NASDAQ as a significant portfolio component?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 8:35 am

So I've been running some portfolio simulations using historical data going back to the early 80's, and I noticed that, for virtually *every* 10-, 15- or 20-year investment window between 1985 and 2017, *every* sane asset allocation that includes a non-trivial amount of NASDAQ (> 5%) beats every version of a portfolio that rebalances the NASDAQ portion into US or international total stock index. That includes the time windows that span both the 2000 "dot com" and 2009 real estate crises. However, NASDAQ is conspicuously missing from Bogle-advised portfolio mixes, as far as I can tell. Is that historical accident, due to the fact that low-cost NASDAQ-tracking index funds have only been available for the last decade or so (**)? Or is NASDAQ not broad enough to sufficiently diversify risk? Or is it something else entirely?

It makes sense to me that NASDAQ should have done well, and continue to do well, for two reasons: (a) it covers the technology sector, which strikes me as having much more room to grow vs. say, manufacturing; and (b) NASDAQ listings tend to be multi-nationals whose income does not depend solely (and perhaps not even in majority part) on US fortunes.

Interested to hear others' thoughts.

(**) PRSCX seems to also track similar returns to NASDAQ, but has 0.83% ER and still only goes back to 1987.

avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: NASDAQ as a significant portfolio component?

Post by avalpert » Mon Oct 23, 2017 8:59 am

Why would the exchange you happen to be listed on be a component for overweighting your investment in a security?

For what it is worth I wouldn't consider an asset allocation that include a weighting to NASDAQ over and above a broad US index a sane asset allocation so it would already fail your criteria.

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 9:30 am

avalpert wrote:
Mon Oct 23, 2017 8:59 am
I wouldn't consider an asset allocation that include a weighting to NASDAQ over and above a broad US index a sane asset allocation
So you're saying you consider NASDAQ to be unsuitably diversified, leading to higher risk than the corresponding returns would justify?

If so, the apparent fact that NASDAQ has consistently and often significantly outperformed the broader US and world markets for its entire (~30-year) existence, even during tech-specific crises, makes it harder for me to accept an out of hand dismissal of the risk/reward story without any details to justify that stance. Similar to how The Bogle Way treats US vs. world equity markets as separate asset classes, even though the US is actually a subset of the world economy, I start to wonder if tech/science (as approximated by NASDAQ, PRSCX, etc.) might have emerged as a separate asset class, even though it is technically just a subset of the US/world economy.

I wasn't able to find any analysis of NASDAQ in these forums, but would be very interested if other Bogleheads have looked closely and determined that NASDAQ has not actually enjoyed the consistent advantage my calculations have suggested exists, or have concluded that the tech/science sector is a temporary bump in the world economy that will be replaced by something else in the next 30-50 years.

For this to really work from a risk diversification perspective, you'd want a US-ex-NASDAQ index fund, along with the world-ex-US index fund, to avoid double-counting... but I suspect no US-ex-NASDAQ fund exists and would probably not make a very appealing investment even if it did: IVV and VTI top-10 holdings have very high overlap with PRSCX and ONEQ, which makes me suspect that removing all portfolio overlap would also remove much of the historic growth IVV and VTI have enjoyed, leading to lower than market returns.

User avatar
Portfolio7
Posts: 356
Joined: Tue Aug 02, 2016 3:53 am

Re: NASDAQ as a significant portfolio component?

Post by Portfolio7 » Mon Oct 23, 2017 9:37 am

A number of portfolios work around a core and satellite type approach, even if 90% of it is your core 3-fund and 10% REIT, or 10% SCV.

I don't see any strong philosophical reason not to own. Such funds are typically diversified, though to a lesser extent than many typical Index funds. I have 5% in a similar fund.

While such funds are diversified, they are just a slice of 'the market', and relatively volatile compared even to the S&P 500. I don't think I'll soon forget the dot.com crash; I don't really like the idea of concentrating more than 5% of my portfolio in such a fund. However, I agree with your assessment at least to some degree - there seems to be good potential for strong earnings there.

However, I did read a piece by the Reformed Broker this weekend, noting that some people have bought technology stocks as insurance against being automated out of a job. Seems dicey to me, but I appreciate the sentiment involved.
An investment in knowledge pays the best interest.

avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: NASDAQ as a significant portfolio component?

