 2011.
Most investors, both institutional and individual, will find that the best way to own common stocks is through an index fund that charges minimal fees. – Warren Buffett

richard
Posts: 7961
Joined: Tue Feb 20, 2007 3:38 pm
Contact:

Re: Article that highest risk is not highest reward

Post by richard » Thu Apr 26, 2012 5:49 am

The study found that the most volatile stocks did not have the highest returns, not that the highest risk stocks don't have the highest returns. Since volatility is not a very good proxy for risk, this should not be a surprise.

See, for example, the Fama French three factor model and many other models developed in the past 50 years, all of which do a better job of identifying factors associated with higher returns. Such factors are generally labelled risk factors, although the exact underlying economic risk is typically unidentified.

Brian2d
Posts: 213
Joined: Thu Jan 20, 2011 9:04 pm

Re: Article that highest risk is not highest reward

Post by Brian2d » Thu Apr 26, 2012 7:01 am

The methodology of a study like this is key:

1. How does it handle new stocks and stocks which disappear due to bankruptcy/etc?
Buy and hold returns on volatile stocks may be lower because you are stuck with losers but don't get to add new stocks which become winners...rebalancing is key. Does the study include rebalancing in some way?

2. Sensitive to time frame. 1990-2011 results may not be repeated.
In this period the risk of getting lower returns occurred (that's why there's risk). May not be the case next time.

User avatar
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Re: Article that highest risk is not highest reward

Post by Lbill » Thu Apr 26, 2012 7:14 am

I think we already know this. It's the reason most people diversify their assets and own bonds as well as stocks and rebalance - this lowers overall portfolio risk and improves returns over the long run.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Thu Apr 26, 2012 7:29 am

The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed. Uncertainty of expected returns (not price volatility) is a reasonable and widely accepted proxy for risk. Ex post realized returns often do not match ex ante expected returns. If they did, there would be no risk, and therefore no risk premium. It must be so.

If volatility, whether of price or returns, could be relied upon to predict long-term returns, then rational investors with long investment horizons would simply invest in high volatility investments. This would drive the price of these investments up, reducing their expected returns, so the long-term returns would no longer be proportional to volatility. High volatility investments must disappoint from time to time, otherwise there would be no risk, and the volatility would decline.

Conclusions cannot be drawn from a single twenty year period. If they could, then one could easily conclude that bonds provide a higher return than stocks.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

richard
Posts: 7961
Joined: Tue Feb 20, 2007 3:38 pm
Contact:

Re: Article that highest risk is not highest reward

Post by richard » Thu Apr 26, 2012 7:53 am

The more I see studies based on historic data, the more I'm convinced that we're better off investing based on fundamental principles, such as the relation between expected return and risk, diversification, the low likelihood of consistently beating the market, low expenses, etc.

We don't have enough market data for historic information to be of much significance for the long term investor and we don't have any assurances that present and future conditions will be sufficiently similar to past conditions to generate comparable outcomes.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Article that highest risk is not highest reward

Post by larryswedroe » Thu Apr 26, 2012 8:20 am

What is shows is that volatility is just ONE MEASURE of risk and clearly not the only thing investors care about.
Best wishes
Larry

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Article that highest risk is not highest reward

Post by pkcrafter » Thu Apr 26, 2012 11:47 am

richard wrote:The more I see studies based on historic data, the more I'm convinced that we're better off investing based on fundamental principles, such as the relation between expected return and risk, diversification, the low likelihood of consistently beating the market, low expenses, etc.

I think you've got it! When I first found the Bogleheads 14 years ago I read everything I could find on theory. It was all new and exciting, but it all kept coming to the same conclusion--the one you've posted above. But that is why investing as a Boglehead really is easy, you don't really need anything more that what you've listed. John Bogle knew this a long time ago.

We don't have enough market data for historic information to be of much significance for the long term investor and we don't have any assurances that present and future conditions will be sufficiently similar to past conditions to generate comparable outcomes.

Right. We don't have enough long term data to be mathematically significant. However, if we did it still would be of little help because the next 10-20 years will not fit the long term data. And the last 10 years of data and the 20 years before that strongly supports that idea.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Article that highest risk is not highest reward

Post by pkcrafter » Thu Apr 26, 2012 12:26 pm

Article quote:
Veteran quantitative-investment analyst and economist Robert Haugen studied every stock market in the world from 1990 to 2011 and found that the average return of the least volatile stocks outperformed the most volatile stocks by an average of 17 percentage points.

I'm very leery of the time frame here, may be data mining.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Article that highest risk is not highest reward

Post by yobria » Thu Apr 26, 2012 12:57 pm

pkcrafter wrote:Article quote:
Veteran quantitative-investment analyst and economist Robert Haugen studied every stock market in the world from 1990 to 2011 and found that the average return of the least volatile stocks outperformed the most volatile stocks by an average of 17 percentage points.

I'm very leery of the time frame here, may be data mining.
Yes, highly suspicious. Volatility is a very easy statistic to capture (unlike, say, book value). So why not show us 50 years of data over 20 countries? No way that series disproves the null hypothesis that more risk = more reward.

For example, small cap stocks have higher volatility, and higher return, over the long term in all studies I've seen.

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: Article that highest risk is not highest reward

Post by GregLee » Fri Apr 27, 2012 1:45 pm

Kevin M wrote:The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed.
Use of the term "axiom" suggests it is taken to be true without evidence.
Greg, retired 8/10.

