Skip navigation

The myth of asset allocation

or Register to post new content in the forum

66 RepliesJump to last post

 

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Jan 5, 2006 4:12 pm

Those performance numbers I posted were for 5 years ending in 2001. The significance of this was to capture the dot.com bubble in the returns.  I thought it was curious that these "value" funds were basically in line with the "growth" funds, and that the govi fund did as well as the small cap fund. What a hoot!  

Again further evidence supporting my previous posts that I've
noticed that during the same market periods some "value" managers
outperform "growth" managers and vice versa.

One could conclude then that this argument of which investment style is better or in favor or out of favor is bogus.  If an investment style is in fact in vogue, then the other style can't do as well. The words value, GARP and growth have become branding words. 

Jamming clients into pre-determined allocation models in order to simplify the process and maximize efficiency is not appealing to me. And as far as "financial planning", I ask how can calculating net worth, life expectancy and time value of money have to do with investment success? 

For purposes of a financial plan, a typical analysis: 1. assumes tax bracket, 2.assumes inflation rate, 3. assumes risk-free returns, 4. assumes equity risk premium and finally 5. assumes that previous investment results will be obtained in order to determine an optimum investment blend. 

Now let me get back to work. My Nordstroms is on fire today.

Jan 5, 2006 5:16 pm

MikeButler said:

dude wrote:

We are not trying to beat an index, we are measured against how the client does relative to their goals. Fine. How about this thought:

I think this is where a lawyer would say you're assuming facts not in evidence. Also, I measure portfolios against a blended index that consists of the equivalent weightings of the sub indexes, EAFA, 400, 600, etc., not the S&P 500.

Reply:

You took this out of context, there is more directly above it that is detailing various arguments and justifications for an  asset allocated approach.  I'm not pointing any fingers at you, just addressing what I have seen to be the vast majority of variations on this theme. 

Jan 5, 2006 6:23 pm

[quote=skeedaddy2] <?:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Those performance numbers I posted were for 5 years ending in 2001. The significance of this was to capture the dot.com bubble in the returns.  I thought it was curious that these "value" funds were basically in line with the "growth" funds, and that the govi fund did as well as the small cap fund. What a hoot!  

[/quote]

Thanks for making that time period clear, and you  make a good point. Especially during that time period when value managers fled a value philosophy and started buying growth stocks and/or became stealth index funds because they got tired of getting beaten-up in a marketing environment where style didn’t matter to the buyer (and therefore didn’t matter to the seller) and every fund was compared to the S&P 500 or the hottest of the hot dot. If I heard it once then I heard it a thousand times, "let’s dump all this Warren Buffett style junk and load up on (fill in hot dot fund name here).”.

The key is to stay away from managers who are able to drift in style (as most fund managers are), a classic example would be the guy running the Legg Mason “Value” Fund that was buying Cisco at the top as a “value” stock.

[quote=skeedaddy2]

 

One could conclude then that this argument of which investment style is better or in favor or out of favor is bogus.  If an investment style is in fact in vogue, then the other style can't do as well.

[/quote]

One could, but one would be wrong .      That’s the danger of using the name of a style drifting fund to ascribe some real style definition. For a clearer case, look at the style and cap indexes, Russell or S&P to get a real picture where marketing induced style drift isn’t present. There you'll see significant differences.

[quote=skeedaddy2]

Jamming clients into pre-determined allocation models in order to simplify the process and maximize efficiency is not appealing to me. 

[/quote]

Or me, see the post with Dude on that subject.

[quote=skeedaddy2]

And as far as "financial planning", I ask how can calculating net worth, life expectancy and time value of money have to do with investment success? 

[/quote]

If you define “success”, as I do, which is did we succeed in achieving a specific client’s specific financial goals, those things are critical. If “success” is posting a number without consideration of the risk involved or what the client needed that money to do (income, wealth preservation, capital appreciation, etc..) then those things don’t matter.

[quote=skeedaddy2]

For purposes of a financial plan, a typical analysis: 1. assumes tax bracket, 2.assumes inflation rate, 3. assumes risk-free returns, 4. assumes equity risk premium and finally 5. assumes that previous investment results will be obtained in order to determine an optimum investment blend. 