Post by avalpert » Mon Oct 23, 2017 9:52 am

overthought wrote:
Mon Oct 23, 2017 9:30 am
avalpert wrote:
Mon Oct 23, 2017 8:59 am
I wouldn't consider an asset allocation that include a weighting to NASDAQ over and above a broad US index a sane asset allocation
So you're saying you consider NASDAQ to be unsuitably diversified, leading to higher risk than the corresponding returns would justify?
Not really. I'm saying it is arbitrarily constructed such that it may be okay at broadly tracking the market overall (with higher risk relative to other options so unless it is the cheapest option in a 401k plan - something I have never seen - it would be hard to justify using for that purposes) but has no logical reason to be consider for overweighting.
If so, the apparent fact that NASDAQ has consistently and often significantly outperformed the broader US and world markets for its entire (~30-year) existence, even during tech-specific crises, makes it harder for me to accept an out of hand dismissal of the risk/reward story without any details to justify that stance.
I don't know that that is an apparent fact - did you do a factor analysis to see what explains the performance? What is the character of listing on NASDAQ that would lead you to believe the market prices those companies to deliver higher returns?
Similar to how The Bogle Way treats US vs. world equity markets as separate asset classes, even though the US is actually a subset of the world economy, I start to wonder if tech/science (as approximated by NASDAQ, PRSCX, etc.) might have emerged as a separate asset class, even though it is technically just a subset of the US/world economy.
NASDAQ is not limited to the tech sector - what makes you since the sector distribution of who happens to list on NASDAQ that existed in the past would persist to the future? At best, you are seeing it as an indirect way to overweight the tech sector - even if we were to think overweighting a sector may be a good idea why would you choose an indirect way that doesn't allow you to control the weighting?

JBTX
Posts: 3098
Joined: Wed Jul 26, 2017 12:46 pm

Re: NASDAQ as a significant portfolio component?

Post by JBTX » Mon Oct 23, 2017 10:11 am

I am sure you can take various sectors and slices of the broad market and some of those slices, by definition, will exceed the broad market. How is that different from overweighting small cap, or growth stocks, or any particular higher risk industry sector?

The NASDAQ is certainly higher risk and more volatile. It took much longer to reach its previous bubble highs than all the other indexes.

Over the long term it isn’t shocking to find that higher risk sectors tend to have higher returns, especially when the markets are hitting new highs.

Valuethinker
Posts: 34680
Joined: Fri May 11, 2007 11:07 am

Re: NASDAQ as a significant portfolio component?

Post by Valuethinker » Mon Oct 23, 2017 10:34 am

JBTX wrote:
Mon Oct 23, 2017 10:11 am
I am sure you can take various sectors and slices of the broad market and some of those slices, by definition, will exceed the broad market. How is that different from overweighting small cap, or growth stocks, or any particular higher risk industry sector?

The NASDAQ is certainly higher risk and more volatile. It took much longer to reach its previous bubble highs than all the other indexes.

Over the long term it isn’t shocking to find that higher risk sectors tend to have higher returns, especially when the markets are hitting new highs.
Except actually higher risk sectors do not, generally, have higher returns. Sometimes called the "low volatility anomaly". Think biotechnology or any of a host of 'ology' sectors and companies which have crashed and burned. Remember the 'lectronic bubble of the early 1960s? Any company with electronic in the name, the share price shot up (see John Brooks "the Go Go Years" and Adam Smith's "Moneygame").

In a sense what is going on here is cherrypicking. US tech stocks have done amazingly well, albeit with an almost complete turnover in leadership more than once (of IBM it was once said "we were the future, once" ;-)).

As opposed to rocketry stocks. Or biotechnology. Or new materials. Or nuclear fusion stocks (KMS Industries). Or Shipstone Corporation ;-)

http://www.technovelgy.com/ct/content.asp?Bnum=1381
Last edited by Valuethinker on Mon Oct 23, 2017 10:38 am, edited 1 time in total.

Valuethinker
Posts: 34680
Joined: Fri May 11, 2007 11:07 am

Re: NASDAQ as a significant portfolio component?

Post by Valuethinker » Mon Oct 23, 2017 10:37 am

overthought wrote:
Mon Oct 23, 2017 8:35 am
So I've been running some portfolio simulations using historical data going back to the early 80's, and I noticed that, for virtually *every* 10-, 15- or 20-year investment window between 1985 and 2017, *every* sane asset allocation that includes a non-trivial amount of NASDAQ (> 5%) beats every version of a portfolio that rebalances the NASDAQ portion into US or international total stock index. That includes the time windows that span both the 2000 "dot com" and 2009 real estate crises. However, NASDAQ is conspicuously missing from Bogle-advised portfolio mixes, as far as I can tell. Is that historical accident, due to the fact that low-cost NASDAQ-tracking index funds have only been available for the last decade or so (**)? Or is NASDAQ not broad enough to sufficiently diversify risk? Or is it something else entirely?

It makes sense to me that NASDAQ should have done well, and continue to do well, for two reasons: (a) it covers the technology sector, which strikes me as having much more room to grow vs. say, manufacturing; and (b) NASDAQ listings tend to be multi-nationals whose income does not depend solely (and perhaps not even in majority part) on US fortunes.