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Fri Apr 27, 2012 3:41 pm

GregLee wrote:
Kevin M wrote:The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed.
Use of the term "axiom" suggests it is taken to be true without evidence.
OK, maybe "axiom" isn't the best word, but I think it fits. From Wikipedia: "Outside logic and mathematics, the term "axiom" is used for any established principle of some field." Seems to me that what I said is an established principle in the field of investing. Just read any investing or finance textbook, and you will see statements to this effect.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: Article that highest risk is not highest reward

Post by GregLee » Fri Apr 27, 2012 4:04 pm

Kevin M wrote:OK, maybe "axiom" isn't the best word, but I think it fits.
I agree that it is an appropriate term. It's just that, in this context, you seemed to be inferring that it's true, while "axiom" means that it's assumed. Just because people regard it as true, that doesn't make it true.
Greg, retired 8/10.

Lumpr
Posts: 313
Joined: Tue May 19, 2009 2:23 pm

Re: Article that highest risk is not highest reward

Post by Lumpr » Fri Apr 27, 2012 5:39 pm

This is not new. There are a number of published academic papers from as early 1970s have come to similar conclusions.

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Fri Apr 27, 2012 7:14 pm

GregLee wrote:
Kevin M wrote:OK, maybe "axiom" isn't the best word, but I think it fits.
I agree that it is an appropriate term. It's just that, in this context, you seemed to be inferring that it's true, while "axiom" means that it's assumed. Just because people regard it as true, that doesn't make it true.
You said "axiom" means "taken to be true without evidence". That's not what I meant. I think "established principle" (per Wikipedia) captures the intended meaning better. An established principle is widely believed to be true, so not only am I inferring that it's true, I'm stating flat out that it is true--as much as any established principle can be true. But I don't think there's a lack of evidence supporting it, not to mention common sense.

You don't think it's true that investors expect higher returns for taking on more risk? That's just basic finance or investing 101. If you're going to challenge that, we probably have no common ground on which to discuss investing.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: Article that highest risk is not highest reward

Post by GregLee » Fri Apr 27, 2012 7:58 pm

Kevin M wrote: ... If you're going to challenge that, we probably have no common ground on which to discuss investing.
You couldn't have made my point more clearly. That is what axioms are like. If I don't accept your axiom, you won't give evidence to support it -- instead you'll just stop talking to me. Because axioms define the boundaries of acceptable discourse.
Greg, retired 8/10.

stlutz
Posts: 4847
Joined: Fri Jan 02, 2009 1:08 am

Re: Article that highest risk is not highest reward

Post by stlutz » Fri Apr 27, 2012 8:52 pm

This thread is kind of amusing to me as a number of the statements made here mirror concerns I've expressed about the conventional wisdom around here in other threads.
The study found that the most volatile stocks did not have the highest returns, not that the highest risk stocks don't have the highest returns. Since volatility is not a very good proxy for risk, this should not be a surprise.

See, for example, the Fama French three factor model and many other models developed in the past 50 years, all of which do a better job of identifying factors associated with higher returns.
Note what's going on here. The way to determine what investments are riskier is to look at which ones performed the best in the past. Really--is that the best way to define risk?
may be data mining.
Not really. Studies on this topic go back to the 70s and consistently show that volatility and return are unrelated at best. Fama and French noted the same in their papers as well--that is why they went off in search of other risk factors--because higher risk did not equal higher return when looking at volatility.

You don't think it's true that investors expect higher returns for taking on more risk? That's just basic finance or investing 101.
I don't think so. For example, in 1999 investors expected very high returns from the stock market and thought it wasn't that risky. In 2009, they expected low returns and and thought stocks were very risky.
Uncertainty of expected returns (not price volatility) is a reasonable and widely accepted proxy for risk. Ex post realized returns often do not match ex ante expected returns. If they did, there would be no risk.
I agree. However, uncertainty makes a stock more volatile; not less.
The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed.
I can't retire off of expected returns; I can only retire with actual returns. The question that low. vol. advocates are asking is, what is the relationship between risk and *actual* returns? If you aren't likely to get the returns you "expect" when you take risks, why take them?

However, to take it further, the approach that higher expected returns go with higher risk is somewhat self-contradictory. The "expected return" is of course a probability-weighted average of the possible outcomes. If investment A has a higher expected return than investment B, that means that investment A has *significantly* higher upside potential and *slightly* higher downside potential than B. That is the only way to move the average up. Is option A really riskier than option B? It seems to me that if the investment is really more risky, then the upside and downside should be expanding pretty much in tandem, which would mean that the mean expected return would stay the same.

An axiomatic statement would be that riskier investments offer higher *potential* return, not higher *expected* return.