[/quote]

Those are assumptions made on a reasonable basis based on reasonable criteria. While they may not constitute a promise, it’s far better to use them than to say they can’t be known to exacting specificity and therefore ignore them. It’s akin to saying flying by instrument isn’t perfect, and therefore venturing off into the night flying completely blind instead.


 

Jan 5, 2006 6:25 pm

[quote=dude]

MikeButler said:

dude wrote:

We are not trying to beat an index, we are measured against how the client does relative to their goals. Fine. How about this thought:

I think this is where a lawyer would say you're assuming facts not in evidence. Also, I measure portfolios against a blended index that consists of the equivalent weightings of the sub indexes, EAFA, 400, 600, etc., not the S&P 500.

Reply:

You took this out of context, there is more directly above it that is detailing various arguments and justifications for an  asset allocated approach.  I'm not pointing any fingers at you, just addressing what I have seen to be the vast majority of variations on this theme. 

[/quote]

I just took it as I read it, and it would seem I misunderstood your point. Could you try me again on it?

Jan 5, 2006 6:26 pm

Ohhhhh…Barf!

Jan 5, 2006 6:36 pm

[quote=dude]

MikeButler said:

dude wrote:

We are not trying to beat an index, we are measured against how the client does relative to their goals. Fine. How about this thought:

I think this is where a lawyer would say you're assuming facts not in evidence. Also, I measure portfolios against a blended index that consists of the equivalent weightings of the sub indexes, EAFA, 400, 600, etc., not the S&P 500.

Reply:

You took this out of context, there is more directly above it that is detailing various arguments and justifications for an  asset allocated approach.  I'm not pointing any fingers at you, just addressing what I have seen to be the vast majority of variations on this theme. 

[/quote]

 Well, I went back to your post and read it. I suppose my answer is the same, just more clearly worded; I simply reject the underlying assumptions of the other school of thought. Many managers DO, REGULARLY beat the index they should be compared to. It’s a matter of knowing who those managers are, how they do what they do, what market cycles favor them and don’t, and how to combine them with other managers.  Portfolios made up of those managers perform superbly and when priced fairly are well worth the fess involved. One thing to consider is why, even though many combine some passive investing, so many large pockets of smart money like endowments and trusts have always managed money in this fashion if it’s worthless.<?:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Jan 5, 2006 6:38 pm

[quote=Dirk Diggler]Ohhhhh....Barf![/quote]

You OK, Dirk? Need a hanky or a bucket?

Jan 5, 2006 7:13 pm

Dirk needs a hug.

Jan 5, 2006 7:21 pm

[quote=FIDO]Dirk needs a hug.[/quote]

I bet bankrep would give him a hug.....

Jan 5, 2006 7:25 pm

[quote=mikebutler222]

[quote=Revealer]MPT is not "modern". Was promulgated in the 50's. Go ahead and index. Still under water after past SIX years. .[/quote]

I haven't seen a single SMA portfolio built using asset allocation or MPT underwater the last six years. Most are up 10% or more on an annualized basis.

[/quote]

avg 10% per annum?  hmmmmm......I find that a little hard to believe....

Jan 5, 2006 7:37 pm

MikeB:

I was illustrating some of the justifications and/or attitudes I have heard concerning the financial planning approach, where some people say that their job is not to find managers that beat their indexes, but to help people meet their financial goals.  I say that if that is the case then they are implicitly saying that they are not adding value to the asset management aspect and should be paid a retainer for doing a plan not a % of assets annually.

In my case, to be honest, I don't believe that I have some magic formula of identifying money managers that will outperform their indices over the next 10 years. 

I can identify managers that have done it in the past (in fact I'm really good at it, with all my software and smarts ), but I don't believe that past performance (by any metric; sharpe ratio, alpha, upside/downside capture ratios, style adherence etc...) is predictive of future performance, and I believe that the vast majority of the nobel winning academics you often cite, would agree.  In fact the vast majority of research on the subject overwhelmingly supports my position, with the exception of more inefficient markets and even then I have never seen an overwhelming majority of emerging markets and small cap funds outperform their indices consistently;  usually it's pretty close to 50/50.  It would seem that the problem would be that someone who follows most allocation models would only put a small portion of the overall portfolio in those asset classes, so any outperformance would have minor effect. 