Interested to hear others' thoughts.

(**) PRSCX seems to also track similar returns to NASDAQ, but has 0.83% ER and still only goes back to 1987.
You could have reached the same conclusion in June of 1999. You'd have been happy -- for about 9 months.

The thing is valuations matter. History will tell us whether Amazon, Tesla, Alphabet, Netflix, Facebook lived up to expectations. Or what happened when Anti Trust authorities broke them up, or a Russian oligarch emerged as Facebook's largest shareholder ;-).

"technology" will be even bigger in the future than it is now. These internets-- you have no idea how big this is going to be. But there were something like 1000 car manufacturers in North America in 1910. Over 100 plane manufacturers in the world in the 1920s. Etc.

pkcrafter
Posts: 12641
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: NASDAQ as a significant portfolio component?

Post by pkcrafter » Mon Oct 23, 2017 10:40 am

overthought, I don't see any problem if you want to hold NASDAQ long term in some reasonable allocation. As for your backtesting, you used an unprecedented time period (the dawn of new technology) and that isn't likely to be repeated. New discoveries and new uses for tech will continue, but impact will be much less than in the 80's and 90's. The NASDAQ didn't draw much attention between 2000 and about 2013, but recent performance is attracting investors again , so I hope you haven't been seduced by that. :happy



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.

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 11:27 am

pkcrafter wrote:
Mon Oct 23, 2017 10:40 am
As for your backtesting, you used an unprecedented time period (the dawn of new technology) and that isn't likely to be repeated. New discoveries and new uses for tech will continue, but impact will be much less than in the 80's and 90's. The NASDAQ didn't draw much attention between 2000 and about 2013, but recent performance is attracting investors again , so I hope you haven't been seduced by that.
My most thorough backtesting assumed four asset classes foreign (VGSTX)/domestic (^GSPC)/nasdaq (^NDX)/bonds (SCSBX). It looked at every 10 and 15 year interval between 1 Jan 1986 and 1 Jul 2007 (or 1 Jul 2002) in six-month increments. Out of those runs, anything from 1995 or later would have spent at least half its time in the bad years, 2000-2013.

For 10-year runs:

A 30/40/0/30 allocation (fairly normal Bogle ratio I believe?) had 5.9% IRR for the run starting Jan 1995, dropped down to -1.7% IRR for the run starting Jan 1999, and has remained above 5.5% for runs starting Jan 2004 or later. Geomean for the 15 bad runs (those starting Jul 1996 through Jul 2003) was 3.24%. Geomean for all runs was 6.57%.

Meanwhile, a 30/15/25/30 allocation (so significant diversion from domestic into nasdaq) had 6.9% IRR for the run starting Jan 1995, bottomed out at -1.6% IRR for the run starting Jul 1999, and has stayed at 6.8% or better for runs starting Jan 2004 or later. Geomean for the 15 bad runs was 4% and geomean for all runs was 8.07%

For 15-year runs:

The 30/15/0/30 allocation saw 5.4% IRR for the run starting Jul 1993, while the run starting Jul 1994 hit bottom at 2%. Things improved to 5.3% IRR for the run starting Jan 1999 and have danced around 5.7% since then. Geomean for the 11 bad runs (those starting Jan 1994 through Jan 1999) was 3.44%. Geomean for all runs was 5.98%.

Meanwhile, the 30/15/25/30 allocation had 6.4% IRR for the Jul 1993 run and the Jul 1994 run hit bottom at 2.8%. Things improved to 6.1% by Jan 1999 and have danced around 6.7% since then. Geomean for the 11 bad runs was 4.37%, and geomean for all runs was 7.18%.

In both cases, the nasdaq-heavy portfolio came out on top, even when I restrict the simulations to the years that should have been its worst case.

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 11:31 am

Valuethinker wrote:
Mon Oct 23, 2017 10:37 am
You could have reached the same conclusion in June of 1999. You'd have been happy -- for about 9 months.
Our replies crossed in flight. My simulations showed Jul 1999 was *the* worst time to start a ten-year run (**), but nasdaq-heavy portfolio still beat the nasdaq-free portfolio even in the worst years (not by a lot, but ahead). In fact, the nasdaq-tilted portfolio came out ahead in *every* ten-year run, and every fifteen-year run except the one starting Jan 1994. Even in that case, the market drop just hit the nasdaq-heavy portfolio six months earlier. The nasdaq-heavy recovered six months earlier, tho: the 15-year run starting Jan 1998 had 5% IRR for nasdaq-heavy vs. just 3.6% IRR for nasdaq-free.

(**) worst start date for 15-year runs was Jul 1994.