If I don't accept your axiom, you won't give evidence to support it -- instead you'll just stop talking to me. Because axioms define the boundaries of acceptable discourse.
Greg--I got exactly the same response you did (further discourse isn't worthwhile) from another forum member when I questioned this axiom in another thread. Somewhere this thing got a little over-ideological. :)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Fri Apr 27, 2012 9:36 pm

GregLee wrote:
Kevin M wrote: ... If you're going to challenge that, we probably have no common ground on which to discuss investing.
You couldn't have made my point more clearly. That is what axioms are like. If I don't accept your axiom, you won't give evidence to support it -- instead you'll just stop talking to me. Because axioms define the boundaries of acceptable discourse.
I didn't say I wouldn't give you evidence. I just meant that this is such a basic financial concept upon which most finance theory is based that if you don't accept it, then it's difficult, if not impossible, to discuss finance theory with you. It's like trying to discuss math without accepting that 2 + 2 = 4. Just read a basic textbook on finance or portfolio theory. That's my evidence. I have three textbooks on investing and one on finance. They all agree on this point.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Fri Apr 27, 2012 10:23 pm

stlutz wrote:Not really. Studies on this topic go back to the 70s and consistently show that volatility and return are unrelated at best. Fama and French noted the same in their papers as well--that is why they went off in search of other risk factors--because higher risk did not equal higher return when looking at volatility.
Huh? You are saying that T-bills aren't less volatile than longer term bonds? That bonds aren't less volatile than stocks? That large-cap stocks aren't less volatile that small-cap stocks? And that each of these is not respectively riskier?

You really can find statements from either Fama or French stating that investors don't expect higher returns from higher-risk investments? The FF 3F model adds two factors (I don't believe they refer to them as risk factors in their paper, but I could be wrong) to the CAPM factor, beta. Using three factors provided a better regression analysis fit to the data than the single factor, beta. I don't understand how you think this shows that volatility and return are unrelated.

You don't think it's true that investors expect higher returns for taking on more risk? That's just basic finance or investing 101.
stlutz wrote:I don't think so. For example, in 1999 investors expected very high returns from the stock market and thought it wasn't that risky. In 2009, they expected low returns and and thought stocks were very risky.
I didn't say investors were always rational. Again, read a basic textbook on investing and finance, and you will see that what I am saying is indeed investing 101. I am not making this up. I am just telling you what the textbooks say. I can look it up and write down some quotes from the textbooks if it will help. Whether or not you believe basic finance theory is another topic.

People use the same type of argument to "prove" that markets are not efficient. I saw a video of Fama responding to this very point. His answer, IIRC, was that if markets were not efficient, investors would be able to consistently profit from the inefficiency, and the evidence does not support that this is the case. Fama's words, not mine.
The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed.
stlutz wrote:I can't retire off of expected returns; I can only retire with actual returns. The question that low. vol. advocates are asking is, what is the relationship between risk and *actual* returns? If you aren't likely to get the returns you "expect" when you take risks, why take them?
There is an enormous amount of evidence that long-term historical returns are generally proportional to the uncertainty of returns (i.e., the dispersion of returns), certainly in the US, and generally in other countries. You only have to look at the data for T-bills, T-bonds and S&P 500 since the late 1920s to see this. The problem for you and me is that we don't know that the expected returns will be realized over our investing time frame. That's the nature of risk; it is real--we may not achieve the expected returns.

A little common sense should tell us that if we could reasonably expect a higher return from a lower risk investment, then we would all gravitate toward that investment. That would drive up the price and lower the expected return. That's exactly what happens in a reasonably efficient market. What are these low-volatility investments that provide high returns that are being advocated?
stlutz wrote:However, to take it further, the approach that higher expected returns go with higher risk is somewhat self-contradictory. The "expected return" is of course a probability-weighted average of the possible outcomes. If investment A has a higher expected return than investment B, that means that investment A has *significantly* higher upside potential and *slightly* higher downside potential than B. That is the only way to move the average up. Is option A really riskier than option B? It seems to me that if the investment is really more risky, then the upside and downside should be expanding pretty much in tandem, which would mean that the mean expected return would stay the same.
A little more study of a good, basic textbook might help correct your misunderstanding. I can see how you are thinking, but it doesn't work that way. The part you're getting wrong is the "slightly higher downside potential". The problem is you are only thinking about the expected returns, not the dispersion of returns. The average can move up a little, but the dispersion of returns can be much broader--that's where the risk comes in.

It's a work in progress, but if you read the first few sections of the Wiki article, Risk and return, it might help clarify these points for you. The info in this article is mostly sourced from textbooks. Look at Figure 1, and it should help clear up your misunderstanding.
stlutz wrote:An axiomatic statement would be that riskier investments offer higher *potential* return, not higher *expected* return.
You can make up your own "axiomatic" statements, or you can use the ones that are widely accepted in finance theory. Let's not get hung up on the word "axiom". One of the most fundamental tenets (do we like that word any better?) of finance theory is that risk and expected return are related. Again, I'm not making this up or telling you what I believe. I'm just telling you what the textbooks say.
If I don't accept your axiom, you won't give evidence to support it -- instead you'll just stop talking to me. Because axioms define the boundaries of acceptable discourse.
stlutz wrote:Greg--I got exactly the same response you did (further discourse isn't worthwhile) from another forum member when I questioned this axiom in another thread. Somewhere this thing got a little over-ideological. :)
I tried to address this above. It's not ideology--unless you define finance theory as ideology. Notice how much text I'm devoting to try and clarify, so it's clearly worthwhile as long as I think there's a chance we can come to some common understanding. If it becomes clear that we can't agree on the basic foundations of investing theory, then it's difficult to have much of a conversation, no?

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Article that highest risk is not highest reward

Post by umfundi » Fri Apr 27, 2012 10:47 pm

Kevin M wrote:The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed. Uncertainty of expected returns (not price volatility) is a reasonable and widely accepted proxy for risk. Ex post realized returns often do not match ex ante expected returns. If they did, there would be no risk, and therefore no risk premium. It must be so.