Jan 5, 2006 10:01 pm

I appreciate these types of discussions, if for no other reason, because

they keep me sharp and informed. As I’ve said before, Mike, I believe that

you do an excellent job of arguing. Just a couple more years of education

and you could have been arguing before the Supreme Court.



Now it seems that you have taken it upon yourself to be the great

advocate for financial planning and asset allocation when I know that a

great many of the participants on this forum share your sentiments and

convictions. Perhaps they lack the time or the articulateness to express

their opinion. So don’t take this as a personal attack. Remember, we’re

professionals.



I had a chance to look into this debate of growth vs. value and came up

with some additional information that might be useful. Since 1980, the

correlation between Russell 1000 Value and Russell 1000 Growth is .82.

In 100 quarterly observations, they have moved in opposite directions

only 5 times or 5% of the time. Since 1979, the average annual difference

in returns between both styles has been only 2.2%.   So all this time we’ve

been splitting hairs on 200 basis points difference.



Laurence Siegel is the guy who manages $10 billion for the Ford

Foundation, who has better information and better portfolio management

capabilities than you or me, hasn’t been able to devise a trigger system to

exploite a change in leadership.



I do agree that diversification is important, it just doesn’t exist amongst

US equities. .82 correlation is not diversification to me. I’m curious does

Warren have any investments in India or China?













Jan 5, 2006 10:03 pm

[quote=joedabrkr][quote=mikebutler222]

[quote=Revealer]MPT is not "modern". Was promulgated in the 50's. Go ahead and index. Still under water after past SIX years. .[/quote]

I haven't seen a single SMA portfolio built using asset allocation or MPT underwater the last six years. Most are up 10% or more on an annualized basis.

[/quote]

avg 10% per annum?  hmmmmm......I find that a little hard to believe....

[/quote]

<?:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Well, here's one right in front of me, selected at random. The allocation is (rounding off) 22.5% RV 1000, 22.5% RG 1000, 22.5% MSCI EAFE, 13% RG 2000, 13% RMCV. CISDM Public Fund Index 7% (for managed futures) These annualized index performance numbers end 6/05; 1yr 11.46%, 3 yr 11.47%, 5 yr 3.08%. The mangers’ performance (total portfolio), net, over the same period was,  1yr 13.65%, 3 yr 14.30%, 5 yr 11.91%.

Jan 5, 2006 10:21 pm

[quote=dude] <?:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

MikeB:

I was illustrating some of the justifications and/or attitudes I have heard .... they are implicitly saying that they are not adding value to the asset management aspect and should be paid a retainer for doing a plan not a % of assets annually.

[/quote]

You may have a point there if that's all they do. What sort of rate retainer are you talking?

[quote=dude]

In my case, to be honest, I don't believe that I have some magic formula of identifying money managers that will outperform their indices over the next 10 years. 

[/quote]

You try, sometimes you do, sometimes you don't and have to fire one. The success rates are higher than you might think, and much better, imho, than picking stocks. I have to say again, the broker that’s picking stocks is, at best, a part time player in a world of full time athletes. You’ll surely do better than the man on the street, but even if you find time away from everything else you do for a living, you can’t match the guy that sits in front of that screen all day long doing nothing but watching the 10 stocks the manager has assigned him. As an example, if something bad happens he pulls the trigger, period. You and me? We have to call the client, explain it, get his permission to sell it and face the consequences of possibly losing the account if you have to make too many of those “hey, we didn’t see it coming, but…” calls.

[quote=dude]

I can identify managers that have done it in the past (in fact I'm really good at it, with all my software and smarts <?:namespace prefix = v ns = "urn:schemas-microsoft-com:vml" />), but I don't believe that past performance (by any metric; sharpe ratio, alpha, upside/downside capture ratios, style adherence etc...) is predictive of future performance, and I believe that the vast majority of the nobel winning academics you often cite, would agree. 