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 11:48 am

Valuethinker wrote:
Mon Oct 23, 2017 10:34 am
In a sense what is going on here is cherrypicking. US tech stocks have done amazingly well, albeit with an almost complete turnover in leadership more than once (of IBM it was once said "we were the future, once" ;-)).

As opposed to rocketry stocks. Or biotechnology. Or new materials. Or nuclear fusion stocks (KMS Industries). Or Shipstone Corporation ;-)
The fact that tech has maintained the lead in spite of heavy turnover in leadership is a good sign IMO: it means that no one specific companies is expected to be the long-term driver of that sector's success.

Those other sectors did not have runs lasting anywhere near 30 years, that I'm aware of.
Valuethinker wrote:
Mon Oct 23, 2017 10:37 am
These internets-- you have no idea how big this is going to be. But there were something like 1000 car manufacturers in North America in 1910. Over 100 plane manufacturers in the world in the 1920s. Etc.
Number of manufacturers != size of market. But regardless, this is the "sector I think will drive economic growth for the next several decades" piece of the portfolio. I think I'd be OK with re-evaluating it every decade or two and pulling out if there are signs of long-term decline ahead. It's not like every car manufacturer simultaneously went from good times to bankrupt in a matter of months. There were pretty clear "signs of the times", spread over many years, to mark our country's exit from manufacturing.

alex_686
Posts: 3332
Joined: Mon Feb 09, 2015 2:39 pm

Re: NASDAQ as a significant portfolio component?

Post by alex_686 » Mon Oct 23, 2017 12:00 pm

overthought wrote:
Mon Oct 23, 2017 11:48 am
Valuethinker wrote:
Mon Oct 23, 2017 10:34 am
In a sense what is going on here is cherrypicking. US tech stocks have done amazingly well, albeit with an almost complete turnover in leadership more than once (of IBM it was once said "we were the future, once" ;-)).

As opposed to rocketry stocks. Or biotechnology. Or new materials. Or nuclear fusion stocks (KMS Industries). Or Shipstone Corporation ;-)
The fact that tech has maintained the lead in spite of heavy turnover in leadership is a good sign IMO: it means that no one specific companies is expected to be the long-term driver of that sector's success.
Let us assume that technology will continue to outperform. This leads to 2 questions.

First, what makes you think that this growth is not already priced into the stocks? I am willing to pay for higher growth but maybe technology valuations are too high at this point. The past has always been a unreliable predictor for the future in the stock market.

Second, why do you think that NASDAQ will continue to be dominated by tech companies and that tech companies will continue to favor the NASDAQ? We know why this was historically true. NASDAQ had much lower listing standards than NYSE. NYSE was for blue chip companies - companies with long histories of steady profits. During the 80s and 90s NASDAQ lowered their standards to let young shaky companies with inconsistent profits to be listed, allowing them to grab lots of tech start ups. This is no longer true. NASDAQ and NYSE have mostly harmonizer their listing requirements.

User avatar
nisiprius
Advisory Board
Posts: 35694
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: NASDAQ as a significant portfolio component?

Post by nisiprius » Mon Oct 23, 2017 12:17 pm

overthought wrote:
Mon Oct 23, 2017 8:35 am
So I've been running some portfolio simulations using historical data going back to the early 80's, and I noticed that, for virtually *every* 10-, 15- or 20-year investment window between 1985 and 2017, *every* sane asset allocation that includes a non-trivial amount of NASDAQ (> 5%) beats every version of a portfolio that rebalances the NASDAQ portion into US or international total stock index...
First of all, used the correct verb tense: "beat," past, not "beats," the present tense with implications that it's a permanent characteristic. Second, did it beat it in risk-adjusted return?

Second, the first thing I happened to try in Portfolio Visualizer... honestly... the first thing that occurred to me was to try a portfolio of

60% VTSMX (Total Stock), 40% VBMFX (Total Bond), portfolio 1, blue,
versus 40% Total Stock, 20% QQQ 40% Total Bond, portfolio 2, red,
over "all available data" (which in this case was limited by availability of QQQ to eighteen years, Apr 1999 - Sep 2017).

As far as I know QQQ is the oldest ETF or fund to track the NASDAQ index. (It tracks the NASDAQ 100).

I picked "month to month" to get every available month, and "no rebalancing" because that seems to be what you are advocating...

...and, no, over that time period the portfolio with the NASDAQ came in as virtually a dead heat in total return, 5.69% without QQQ and 5.70% with QQQ, but with considerably higher standard deviation, 7.99% without it and 9.36% with it, and thus it lowered risk-adjusted return as measured by the Sharpe ratio, 0.52 without it and 0.46 with it.

Actually there isn't a huge difference, but I certainly don't see any benefit to it. Again, this is over the lifetime of QQQ--but how else could an individual investor have invested in or overweighted the NASDAQ?