If volatility, whether of price or returns, could be relied upon to predict long-term returns, then rational investors with long investment horizons would simply invest in high volatility investments. This would drive the price of these investments up, reducing their expected returns, so the long-term returns would no longer be proportional to volatility. High volatility investments must disappoint from time to time, otherwise there would be no risk, and the volatility would decline.

Conclusions cannot be drawn from a single twenty year period. If they could, then one could easily conclude that bonds provide a higher return than stocks.

Kevin
Kevin pretty much summarizes what I have been trying to articulate for myself.

In a perfectly rational world, higher return for higher risk would balance out, so actual return would not depend on risk. (For example, zero risk return is 2%. High risk return is potentially 7%, but due to defaults is actually 2%.)

But, investors are risk-averse. So, higher risk does provide higher actual returns.

Keith
Déjà Vu is not a prediction

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Article that highest risk is not highest reward

Post by 555 » Fri Apr 27, 2012 10:49 pm

Kevin M wrote:
GregLee wrote:
Kevin M wrote:`... If you're going to challenge that, we probably have no common ground on which to discuss investing.'
`You couldn't have made my point more clearly. That is what axioms are like. If I don't accept your axiom, you won't give evidence to support it -- instead you'll just stop talking to me. Because axioms define the boundaries of acceptable discourse.'
`I didn't say I wouldn't give you evidence. I just meant that this is such a basic financial concept upon which most finance theory is based that if you don't accept it, then it's difficult, if not impossible, to discuss finance theory with you. It's like trying to discuss math without accepting that 2 + 2 = 4. Just read a basic textbook on finance or portfolio theory. That's my evidence. I have three textbooks on investing and one on finance. They all agree on this point.'
What exactly is this `point' you refer to. Could you please give a precise statement of it?

getRichSlower
Posts: 274
Joined: Tue Aug 16, 2011 6:40 pm

Re: Article that highest risk is not highest reward

Post by getRichSlower » Fri Apr 27, 2012 11:22 pm

Kevin M wrote: I didn't say I wouldn't give you evidence. I just meant that this is such a basic financial concept upon which most finance theory is based that if you don't accept it, then it's difficult, if not impossible, to discuss finance theory with you. It's like trying to discuss math without accepting that 2 + 2 = 4. Just read a basic textbook on finance or portfolio theory. That's my evidence. I have three textbooks on investing and one on finance. They all agree on this point.

Kevin
You're greatly simplifying the academic consensus, to the point of being misleading. Here's a quote from Dr. Delong, a highly respected economist at UCB:
"Our models predict that in normal times, with the ability to diversify portfolios that exists today, the risk discount on assets like corporate equities should be around 1% per year. It is more like 5% per year in normal times...

james22
Posts: 1223
Joined: Tue Aug 21, 2007 2:22 pm

Re: Article that highest risk is not highest reward

Post by james22 » Fri Apr 27, 2012 11:55 pm

One of the interesting points that both Warren Buffett and Howard Marks have stressed over the years is that risk - viewed as the risk of losing significant amounts of money - moves in the same direction as valuations. So as valuations become rich, risk increases, and as valuations become depressed, risk declines. At the same time, rich valuations imply weak long-term prospective returns, while depressed valuations imply strong long-term prospective returns. As a result, both Marks and Buffett suggest that risk is lowest precisely when prospective returns are the highest, and risk is highest precisely when prospective returns are the worst.

http://www.bogleheads.org/forum/viewtop ... 10&t=75211
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

stlutz
Posts: 4847
Joined: Fri Jan 02, 2009 1:08 am

Re: Article that highest risk is not highest reward

Post by stlutz » Sat Apr 28, 2012 12:32 am

Kevin--Thanks for the comments. The article that the OP referenced and the studies that I was referring to were comparing the actual returns of high volatility stocks vs. low volatility stocks, not stocks vs. T-bills. That has been where most such research has been.
You really can find statements from either Fama or French stating that investors don't expect higher returns from higher-risk investments
Not what I said at all. Indeed, Fama/French say that higher returns definitely come from higher risk investments. What I indicated what was their research showed that higher volatility stocks did not provide better long-term returns than low-volatility stocks--the CAPM failed. That is why they went in search of other ways to define risk. I don't think we disagree all that much on this. Where I disagree with the predominant view here is that I think the failure of CAPM should have equally provoked a reexamination about the assumptions around the risk/return relationship. That second look has only been happening very recently.
One of the most fundamental tenets (do we like that word any better?) of finance theory is that risk and expected return are related. Again, I'm not making this up or telling you what I believe. I'm just telling you what the textbooks say.
Again, we agree on what most textbooks say. I'm questioning whether the textbook position is logical/borne out by reality.
A little more study of a good, basic textbook might help correct your misunderstanding. I can see how you are thinking, but it doesn't work that way. The part you're getting wrong is the "slightly higher downside potential". The problem is you are only thinking about the expected returns, not the dispersion of returns.
Expected returns takes into account return dispersion; otherwise they are optimistic returns. My point is not theoretical; it's mathematical. The only way to move an average from 10 to 12 is to have the positive changes in the set be bigger than than the negative ones.

When someone says that investment A has a dramatically higher expected return than investment B (e.g "stocks have an expected return premium of 6% per year"), the "risk" has a wink, wink character to it. The promised payoffs are simply far and above the downside potential. If the premium is that big, that would be an indication to me of market inefficiency, not market efficiency, as people would be dumb not to take that risk. The view that avoiding such risk is dumb is, for example, why Vanguard has such a high stock allocation in their Target Retirement funds. If markets are efficient/rational, I don't think the equity risk premium would be that high. But of course, rationality is in the eye of the beholder, so people can disagree as to what is rational.

