Utilities - Does anyone own some?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Always passive
Posts: 691
Joined: Fri Apr 14, 2017 4:25 am
Location: Israel

Utilities - Does anyone own some?

Post by Always passive »

The correlation of Utilities and the total market is about .30-.40. It seems to me like a decent diversifier with a comparable return to the market (SPY). Does anyoner own some? Why? I am looking at Fidelity FUTY
Silk McCue
Posts: 4825
Joined: Thu Feb 25, 2016 7:11 pm

Re: Utilities - Does anyone own some?

Post by Silk McCue »

Yes. Inside VG Total Stock Market where it belongs for me as a long term retirement investment.

Cheers
Topic Author
Always passive
Posts: 691
Joined: Fri Apr 14, 2017 4:25 am
Location: Israel

Re: Utilities - Does anyone own some?

Post by Always passive »

Silk McCue wrote: Fri Jun 12, 2020 10:01 am Yes. Inside VG Total Stock Market where it belongs for me as a long term retirement investment.

Cheers
Thank you. Extremely useful
User avatar
eye.surgeon
Posts: 684
Joined: Wed Apr 05, 2017 1:19 pm
Location: California

Re: Utilities - Does anyone own some?

Post by eye.surgeon »

Always passive wrote: Fri Jun 12, 2020 10:13 am
Silk McCue wrote: Fri Jun 12, 2020 10:01 am Yes. Inside VG Total Stock Market where it belongs for me as a long term retirement investment.

Cheers
Thank you. Extremely useful
Not to pile on but that answer is pretty consistent with the own-the-haystack philosophy of boglehead investing. Utilities are a decent diversifier as far as they are represented in the total market.
Last edited by eye.surgeon on Fri Jun 12, 2020 10:19 am, edited 1 time in total.
"I would rather be certain of a good return than hopeful of a great one" | Warren Buffett
02nz
Posts: 5682
Joined: Wed Feb 21, 2018 3:17 pm

Re: Utilities - Does anyone own some?

Post by 02nz »

eye.surgeon wrote: Fri Jun 12, 2020 10:17 am
Always passive wrote: Fri Jun 12, 2020 10:13 am
Silk McCue wrote: Fri Jun 12, 2020 10:01 am Yes. Inside VG Total Stock Market where it belongs for me as a long term retirement investment.

Cheers
Thank you. Extremely useful
Not to pile on but that answer is pretty consistent with the own-the-haystack philosophy of boglehead investing.
And with OP's user name! :happy
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

Always passive wrote: Fri Jun 12, 2020 9:45 am The correlation of Utilities and the total market is about .30-.40. It seems to me like a decent diversifier with a comparable return to the market (SPY). Does anyoner own some? Why? I am looking at Fidelity FUTY
Increasing the weighting of utility stocks is, indeed, a reasonable way to improve the diversification of a portfolio. The effect is going to be small, but probably noticeable.

Among the sector funds, I agree that FUTY is the best choice with Vanguard's VPU being a very close second.

Because my portfolio has such a strong SCV tilt in my US stocks, I ended up building a small basket of individual utility stocks instead of using a fund but FUTY would largely have accomplished the same effect.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Tipro
Posts: 15
Joined: Sun Jul 14, 2013 8:22 pm

Re: Utilities - Does anyone own some?

Post by Tipro »

Deleted
Last edited by Tipro on Fri Jun 12, 2020 10:55 am, edited 1 time in total.
User avatar
CyclingDuo
Posts: 3725
Joined: Fri Jan 06, 2017 9:07 am

Re: Utilities - Does anyone own some?

Post by CyclingDuo »

Always passive wrote: Fri Jun 12, 2020 9:45 amThe correlation of Utilities and the total market is about .30-.40. It seems to me like a decent diversifier with a comparable return to the market (SPY). Does anyoner own some? Why? I am looking at Fidelity FUTY
Yes, we own some outside of our index funds. Prefer owning the individual stocks themselves with a diverse group of utilities (no ER fees that way), but one could also own them via an ETF such as XLU, VPU, FUTY.
"Save like a pessimist, invest like an optimist." - Morgan Housel
User avatar
Chrono Triggered
Posts: 330
Joined: Wed Feb 17, 2016 12:55 pm

Re: Utilities - Does anyone own some?