So I have to think it isn't completely cut-and-dried and has not, in the past, been quite the sure thing you're suggesting.

Source

Image
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
nisiprius
Advisory Board
Posts: 35694
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: NASDAQ as a significant portfolio component?

Post by nisiprius » Mon Oct 23, 2017 12:24 pm

P.S. If I substitute the ONEQ ETF which tracks the NASDAQ Composite, the time frame shortens to Nov 2003 - Sep 2017. In this case the 40/20/40 does have higher return than the portfolio without the NASDAQ, but it is counterbalanced by higher standard deviation, a measure of volatility, one form of risk, and has exactly the same Sharpe ratio. No better, no worse.

Source
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.

EddyB
Posts: 469
Joined: Fri May 24, 2013 3:43 pm

Re: NASDAQ as a significant portfolio component?

Post by EddyB » Mon Oct 23, 2017 1:00 pm

overthought wrote:
Mon Oct 23, 2017 8:35 am
So I've been running some portfolio simulations using historical data going back to the early 80's, and I noticed that, for virtually *every* 10-, 15- or 20-year investment window between 1985 and 2017, *every* sane asset allocation that includes a non-trivial amount of NASDAQ (> 5%) beats every version of a portfolio that rebalances the NASDAQ portion into US or international total stock index. That includes the time windows that span both the 2000 "dot com" and 2009 real estate crises. However, NASDAQ is conspicuously missing from Bogle-advised portfolio mixes, as far as I can tell. Is that historical accident, due to the fact that low-cost NASDAQ-tracking index funds have only been available for the last decade or so (**)? Or is NASDAQ not broad enough to sufficiently diversify risk? Or is it something else entirely?

It makes sense to me that NASDAQ should have done well, and continue to do well, for two reasons: (a) it covers the technology sector, which strikes me as having much more room to grow vs. say, manufacturing; and (b) NASDAQ listings tend to be multi-nationals whose income does not depend solely (and perhaps not even in majority part) on US fortunes.

Interested to hear others' thoughts.

(**) PRSCX seems to also track similar returns to NASDAQ, but has 0.83% ER and still only goes back to 1987.
I'm not sure I understand the premise. A significant portion of the S&P 500 (a popular index) is Nasdaq-listed stocks. A significant portion of "the total market" is Nasdaq-listed stocks. Why do you say that Nasdaq is "conspicuously missing from Bogle-advised portfolio mixes"?

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 2:41 pm

nisiprius wrote:
Mon Oct 23, 2017 12:17 pm
Second, the first thing I happened to try in Portfolio Visualizer... honestly... the first thing that occurred to me was to try a portfolio of

60% VTSMX (Total Stock), 40% VBMFX (Total Bond), portfolio 1, blue,
versus 40% Total Stock, 20% QQQ 40% Total Bond, portfolio 2, red,
over "all available data" (which in this case was limited by availability of QQQ to eighteen years, Apr 1999 - Sep 2017).

As far as I know QQQ is the oldest ETF or fund to track the NASDAQ index. (It tracks the NASDAQ 100).

I picked "month to month" to get every available month, and "no rebalancing" because that seems to be what you are advocating...
I implemented some rebalancing by steering each new (weekly) deposit into whichever piece of the portfolio would be left the least overweight by doing so. ^NDX got ahead of the new money and managed to become a little overweight in a few runs, but overall things stayed pretty well balanced to target allocations. I used ^NDX as a proxy, since as you say there's not really anything available before QQQ. I also used ^GSPC as a proxy for total market. Modern index funds seem to track both nearly perfectly, so that seemed reasonable.
nisiprius wrote:
Mon Oct 23, 2017 12:17 pm
Actually there isn't a huge difference, but I certainly don't see any benefit to it. Again, this is over the lifetime of QQQ--but how else could an individual investor have invested in or overweighted the NASDAQ?
I don't especially care if it would have been possible to invest in NASDAQ 30 years ago (I was in grade school). I was just trying to suss out the impact of a market segment that happens to be well-represented by that index. If someone chose to get into it today, there are plenty of options. Even back in the day, PRSCX was available for people willing to pay the higher ER.

That web site is new to me, BTW, thanks for the tip. I tried to set up the closest I could to my simulation as follows:
Source.

It goes back to 1987 (limited by PRSCX). It agrees that putting $100/mo into a standard three-fund portfolio and rebalancing monthly beats my proposed portfolio, though my proposed has higher Sharpe and Sortino ratios so it's somehow considered safer? Meanwhile, the 2-fund portfolio (with lowest return) has the best Sharpe and Sortino ratios by a wide margin (and lowest returns) while the S&P 500 index benchmark has the most money with the worst Sharpe and Sortino values. I'm new to both those metrics, and I'll admit that's a bit confusing to me.