To go back to your question of stocks vs. bonds, the average investor has not actually earned much of a risk premium from stocks. As we talk about here all of the time, much of the equity premium is lost due to costs and adverse market timing (buying high and selling low). The average 401K investor underperforms the S&P by 6%/yr. The so-called equity risk premium is only available to those who do it correctly, by either buying/selling the right stocks at the right times (Warren Buffett), or by low cost, low turnover, non-market-timing investing (Bogleheadism). So, it's not a risk premium, but one you can only access by exhibiting the right behaviors. (And, if markets are efficient, then if everyone becomes a Boglehead, the premium should go down).
But, investors are risk-averse. So, higher risk does provide higher actual returns.
Keith--Not if you are looking at investors' actual perceptions of risk. Again, stocks were considered fairly low risk in 1999, and high risk in 2009. Right now "dividend stocks" are widely considered by a lot of smart people to be relatively low risk alternatives to the paltry yields on bonds. After the run-up that has happened in these securities as a result, I doubt they will deliver the expected returns in the future nor will they be as low risk as investors consider them to be. (i.e. I "expect" such stocks to be high risk/low return going forward). These should not be terribly controversial viewpoints around here.

Tuxx
Posts: 863
Joined: Tue Mar 30, 2010 1:19 am

Re: Article that highest risk is not highest reward

Post by Tuxx » Sat Apr 28, 2012 12:35 am

This article remained me of the Microcap Craze.

I have not heard the term "Microcap" in at least a decade. Amazing how quickly the fads come and go.

User avatar
grayfox
Posts: 4939
Joined: Sat Sep 15, 2007 4:30 am

Re: Article that highest risk is not highest reward

Post by grayfox » Sat Apr 28, 2012 2:21 am

stlutz wrote: However, to take it further, the approach that higher expected returns go with higher risk is somewhat self-contradictory. The "expected return" is of course a probability-weighted average of the possible outcomes. If investment A has a higher expected return than investment B, that means that investment A has *significantly* higher upside potential and *slightly* higher downside potential than B. That is the only way to move the average up. Is option A really riskier than option B? It seems to me that if the investment is really more risky, then the upside and downside should be expanding pretty much in tandem, which would mean that the mean expected return would stay the same.
Suppose there were two investments A and B. If you purchase A today, in one year you will get $105, guaranteed. No uncertianty and no risk.

If you purchase B today, in one year you get either $100 or $110, with equal probability. (The expected payoff is 0.5*100+0.5*110 = $105) There is uncertainty with consequences, so there is risk.

The current market price for A is $100, so its expected return is 5%

How much should you pay for investment B? You seem to be saying that you would pay $100 which would have the same 5% expected return as A.

My answer:
Not many investors would choose B if they could get the same expected return with A with no risk. Extremely risk-averse investors would choose A and never B. Others might choose B, but only if there is a discount. Some might choose B if there was a 20% discount, so they would offer $80. Others might feel that a 10% discount is acceptable so they offer $90. Some might even accept only 5% discount, and pay $95. There may even be a few gamblers that would pay $100 just for the thrill. (Some might even pay $105, for zero expected return, or even more and accept a negative expected return.)

Eventually the market for A and B well sort it out and come up with a equilibrium price. Normally B will be trading at a significant discount to A. B will have a higher expected return than A, so there will be a risk premium. The exact amount of the risk premium depends on the risk preferences of all of the investors.
Last edited by grayfox on Sat Apr 28, 2012 2:33 am, edited 4 times in total.

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 2:27 am

555 wrote: What exactly is this `point' you refer to. Could you please give a precise statement of it?
Sure, I'll give you a quote from one textbook: Essentials of Investments by Bodie, Kane and Marcus (2008), p. 11 (note how early in the text this is explained--page 11):
If you want higher expected returns, you will have to pay a price in terms of accepting higher investment risk ... We conclude that there should be a risk-return tradeoff in the securities markets, with higher-risk assets priced to offer higher expected returns than lower-risk assets.
The text represented by the ... is a paragraph that explains basically what I explained above about lower-risk assets with higher returns attracting buyers that drive the prices up, thus lowering their expected return (and vice versa).

This is representative of the principle that is explained in all investment and finance texts I have read, and in all investment books I've read (by authors such as Bernstein, Swedroe, and many others). This is just such a basic concept that it's hard for me to fathom why people here are challenging it.

Kevin
Last edited by Kevin M on Sat Apr 28, 2012 3:20 am, edited 1 time in total.
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 2:34 am

getRichSlower wrote: You're greatly simplifying the academic consensus, to the point of being misleading. Here's a quote from Dr. Delong, a highly respected economist at UCB:
"Our models predict that in normal times, with the ability to diversify portfolios that exists today, the risk discount on assets like corporate equities should be around 1% per year. It is more like 5% per year in normal times...
I'm totally baffled by your statement. I don't see how a quote from one academic can represent academic consensus, and I'm not even sure what the point of the quote is. I am simply summarizing a very fundamental principle of the trade-off between risk and return that is stated in every credible investment book I have ever read, whether textbook or "retail investor" book. Absolutely baffled. :confused

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 3:18 am

stlutz wrote:Not what I said at all. Indeed, Fama/French say that higher returns definitely come from higher risk investments.