[/quote]

I think you're wrong about what experts say on that. You've heard that line often about mutual funds (most can’t beat the S&P 500, blah, blah, blah), but let's not forget all the drags on mutual funds aside from the obvious fees and taxes (the in and out at the worst time behavior of owners of funds, for example). Your software will get you a long way, but results alone aren't enough. You have to dig deeper, like attribution reports that explain WHY they did well. Who knows, you might be looking at a manager who got lucky with a single high flier or two and has been dining on that ever since.

 

 

 

[quote=dude]

 

In fact the vast majority of research on the subject overwhelmingly supports my position, with the exception of more inefficient markets and even then I have never seen an overwhelming majority of emerging markets and small cap funds outperform their indices consistently;  usually it's pretty close to 50/50.  

[/quote]

No, sorry, Dude, but that’s not true. The overwhelming majority of research does not say active management can’t beat indexes (and we’re to a degree mixing apples and oranges, even if you completely agree that passive beats active investing, that doesn’t mean you have to reject asset allocation along styles and cap, it just means you’ll use indexes to do it) and when you’re talking outside the S&P 500 to the other indexes there are plenty of managers who beat them like drums year in and year out.

[quote=dude]

It would seem that the problem would be that someone who follows most allocation models would only put a small portion of the overall portfolio in those asset classes, so any outperformance would have minor effect. 

[/quote]

 

Well, I can’t speak to “most models”, but I can say I typically only use 45 to 50% in LC and the balance, including 20% plus in international and the out performance is anything but minor.

Jan 5, 2006 10:21 pm

[quote=dude] I have heard concerning the financial planning approach,

where some people say that their job is not to find managers that beat

their indexes, but to help people meet their financial goals.

[/quote]



That’s a ridiculous position to take. I’m curious, does the CPI include

prescription drugs into the calculation? I know they don’t include energy

and food.



Let’s see…I’m a retiree, my house is paid for, my cars are paid for, but my

heating bill is up 40% and my food costs are alot higher than I thought

they would be. Oh, and those association fees on that condo in Florida

are killing me. Why did my financial plan only use a 2.5% inflation rate in

its calculation?

Jan 5, 2006 11:05 pm

[quote=skeedaddy]I appreciate these types of discussions, if for no other reason, because they keep me sharp and informed. As I've said before, Mike, I believe that you do an excellent job of arguing. Just a couple more years of education and you could have been arguing before the Supreme Court. [/quote]<?:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Thanks, I think. Law school was always out of te question, my parents were married  <?:namespace prefix = v ns = "urn:schemas-microsoft-com:vml" />

[quote=skeedaddy]

 So don't take this as a personal attack. Remember, we're professionals.

[/quote]

Hmmm, where's that kevlar and flak jacket I had around here....

[quote=skeedaddy]I had a chance to look into this debate of growth vs. value and came up with some additional information that might be useful. Since 1980, the correlation between Russell 1000 Value and Russell 1000 Growth is .82.

[/quote]

That's a pretty striking divergence between categories of the same cap. Furthermore, as I'm sure you're aware, that 25 year time horizon does a great job of masking the many rotations where divergence was far, far greater. It's not much different than the long term studies (think the classic Ibottson mountain chart that shows the lines so smoothly assending and the results so similar between cap classes) that minimize differences.

[quote=skeedaddy]

that In 100 quarterly observations, they have moved in opposite directions only 5 times or 5% of the time.

[/quote]

We wouldn’t expect them to move in opposite directions very often, they’re far too similar. What is key is how far apart they are at times, even if they’re both positive or negative.

An example, in 1998 LCG hit 38.7%, LCV was at 15.68. Your 25 year time horizon may swallow up that massive difference, clients won’t. Better still, rebalancing, which isn’t present in your figures accelerates that advantage of being in both categories instead of one.

BTW, as a further example, back to 1998, SMV was at   -6.43%, SCG was at 1.24% and EAFE was at 20.34%. With rebalancing you wouldn’t have allowed that LCG to run wild after 1998 and would have ensured that some money returned to SMV (instead of allowing it to shrink more and more) so that you would have had a reasonable foothold in it in 2000 when it returned to 22.8% and your formerly high-flying LCG 22.43%.