Narrowing the date range from Jan 1994 to Dec 2010, the 4-fund portfolio beats the others on return but the standard three-fund comes close and has markedly lower risk. I'm not seeing the same benefit you saw from a 2-fund portfolio, though. The 3-fund seems to beat it.

So I guess the lesson here is that my simulation was doing something different from the portfolio visualizer (for better or for worse), and that the differences are indeed fairly small (2-3% difference in final value after 15+ years). So if I went this way, it would fall in the usual "some people like to spice things up a bit" category rather than "now for something entirely new."

User avatar
Ethelred
Posts: 337
Joined: Sun Oct 30, 2016 9:38 am

Re: NASDAQ as a significant portfolio component?

Post by Ethelred » Mon Oct 23, 2017 2:53 pm

overthought wrote:
Mon Oct 23, 2017 2:41 pm
I implemented some rebalancing by steering each new (weekly) deposit into whichever piece of the portfolio would be left the least overweight by doing so. ^NDX got ahead of the new money and managed to become a little overweight in a few runs, but overall things stayed pretty well balanced to target allocations. I used ^NDX as a proxy, since as you say there's not really anything available before QQQ. I also used ^GSPC as a proxy for total market. Modern index funds seem to track both nearly perfectly, so that seemed reasonable.
I could be wrong, but it looks like GSPC and NDX both exclude dividends. I wouldn't be surprised if that entirely explains the differences you're observing.

Jags4186
Posts: 1998
Joined: Wed Jun 18, 2014 7:12 pm

Re: NASDAQ as a significant portfolio component?

Post by Jags4186 » Mon Oct 23, 2017 3:13 pm

Top 10 holdings in QQQ, Nasdaq 100 index account for 52% of the holdings. Apple alone accounts for 12%.
Top 10 holdings in VTI, Total Stock Market index account for 16% of the holdings.

QQQ will perform significantly better when a few stocks have spectacular years vs VTI.

You could simply buy VTI and then add some of the top holdings of QQQ if you want to gamble a little bit.

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 3:16 pm

Ethelred wrote:
Mon Oct 23, 2017 2:53 pm
I could be wrong, but it looks like GSPC and NDX both exclude dividends. I wouldn't be surprised if that entirely explains the differences you're observing.
I use Yahoo Finance historic price quotes and export to CSV, which includes an adjusted returns column that's supposed to account for reinvested dividents (which is ironic, since none of their charts seem to be able to get that right). When I plot it vs. index funds like QQQ and VTI with reinvested dividends, there's a near-perfect match. So I'm reasonably confident that dividends are accounted for correctly.

overthought
Posts: 201
Joined: Tue Oct 17, 2017 3:44 am

Re: NASDAQ as a significant portfolio component?

Post by overthought » Mon Oct 23, 2017 3:17 pm

Jags4186 wrote:
Mon Oct 23, 2017 3:13 pm
You could simply buy VTI and then add some of the top holdings of QQQ if you want to gamble a little bit.
Yeah, but that would require more work than just buying QQQ if I want to gamble a little bit. And probably bigger monthly cash flow, too.

Valuethinker
Posts: 34680
Joined: Fri May 11, 2007 11:07 am

Re: NASDAQ as a significant portfolio component?

Post by Valuethinker » Mon Oct 23, 2017 5:06 pm

overthought wrote:
Mon Oct 23, 2017 11:48 am
Valuethinker wrote:
Mon Oct 23, 2017 10:34 am
In a sense what is going on here is cherrypicking. US tech stocks have done amazingly well, albeit with an almost complete turnover in leadership more than once (of IBM it was once said "we were the future, once" ;-)).

As opposed to rocketry stocks. Or biotechnology. Or new materials. Or nuclear fusion stocks (KMS Industries). Or Shipstone Corporation ;-)
The fact that tech has maintained the lead in spite of heavy turnover in leadership is a good sign IMO: it means that no one specific companies is expected to be the long-term driver of that sector's success.

Those other sectors did not have runs lasting anywhere near 30 years, that I'm aware of.
Tobacco certainly has. Pharmaceuticals probably has -- just not the last 15 years.

Valuethinker wrote:
Mon Oct 23, 2017 10:37 am
These internets-- you have no idea how big this is going to be. But there were something like 1000 car manufacturers in North America in 1910. Over 100 plane manufacturers in the world in the 1920s. Etc.
Number of manufacturers != size of market. But regardless, this is the "sector I think will drive economic growth for the next several decades" piece of the portfolio. I think I'd be OK with re-evaluating it every decade or two and pulling out if there are signs of long-term decline ahead. It's not like every car manufacturer simultaneously went from good times to bankrupt in a matter of months. There were pretty clear "signs of the times", spread over many years, to mark our country's exit from manufacturing.
The point is that if you look at early car manufacturers, or early PC manufacturers, most of them failed. That's true of any emergent industry.
And the same for plane manufacturers. Who now remembers McLaughlin? (became part of Oldsmobile). Studebaker and Packard? And those weren't even the earliest car companies. Osborne Computing? Southwest Technical Products? Altair?