Good. So, we have some common ground, and we can have a discussion about the fine points.
stlutz wrote:What I indicated what was their research showed that higher volatility stocks did not provide better long-term returns than low-volatility stocks--the CAPM failed.

OK, we agree that their research presented a challenge to the validity of CAPM when looking at the empirical evidence. I think the issue here is conflating the concept of volatility with the concept of using one or more factors in a regression analysis to explain the cross-sectional variation of returns. Whether you use one factor or three factors, you are still doing a regression analysis of a security or portfolio of securities against one or more benchmark portfolios. In CAPM, the benchmark portfolio is the market portfolio (as represented by some stock market index, like S&P 500 or total US stock market). In the 3F model, three benchmark portfolios are used: market, SmB, and HmL. You seem to be equating beta with volatility, but it's just the regression coefficient in the CAPM model. In the 3F model, there are three regression coefficients instead of one. How does this support your conclusion about volatility?

So, I don't think the way you state the results of their research is accurate. Do you agree that small-cap stocks are generally more volatile than large-cap stocks? One of the factors is the regression coefficient with small-cap stocks (minus large-cap stocks). Think about how that can possibly support your statement.
stlutz wrote:Again, we agree on what most textbooks say. I'm questioning whether the textbook position is logical/borne out by reality.
Good again, some common ground. Don't mean to be condescending, but have you actually read one of the textbooks? They actually use historical data (reality), as well as logic, to demonstrate the risk/return relationship.
A little more study of a good, basic textbook might help correct your misunderstanding. I can see how you are thinking, but it doesn't work that way. The part you're getting wrong is the "slightly higher downside potential". The problem is you are only thinking about the expected returns, not the dispersion of returns.
stlutz wrote:Expected returns takes into account return dispersion; otherwise they are optimistic returns. My point is not theoretical; it's mathematical. The only way to move an average from 10 to 12 is to have the positive changes in the set be bigger than than the negative ones.
Your first statement is not correct; expected return and return dispersion are distinct characteristics of a frequency distribution of returns. You can have the same expected return for two investments but much different dispersion of returns. "Expected return" is just a fancy term for the (weighted) "mean" or "average" return, whether speaking of historical or estimated future returns. Dispersion of returns describes the variation in the returns (how much the returns are "spread out"), and typically is measured by variance or standard deviation.

I would much rather invest in something with an expected return of 5% and a standard deviation of 0 than an investment with an expected return of 5% and a standard deviation of 3 pp (percentage points). Similarly, I would definitely prefer an investment with an expected return of 6% rather than one with an expected return of 5% if they both had the same risk (uncertainty or dispersion of returns). This is the basic tenet of risk aversion. I also would rather invest in an investment with a guaranteed return of 5% than an investment with an expected return of 7% and a standard deviation of 4 pp. This is where the difference in individual investors' risk aversion comes into play. Each investor has a unique preference for accepting a higher expected return given some higher uncertainty of return.

I feel we still have enough common ground for rational discourse. How about you?

Did you take a look at the Wiki article I referenced? I would be interested if you find it understandable, since you seem to have a level of understanding that it is targeted for. Since it is all based on credible sources (i.e., I did not make it up), I would welcome additional inputs based on credible sources that could enhance it. Later sections could get into some of the challenges to the theories (which the textbooks do include--just haven't gotten that far yet).

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Article that highest risk is not highest reward

Post by larryswedroe » Sat Apr 28, 2012 8:34 am

FWIW
Spoke with a leading researcher on this issue other day, though in limited light. Was related to small value. Once you used multiple screens for value, not just BtM, screening for low beta added no value at all
Larry

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Article that highest risk is not highest reward

Post by 555 » Sat Apr 28, 2012 9:35 am

@Kevin M Read this.
http://www.efficientfrontier.com/ef/499/inept.htm
If you go into a casino there are plenty of high risk low return opportunities. At least in the casino, for most choices, the risks and returns are well understood in advance, which can't be said for many decisions elsewhere. I doubt anyone here misunderstands the basics of risk and return, but there are many subtleties.

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: Article that highest risk is not highest reward

Post by GregLee » Sat Apr 28, 2012 10:12 am

stlutz wrote:The only way to move an average from 10 to 12 is to have the positive changes in the set be bigger than than the negative ones.

When someone says that investment A has a dramatically higher expected return than investment B (e.g "stocks have an expected return premium of 6% per year"), the "risk" has a wink, wink character to it. The promised payoffs are simply far and above the downside potential.
Careful about these terms "positive", "negative", "downside". You are using them to mean greater than or less than zero, but in this context, they would often be interpreted to mean greater than or less than the expected value.
Greg, retired 8/10.

User avatar
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Re: Article that highest risk is not highest reward

Post by Lbill » Sat Apr 28, 2012 10:27 am

larryswedroe wrote:FWIW
Spoke with a leading researcher on this issue other day, though in limited light. Was related to small value. Once you used multiple screens for value, not just BtM, screening for low beta added no value at all
Larry
What about the reverse?
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: Article that highest risk is not highest reward

Post by GregLee » Sat Apr 28, 2012 10:43 am

Kevin M wrote:I would much rather invest in something with an expected return of 5% and a standard deviation of 0 than an investment with an expected return of 5% and a standard deviation of 3 pp (percentage points).
I don't doubt that you would, but is that a rational choice? After all, you expect the same profit for the two investments, but for the second one, you have a better chance of making more than you expect. And it gets more difficult when we consider a more realistic choice between an investment with a slightly lower expected return and one with much greater risk. I find it easier to believe that people will choose the lower expected return than I find it to understand why this is regarded as rational investor behavior.
Last edited by GregLee on Sat Apr 28, 2012 12:40 pm, edited 1 time in total.
Greg, retired 8/10.