Post by Chrono Triggered »

Check out this thread if you haven't already seen it.
viewtopic.php?t=275899
User avatar
jason2459
Posts: 616
Joined: Wed May 06, 2020 7:59 pm

Re: Utilities - Does anyone own some?

Post by jason2459 »

I view sector tilts as alternatives in my AA. I allow a small percentage to them. my "alternatives" are split between REIT, Utilities, and a recent addition Energy. I hold them in my rIRA so I can't allocate to much to them even if I was tempted too. It's like my gambling and drinking money as I don't go to casinos.
dukeblue219
Posts: 841
Joined: Fri Jan 29, 2016 12:40 pm

Re: Utilities - Does anyone own some?

Post by dukeblue219 »

I have a 5% utility tilt in taxable right now, alongside 10% cash. My taxable funds are not strictly intended for retirement but some has been earmarked for potential use in the 3-5 year range. For that reason I've chosen to reduce volatility of portion of my portfolio.

No, I don't expect broad approval from BHers ;)
92irish
Posts: 259
Joined: Thu Mar 01, 2007 12:05 pm

Re: Utilities - Does anyone own some?

Post by 92irish »

About a month ago, I purchased some VPU (Vanguard Utilities ETF) for my retirement account. I had a bond mature and with rates so incredible low, I chose to increase my equity exposure with some utilities rather than reinvesting in bonds. I know it is not a bond substitute per se, but they do create a high level of income with the possibility of slow appreciation. We will see how it turns out over the years, but with utility stock prices in a bit of COVID dip and bond yields at ridiculous lows I like my chances.
User avatar
CyclingDuo
Posts: 3725
Joined: Fri Jan 06, 2017 9:07 am

Re: Utilities - Does anyone own some?

Post by CyclingDuo »

92irish wrote: Fri Jun 12, 2020 12:50 pm About a month ago, I purchased some VPU (Vanguard Utilities ETF) for my retirement account. I had a bond mature and with rates so incredible low, I chose to increase my equity exposure with some utilities rather than reinvesting in bonds. I know it is not a bond substitute per se, but they do create a high level of income with the possibility of slow appreciation. We will see how it turns out over the years, but with utility stock prices in a bit of COVID dip and bond yields at ridiculous lows I like my chances.
Utility companies for gas, electric, water, and other forms of power often operate with government regulatory protection that acts as a barrier to entry in a market, which shields companies from competition.

Retirees, conservative investors, and other investors more interested in income gravitate towards utilities.

Stock dividends from utility companies often prove to outyield other fixed-income investments and have less volatility than other equities.

Utilities tend to be very resistant to economic cycles because demand for utilities does not change much compared with most other industries, even in the deepest recessions.


https://www.investopedia.com/ask/answer ... 20equities.

Obviously, there is risk. The California fires put a utility such as PG&E as an investment at extreme risk the past few years.

Current worries would concern the pandemic and shutdown that will impact utilities in the short to intermediate term that operate in large districts where manufacturing was shut down, malls were closed, restaurants were closed, lights were turned off, gas was not be used, water was not being used, etc... .

https://www.indystar.com/story/news/env ... 308939002/

https://www.utilitydive.com/news/tracki ... or/574514/

CyclingDuo
"Save like a pessimist, invest like an optimist." - Morgan Housel
User avatar
abuss368
Posts: 22028
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: Utilities - Does anyone own some?

Post by abuss368 »

Always passive wrote: Fri Jun 12, 2020 9:45 am The correlation of Utilities and the total market is about .30-.40. It seems to me like a decent diversifier with a comparable return to the market (SPY). Does anyoner own some? Why? I am looking at Fidelity FUTY
Outside of what is market weight and held in Total Stock I do not. Jack Bogle has often said most investors could go their entire lifetime without the need for a sector fund.