High growth attracts more capital which drives down returns. Look at the technologically most exciting industry of the last 15-20 years - oil fracking. Those companies have found lots of oil, but they haven't made much money for shareholders.

A lot of Nasdaq stocks are basically science projects-- or at least they were in 2000. And there were the Chinese reversals into Nasdaq shell companies (that one ended badly) ...

Valuethinker
Posts: 34680
Joined: Fri May 11, 2007 11:07 am

Re: NASDAQ as a significant portfolio component?

Post by Valuethinker » Mon Oct 23, 2017 5:08 pm

Jags4186 wrote:
Mon Oct 23, 2017 3:13 pm
Top 10 holdings in QQQ, Nasdaq 100 index account for 52% of the holdings. Apple alone accounts for 12%.
Top 10 holdings in VTI, Total Stock Market index account for 16% of the holdings.

QQQ will perform significantly better when a few stocks have spectacular years vs VTI.

You could simply buy VTI and then add some of the top holdings of QQQ if you want to gamble a little bit.
And there you have part of the answer.

If it were not for the incredible performance of Apple since 1993, Nasdaq would not have done as well. If Samsung or Nokia say had dominated the smartphone market. What else is in the top 10 ?

I see, Alphabet, Amazon, Intel, Cisco, Comcast.

Jags4186
Posts: 1998
Joined: Wed Jun 18, 2014 7:12 pm

Re: NASDAQ as a significant portfolio component?

Post by Jags4186 » Mon Oct 23, 2017 7:40 pm

Valuethinker wrote:
Mon Oct 23, 2017 5:08 pm
Jags4186 wrote:
Mon Oct 23, 2017 3:13 pm
Top 10 holdings in QQQ, Nasdaq 100 index account for 52% of the holdings. Apple alone accounts for 12%.
Top 10 holdings in VTI, Total Stock Market index account for 16% of the holdings.

QQQ will perform significantly better when a few stocks have spectacular years vs VTI.

You could simply buy VTI and then add some of the top holdings of QQQ if you want to gamble a little bit.
And there you have part of the answer.

If it were not for the incredible performance of Apple since 1993, Nasdaq would not have done as well. If Samsung or Nokia say had dominated the smartphone market. What else is in the top 10 ?

I see, Alphabet, Amazon, Intel, Cisco, Comcast.
The top 5 companies account for 42% of the NASDAQ 100

Apple
Alphabet
Microsoft
Amazon
Facebook

Valuethinker
Posts: 34680
Joined: Fri May 11, 2007 11:07 am

Re: NASDAQ as a significant portfolio component?

Post by Valuethinker » Tue Oct 24, 2017 4:30 am

Jags4186 wrote:
Mon Oct 23, 2017 7:40 pm
Valuethinker wrote:
Mon Oct 23, 2017 5:08 pm
Jags4186 wrote:
Mon Oct 23, 2017 3:13 pm
Top 10 holdings in QQQ, Nasdaq 100 index account for 52% of the holdings. Apple alone accounts for 12%.
Top 10 holdings in VTI, Total Stock Market index account for 16% of the holdings.

QQQ will perform significantly better when a few stocks have spectacular years vs VTI.

You could simply buy VTI and then add some of the top holdings of QQQ if you want to gamble a little bit.
And there you have part of the answer.

If it were not for the incredible performance of Apple since 1993, Nasdaq would not have done as well. If Samsung or Nokia say had dominated the smartphone market. What else is in the top 10 ?

I see, Alphabet, Amazon, Intel, Cisco, Comcast.
The top 5 companies account for 42% of the NASDAQ 100

Apple
Alphabet
Microsoft
Amazon
Facebook
You can see then the problem? The lack of diversification.

Whilst I believe that technology really will revolutionize the world (but not just info tech: biotechnology; new materials & nanontechnology ; changes in energy technology) it's impossible to do a better job than the market in picking the winners.

Facebook and Alphabet are basically utility business models. Monopoly provision of a public service. In time, the market will not grow as much/ countries will find ways of restricting that or taking it to national control.

Microsoft is something similar and is much further down the track of being dull and worthy.

Why take an excessive risk on one sector, and one group of companies?

Jags4186
Posts: 1998
Joined: Wed Jun 18, 2014 7:12 pm

Re: NASDAQ as a significant portfolio component?