User avatar
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Re: Article that highest risk is not highest reward

Post by Lbill » Sat Apr 28, 2012 11:49 am

Not exactly a new finding that risk (defined as market risk, or beta) is not related to returns. Fama & French found this in 1992:
Our tests do not support the central production of the [CAPM], that average stock returns are positively related to beta.
http://www.bengrahaminvesting.ca/Resear ... eturns.pdf

This was discussed by Bill Bernstein in 2002:
Ten years ago this month, Eugene Fama and Kenneth French fired the shot heard ’round the world. Its echoes still plainly reverberate today in boardrooms and trading floors.
{snip]
The corollary of their work was that once one considered the size and value factor "loadings" of a diversified U.S. all-stock portfolio, the market loading—Sharpe’s famous "beta"—did not explain return. In other words, portfolios of high-beta stocks did not have higher returns than portfolios of low-beta stocks. Beta was dead.
http://www.efficientfrontier.com/ef/702/3FM-10.htm
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
NoRoboGuy
Posts: 732
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Re: Article that highest risk is not highest reward

Post by NoRoboGuy » Sat Apr 28, 2012 12:32 pm

I am not seeing the fundamental argument. Risk is transitory. For example, most would agree Bank of America and AIG were regarded as a very low volatility stocks...until they weren't.
There is no free lunch.

stlutz
Posts: 4847
Joined: Fri Jan 02, 2009 1:08 am

Re: Article that highest risk is not highest reward

Post by stlutz » Sat Apr 28, 2012 2:48 pm

I think the issue here is conflating the concept of volatility with the concept of using one or more factors in a regression analysis to explain the cross-sectional variation of returns.
No, my complaint on other threads is finding what performed the best and deciding that those factors are therefore the best measurements of risk. Remember, a "risk factor" is not the the same as "risk".

People often say that volatility is not a good, or at least a very incomplete way, of measuring risk. Fair enough. However, nobody here has ever pointed me to the paper where the researchers came up with a list of say, the 10 best measurements of a the riskiness of a stock, score all companies based on those measurements, and report the results. One can pick up bits and pieces here and there (e.g. companies with high ratios of debt to equity tend to outperform those with low ratios, which would support the standard view. On the other hand, companies in financial distress tend to underperform, which undermines the standard view), but never a real comprehensive model of the riskiness of a stock. I'm sure plenty of practitioners have done such research--the fact that we haven't heard anything about it suggests to me that the results don't really provide a good way to get market-beating returns.

The standard approach to economics and financial theory has been to assume that people are rational decision makers--"rational" being defined that way that the economists viewed rational. Those assumptions have been widely critiqued in areas beyond investing. Such material is worth checking out, and unlike a textbook (which yes, I have read), is actually interesting. :D

(Kevin--side note: I have read the wiki entry and thought it was a good explanation of the standard view. I actually posted my comments in the thread you had on it).

stlutz
Posts: 4847
Joined: Fri Jan 02, 2009 1:08 am

Re: Article that highest risk is not highest reward

Post by stlutz » Sat Apr 28, 2012 3:04 pm

If you follow this forum a lot, in practice people here seem to actually be skeptical of the view that loading up on risk adds more expected (as opposed to potential) return. When someone asks for help and says something like, "I'm 45 and would like to be 90% stocks", people here generally find a nice way to say that such an approach isn't smart. Yobria who is a strong supporter of efficient markets says that one can expect small stocks to outperform big by about 1% per year. Seems pretty reasonable even though I disagree with him on many theoretical considerations. Larry Swedroe has argued very persuasively that taking any credit risk with bonds (i.e. holding anything but Treasuries) both adds risk and subtracts return, when looking at one's portfolio as a whole.

While I know it's not a popular view, I'd suggest that our theory should follow our practice--be cautious about what risk can do for you. It probably won't make up for not saving enough. It probably won't make up for spending too much in retirement. It won't make up for your bad market timing calls. It won't make up for those high expenses in trading stocks or buying actively manged funds.

stlutz
Posts: 4847
Joined: Fri Jan 02, 2009 1:08 am

Re: Article that highest risk is not highest reward

Post by stlutz » Sat Apr 28, 2012 3:08 pm

Careful about these terms "positive", "negative", "downside". You are using them to mean greater than or less than zero, but in this context, they would often be interpreted to mean greater than or less than the expected value.
Agreed--I have been seeking to oversimplify to make a point w/o getting bogged down in math and to keep the thread readable. :happy

getRichSlower
Posts: 274
Joined: Tue Aug 16, 2011 6:40 pm

Re: Article that highest risk is not highest reward

Post by getRichSlower » Sat Apr 28, 2012 3:18 pm

Kevin M wrote: I'm totally baffled by your statement. I don't see how a quote from one academic can represent academic consensus, and I'm not even sure what the point of the quote is. I am simply summarizing a very fundamental principle of the trade-off between risk and return that is stated in every credible investment book I have ever read, whether textbook or "retail investor" book. Absolutely baffled. :confused
Kevin
If your point is merely that most people will only invest in assets with higher perceived risk if those assets have higher perceived expected values, then your statement is non-controversial, but also next to useless for making investment decisions. If you believe that an equity risk premium of 5%, 3% or even 2% is some type of economic axiom, then that is relevant to investment decisions, but is very controversial.