Unless you are looking to increase your passive income stream I would pass. If you are trying to build passive income there are other alternatives.
John C. Bogle: “Simplicity is the master key to financial success."
User avatar
JaneyLH
Posts: 517
Joined: Wed Oct 16, 2013 7:16 pm

Re: Utilities - Does anyone own some?

Post by JaneyLH »

I used to, when I had a Financial Advisor.

Now 3-Fund Portfolio for the most part.
Linux
Posts: 5
Joined: Sat Sep 21, 2019 6:14 pm

Re: Utilities - Does anyone own some?

Post by Linux »

Always Passive, as shown in my signature I have about 10% of my portfolio in utilities. It looks like a very good diversifier to me. It's had a reliably low correlation to the market for the last 70 years or so (I don't have the analysis on hand). I like diversifying my equities, and I'm not convinced the factors investing route is the way to go.
50% NTSX, 30% VXUS, 10% VPU, 10% TIAA Real Estate
User avatar
Harry Livermore
Posts: 535
Joined: Thu Apr 04, 2019 5:32 am

Re: Utilities - Does anyone own some?

Post by Harry Livermore »

I own some water utilities and have held most of them for a long time. Pretty nice dividend yields, especially "yield on cost", and when the market goes down, they seem to go up, so I would agree that they are a diversifier. Obviously, if you want to adhere strictly to the BH philosophy and "own the haystack" you will have some exposure to them, with whatever stabilizing effect that brings.
Cheers
superinvestor
Posts: 34
Joined: Tue Apr 21, 2020 3:51 pm

Re: Utilities - Does anyone own some?

Post by superinvestor »

vineviz wrote: Fri Jun 12, 2020 10:39 am

Increasing the weighting of utility stocks is, indeed, a reasonable way to improve the diversification of a portfolio. The effect is going to be small, but probably noticeable.

Buying more of what you already have is not diversification. Especially when it takes you farther away from a market cap weighted TSM.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

superinvestor wrote: Sat Jun 13, 2020 6:28 am
vineviz wrote: Fri Jun 12, 2020 10:39 am Increasing the weighting of utility stocks is, indeed, a reasonable way to improve the diversification of a portfolio. The effect is going to be small, but probably noticeable.
Buying more of what you already have is not diversification. Especially when it takes you farther away from a market cap weighted TSM.
That's not consistent with the way diversification works.

"Buying more of what you already have" can either increase or decrease diversification: the direction of the effect depends entirely on three things: the variance of what you're buying more of, the correlation of what you're buying more of with what you already own, and how much of it you are buying.

Strict market cap weighted portfolios have a few special properties, but being maximally diversified isn't one of them.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Housedoc
Posts: 719
Joined: Thu May 24, 2018 4:25 pm
Location: Alabama

Re: Utilities - Does anyone own some?

Post by Housedoc »

I have some Southern Company and Duke shares purchased long time ago before I began index funds. I switched to dividend in cash from reinventment at retirement. Nice to treat ourselves quarterly to some fun. Need Covid to slow so fun can happen.
superinvestor
Posts: 34
Joined: Tue Apr 21, 2020 3:51 pm

Re: Utilities - Does anyone own some?

Post by superinvestor »

vineviz wrote: Sat Jun 13, 2020 7:01 am

That's not consistent with the way diversification works.

"Buying more of what you already have" can either increase or decrease diversification: the direction of the effect depends entirely on three things: the variance of what you're buying more of, the correlation of what you're buying more of with what you already own, and how much of it you are buying.

Strict market cap weighted portfolios have a few special properties, but being maximally diversified isn't one of them.
Personally I do not include volatility in my definition of risk. I don't care what my volatility is as long as I am getting market returns. Which I am basically getting with VTI minus tracking error and fees.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

superinvestor wrote: Sat Jun 13, 2020 7:42 am
vineviz wrote: Sat Jun 13, 2020 7:01 am

That's not consistent with the way diversification works.