 There's where money was saved and made.

[quote=skeedaddy]

 Since 1979, the average annual difference in returns between both styles has been only 2.2%.   So all this time we've been splitting hairs on 200 basis points difference.

[/quote]

It’s only 2.2% if you ignore every other cap and style and look over the long, long horizon.

Then again, if you measured the height of only white males, age 35,  living in <?:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" />Minnesota during a 25 year period between 1979 to today, you’d find, I’m guessing, a much smaller divergence than if you measured the height of all males round the world since time began.

[quote=skeedaddy]



Laurence Siegel is the guy who manages $10 billion for the Ford Foundation, who has better information and better portfolio management capabilities than you or me, hasn't been able to devise a trigger system to exploite a change in leadership.

[/quote]

I don’t know anyone who HAS found a way to exploit leadership changes. That would require a predictive ability to determine which cap and style was going to be in favor next (gee, sounds like stock picking).

Then again, that’s not what we’re trying to do. We’re trying to maximize return on each unit of risk taken. You can be pretty certain that Laurance, like every other manager of large endowments and foundations I’ve ever heard of, pays very close attention to cap and style diversification and doesn’t have one manager running every dollar he has.

[quote=skeedaddy]

I do agree that diversification is important, it just doesn't exist amongst US equities. .82 correlation is not diversification to me.

[/quote]

Fair enough (although .82 is a remarkable variation given the timeframe and the similarities of the two LC categories). Then again, just looking at LCG and LCV isn’t the entirety of US equities.

[quote=skeedaddy]I'm curious does Warren have any investments in India or China?

[/quote]

 

I couldn’t tell you, since the only way to hire Warren is via BRK, I don’t follow his holdings in detail.

 

However, I can tell you two things, India and China alone aren’t a reasonable representation of the totality of international markets AND there are managers using Warren’s Graham and Dodd bible that are buying in international markets.

Jan 5, 2006 11:08 pm

[quote=skeedaddy] [quote=dude] I have heard concerning the financial planning approach, where some people say that their job is not to find managers that beat their indexes, but to help people meet their financial goals. 

[/quote]

That's a ridiculous position to take. I'm curious, does the CPI include
prescription drugs into the calculation? I know they don't include energy and food.

[/quote]

The topline CPI does take food and energy into consideration, you may be thinking of the core number, which doesn't. That's why the core rate is only used to look at ultra-short time periods where the swings in fod and energy would otherwise distort the numbers).

Jan 5, 2006 11:46 pm

Mike,

My concept of the retainer fee is that there would be a one time up front fee of say $1,000 (specifics not important here) for the initial planning work and an on going annual flat fee based on services provided (like on call financial advice, education, plan updating etc... ) and time expected to work on the project.  Like how a lawyer is paid.

Now I'm not preaching this as an end all be all, only right way, dogma.  I'm merely trying to open some discussion about how to best price our services.

Mike it seems like you and I agree on more areas than we disagree when it comes to compensation guidelines.  You have to agree though that the vast majority of advisors using asset allocation are doing just as I described: Ibbotsson Strategic allocation model (or something very close), 5 risk profiles, static allocation with annual or quarterly rebalancing. 

I still don't know if I agree with you that past performance metrics (including attribution analysis) will be predictive of future results for a particular manager.  I think that there are some fundamental errors to the ways academics carve up the investment markets (value vs. growth, volatility = risk, etc....).  It is my belief that quantifying anything that is the direct result of millions (if not billions) of inputs having their basis in human emotion is akin to hearding cats, or at least coming up with mathmatical equations to describe the various dynamics of hearding cats. 

Jan 9, 2006 1:02 am

[quote=dude]

Like how a lawyer is paid.

[/quote]

Be careful what you wish for.  I have many lawyer friends, and they would all tell you that their system of compensation is what they hate most about being a lawyer. 

And they will also tell you that the highest paid attorneys are those working on a contingency fee.  The contingency fee being the closest thing they have to our system of compensation.

Jan 9, 2006 3:12 pm

You guys are starting to make my head hurt…