allsop
Posts: 1046
Joined: Sun Jun 15, 2008 7:08 am

Re: Article that highest risk is not highest reward

Post by allsop » Sat Apr 28, 2012 3:49 pm

An offer of high return with no risk is a very good sign of attempted fraud, for a reason.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Article that highest risk is not highest reward

Post by yobria » Sat Apr 28, 2012 4:56 pm

stlutz wrote:If you follow this forum a lot, in practice people here seem to actually be skeptical of the view that loading up on risk adds more expected (as opposed to potential) return. When someone asks for help and says something like, "I'm 45 and would like to be 90% stocks", people here generally find a nice way to say that such an approach isn't smart. Yobria who is a strong supporter of efficient markets says that one can expect small stocks to outperform big by about 1% per year.
Just to clarify my views, I believe that investors are rational, and so demand more return for more risk taken. And I like volatility as a proxy for risk, since asset price movements reflect all realized risks, from earnings misses to unexpected competition to corporate fraud. Since the future will look different from the past anyway, more refined measures of risk are probably a waste of time, and easily abused to sell something or make a certain point. I expect small caps to return about 1% over large (though perhaps not at today's valuations) due to their higher historical volatility. But there's no free lunch - if the small cap risk shows up over your time horizon, you'll wish you'd never overwieghted them. If you tilt, tilt gently.

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 5:18 pm

getRichSlower wrote: If your point is merely that most people will only invest in assets with higher perceived risk if those assets have higher perceived expected values, then your statement is non-controversial
Yep, that's pretty much all I ever stated.
Kevin M wrote:The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed.
getRichSlower wrote:, but also next to useless for making investment decisions.
Interesting point of view. I disagree. I've talked with many novice investors who do not understand this. One of most important things I try to get them to understand is that they cannot expect higher returns without taking more risk. It is not uncommon for us to see statements on this forum like, "I want to make 8% per year, and I don't want to take much risk".

getRichSlower wrote:If you believe that an equity risk premium of 5%, 3% or even 2% is some type of economic axiom, then that is relevant to investment decisions, but is very controversial.
I never said anything like that. How did you get that impression? Risk premiums are variable and difficult if not impossible to determine with any precision. I'm just talking about a general principle (which we appear to agree on). :sharebeer

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 5:23 pm

GregLee wrote:
Kevin M wrote:I would much rather invest in something with an expected return of 5% and a standard deviation of 0 than an investment with an expected return of 5% and a standard deviation of 3 pp (percentage points).
I don't doubt that you would, but is that a rational choice? After all, you expect the same profit for the two investments, but for the second one, you have a better chance of making more than you expect.
For me it is a rational choice. It makes more sense if you quote the entire paragraph instead of just that one line:
I would much rather invest in something with an expected return of 5% and a standard deviation of 0 than an investment with an expected return of 5% and a standard deviation of 3 pp (percentage points). Similarly, I would definitely prefer an investment with an expected return of 6% rather than one with an expected return of 5% if they both had the same risk (uncertainty or dispersion of returns). This is the basic tenet of risk aversion. I also would rather invest in an investment with a guaranteed return of 5% than an investment with an expected return of 7% and a standard deviation of 4 pp. This is where the difference in individual investors' risk aversion comes into play. Each investor has a unique preference for accepting a higher expected return given some higher uncertainty of return.
(Bolded the key points to address your comments).

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 5:34 pm

OK, read it. Agree with everything in it. What is the very first sentence of the article?
The interplay between return and risk is the heart and soul of the financial markets. You want high returns? Fine, prepare to have your bells seriously rung every now and then. You want safety of capital? OK, but forget about retiring to La Jolla or Provence.
Other than Bill's witty writing style, how is this different than my original statement that for some reason seems to have generated so much controversy (much to my surprise)?
Kevin M wrote:The fundamental axiom of investing that investors demand higher expected returns from investments that are perceived to have higher risk has not been repealed. Uncertainty of expected returns (not price volatility) is a reasonable and widely accepted proxy for risk. Ex post realized returns often do not match ex ante expected returns. If they did, there would be no risk, and therefore no risk premium. It must be so.
Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10300
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Article that highest risk is not highest reward

Post by Kevin M » Sat Apr 28, 2012 5:55 pm

I hadn't even read the article yet, but just did. So how many people here are actually planning on putting a majority of their portfolio in the new ETFs designed to take advantage of this apparent anomaly?
LAUNCH OF ETFS

From its launch May 27 through last Tuesday, the Russell 1000 Low-Volatility ETF (LVOL) gained 5%, while its counterpart, the Russell High-Volatility ETF (HVOL), declined by 4%.

The Russell 2000 Low-Volatility ETF (SLVY), over the same period, gained 2.6%, while its counterpart, the Russell 2000 High-Volatility ETF (SHVY), fell by 15.3%.
I suspect most of us are either going to stick with a total market approach (beta = 1, by definition, with no loading on the other two factors, SmB or HmL), or tilt to small and value based on the FF research (load on the other two factors). And I suspect most of us will continue to hold some fixed income because it is less risky than stocks (lower dispersion of returns, lower expected return).

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Post Reply