"Buying more of what you already have" can either increase or decrease diversification: the direction of the effect depends entirely on three things: the variance of what you're buying more of, the correlation of what you're buying more of with what you already own, and how much of it you are buying.

Strict market cap weighted portfolios have a few special properties, but being maximally diversified isn't one of them.
Personally I do not include volatility in my definition of risk. I don't care what my volatility is as long as I am getting market returns. Which I am basically getting with VTI minus tracking error and fees.
That very well may be true. Nonetheless, diversification has a specific meaning and that meaning is inseparably linked to the concepts of correlation and variance.

For investors who are in the early to middle phases of accumulation, portfolio diversification and portfolio volatility (not the same thing, by the way) take a back seat to expected returns.

For investors who are in decumulation (i.e. making withdrawals) or approaching that situation, however, portfolio diversification and portfolio volatility play a crucial role in determining the amount of lifetime consumption the portfolio can support.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
CyclingDuo
Posts: 3725
Joined: Fri Jan 06, 2017 9:07 am

Re: Utilities - Does anyone own some?

Post by CyclingDuo »

vineviz wrote: Sat Jun 13, 2020 7:49 am
superinvestor wrote: Sat Jun 13, 2020 7:42 am
vineviz wrote: Sat Jun 13, 2020 7:01 am

That's not consistent with the way diversification works.

"Buying more of what you already have" can either increase or decrease diversification: the direction of the effect depends entirely on three things: the variance of what you're buying more of, the correlation of what you're buying more of with what you already own, and how much of it you are buying.

Strict market cap weighted portfolios have a few special properties, but being maximally diversified isn't one of them.
Personally I do not include volatility in my definition of risk. I don't care what my volatility is as long as I am getting market returns. Which I am basically getting with VTI minus tracking error and fees.
That very well may be true. Nonetheless, diversification has a specific meaning and that meaning is inseparably linked to the concepts of correlation and variance.

For investors who are in the early to middle phases of accumulation, portfolio diversification and portfolio volatility (not the same thing, by the way) take a back seat to expected returns.

For investors who are in decumulation (i.e. making withdrawals) or approaching that situation, however, portfolio diversification and portfolio volatility play a crucial role in determining the amount of lifetime consumption the portfolio can support.
Well said!
"Save like a pessimist, invest like an optimist." - Morgan Housel
Rosencrantz1
Posts: 561
Joined: Tue Sep 10, 2019 12:28 pm

Re: Utilities - Does anyone own some?

Post by Rosencrantz1 »

92irish wrote: Fri Jun 12, 2020 12:50 pm About a month ago, I purchased some VPU (Vanguard Utilities ETF) for my retirement account. I had a bond mature and with rates so incredible low, I chose to increase my equity exposure with some utilities rather than reinvesting in bonds. I know it is not a bond substitute per se, but they do create a high level of income with the possibility of slow appreciation. We will see how it turns out over the years, but with utility stock prices in a bit of COVID dip and bond yields at ridiculous lows I like my chances.
Makes sense. I like your idea. I may have to think about doing some of the same - especially with utilities in a 'bit of a covid dip'.
User avatar
Uncorrelated
Posts: 1100
Joined: Sun Oct 13, 2019 3:16 pm

Re: Utilities - Does anyone own some?

Post by Uncorrelated »

vineviz wrote: Fri Jun 12, 2020 10:39 am
Always passive wrote: Fri Jun 12, 2020 9:45 am The correlation of Utilities and the total market is about .30-.40. It seems to me like a decent diversifier with a comparable return to the market (SPY). Does anyoner own some? Why? I am looking at Fidelity FUTY
Increasing the weighting of utility stocks is, indeed, a reasonable way to improve the diversification of a portfolio. The effect is going to be small, but probably noticeable.
There is no evidence that over-weighting utilities is a reasonable way to improve the diversification of a portfolio. Low correlation is not a sufficient metric for portfolio improvement.