Post by Jags4186 » Tue Oct 24, 2017 7:04 am

Valuethinker wrote:
Tue Oct 24, 2017 4:30 am
You can see then the problem? The lack of diversification.

Whilst I believe that technology really will revolutionize the world (but not just info tech: biotechnology; new materials & nanontechnology ; changes in energy technology) it's impossible to do a better job than the market in picking the winners.

Facebook and Alphabet are basically utility business models. Monopoly provision of a public service. In time, the market will not grow as much/ countries will find ways of restricting that or taking it to national control.

Microsoft is something similar and is much further down the track of being dull and worthy.

Why take an excessive risk on one sector, and one group of companies?
I’m not the OP and I agree with you. If anything, if I wanted to gamble I’d look to smaller companies that have a chance to pop. I’m sure Apple and Microsoft will continue to be respectable performers but it’s hard to imagine that they can be real difference makers to a portfolio if added today.

Of course, what do I know? You look at Amazon and you look at Microsoft. One prints money and one loses it hand over fist. Yet one has performed significantly better the last several years.

User avatar
TD2626
Posts: 609
Joined: Thu Mar 16, 2017 3:40 pm

Re: NASDAQ as a significant portfolio component?

Post by TD2626 » Tue Oct 24, 2017 8:01 pm

Just my opinions:

Total Market funds already have NASDAQ stocks as a significant component.

The real question is "Is there any theoretical reason to overweight,relative to the total market, the stocks that happen to be on one particular exchange or index". I don't think so. First of all, the NASDAQ may be overweight some sectors, but it's hardly a pure sector fund. Second, without overwhelming evidence, I usually don't feel like there are very good arguments to pursue one sector over another - what do you know that the market doesn't regarding a sector's valuation? Stock markets are (over periods of many decades) supposed to be fairly efficient due to the efficient market hypothesis, so if the EMH holds there'd be no need for a long-term, buy-and-hold investor to overweight the sectors that are over-represented on the NASDAQ.

I think that the "burden of proof" regarding whether a deviation from global cap weight is an improvement rests on those proposing the deviation from global cap weight. What reasons are there to expect outperformance? Not raw outperformance - but risk-adjusted outperformance?

User avatar
CyclingDuo
Posts: 1375
Joined: Fri Jan 06, 2017 9:07 am

Re: NASDAQ as a significant portfolio component?

Post by CyclingDuo » Wed Oct 25, 2017 7:35 am

TD2626 wrote:
Tue Oct 24, 2017 8:01 pm
Just my opinions:

Total Market funds already have NASDAQ stocks as a significant component.

The real question is "Is there any theoretical reason to overweight,relative to the total market, the stocks that happen to be on one particular exchange or index". I don't think so. First of all, the NASDAQ may be overweight some sectors, but it's hardly a pure sector fund. Second, without overwhelming evidence, I usually don't feel like there are very good arguments to pursue one sector over another - what do you know that the market doesn't regarding a sector's valuation? Stock markets are (over periods of many decades) supposed to be fairly efficient due to the efficient market hypothesis, so if the EMH holds there'd be no need for a long-term, buy-and-hold investor to overweight the sectors that are over-represented on the NASDAQ.
At least with regard to the thread mentioning the QQQ ETF, it is Technology, Consumer Cyclical, and Healthcare "heavy":

Image

User avatar
Portfolio7
Posts: 356
Joined: Tue Aug 02, 2016 3:53 am

Re: NASDAQ as a significant portfolio component?

Post by Portfolio7 » Wed Oct 25, 2017 10:04 pm

CyclingDuo wrote:
Wed Oct 25, 2017 7:35 am

At least with regard to the thread mentioning the QQQ ETF, it is Technology, Consumer Cyclical, and Healthcare "heavy":
My particular Asset Allocation appears to be light on Tech and Healthcare vs the total market sector weighting (I write 'appears' because some of my portfolio funds are proprietary, though very low cost. Sector detail just is buried and I haven't tracked it down.) FOCKX, the Fidelity OTC fund, helps offset some of the gap. With a 5% allocation, I'm apparently still a bit low in those sectors.
An investment in knowledge pays the best interest.

User avatar
TinkerPDX
Posts: 266
Joined: Sun Jun 08, 2014 4:38 pm

Re: NASDAQ as a significant portfolio component?

Post by TinkerPDX » Wed Oct 25, 2017 10:16 pm

30 years isn't that long.

Nasdaq has killed this year - beware performance chasing.

Almost all of the Nasdaq stocks are also covered in sp 500: https://www.etfresearchcenter.com/tools ... Spy&f2=QQQ, and certainly in a total market fund. So it's not a question of whether to include it, but whether to overweight--in other words, NASDAQ already is a significant portion of a portfolio that holds us total market or even S&P/ US large.

Post Reply