According to https://papers.ssrn.com/sol3/papers.cfm ... id=2965146, utilities have low market exposure (.42), some value exposure (.32), and very high corporate bonds exposure (.78). This explains the seemingly low correlation. If you think the characteristics of utilities are attractive, you are better off tilting towards value and corporate bonds. This results in the same factor exposure but lower idiosyncratic risk.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

Uncorrelated wrote: Sun Jun 14, 2020 3:49 am There is no evidence that over-weighting utilities is a reasonable way to improve the diversification of a portfolio. Low correlation is not a sufficient metric for portfolio improvement.
Throwing a subjective word like "reasonable" into a sentence after the word "evidence" makes your claim neither defensible nor refutable. In other words, not very helpful.

The paper you mentioned is not completely wrong: utility stocks DO have factor exposures that differ from the market, which is (obviously) what makes them a potential source of diversification.

The factor exposures aren't limited to value and interest rate sensitivity ("corporate bonds exposure" isn't a factor, and utilities actually have a negative exposure to the fixed income "credit" factor). Utility stocks (depending on the fund composition) typically also have exposure to momentum, quality, and low volatility factors.

Plus, a full factor regression for utility stocks typically has an r-squared of around 50% or so compared to >90% for most other index funds. This tells you that typical factor models aren't fully describing the diversification benefits of utility stocks: you can't simply replicate their addition by tuning up the value exposure.

With plenty of low-cost utility index funds available, replacing 25% to 30% (for example) of a TSM allocation with utilities will improve the diversification of the portfolio without incurring excessive fund expenses.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
tibbitts
Posts: 12237
Joined: Tue Feb 27, 2007 6:50 pm

Re: Utilities - Does anyone own some?

Post by tibbitts »

I would have bought some of Vanguard's utility fund, were it not for what seems like (to me) an outrageous $100,000 minimum. Even at $50k I probably would have. Maybe not so much now, but at one time.
stan1
Posts: 9134
Joined: Mon Oct 08, 2007 4:35 pm

Re: Utilities - Does anyone own some?

Post by stan1 »

tibbitts wrote: Sun Jun 14, 2020 9:06 am I would have bought some of Vanguard's utility fund, were it not for what seems like (to me) an outrageous $100,000 minimum. Even at $50k I probably would have. Maybe not so much now, but at one time.
There is an ETF share class VPU. If you refuse to use ETFs that's a different issue.
tibbitts
Posts: 12237
Joined: Tue Feb 27, 2007 6:50 pm

Re: Utilities - Does anyone own some?

Post by tibbitts »

stan1 wrote: Sun Jun 14, 2020 9:08 am
tibbitts wrote: Sun Jun 14, 2020 9:06 am I would have bought some of Vanguard's utility fund, were it not for what seems like (to me) an outrageous $100,000 minimum. Even at $50k I probably would have. Maybe not so much now, but at one time.
There is an ETF share class VPU. If you refuse to use ETFs that's a different issue.
I don't refuse to own ETFs, but it would require opening yet another Vanguard account, besides my taxable, Roth, traditional, rollover, and SEP mutual fund accounts. I don't think the complexity would be worthwhile. I do think the $100,000 minimum, certainly vs. $50,000, is unreasonable.
james22
Posts: 1706
Joined: Tue Aug 21, 2007 2:22 pm

Re: Utilities - Does anyone own some?

Post by james22 »

Buy $100,000, sell $50,000+ the next day?
User avatar
Uncorrelated
Posts: 1100
Joined: Sun Oct 13, 2019 3:16 pm

Re: Utilities - Does anyone own some?

Post by Uncorrelated »

vineviz wrote: Sun Jun 14, 2020 8:38 am
Uncorrelated wrote: Sun Jun 14, 2020 3:49 am There is no evidence that over-weighting utilities is a reasonable way to improve the diversification of a portfolio. Low correlation is not a sufficient metric for portfolio improvement.
Throwing a subjective word like "reasonable" into a sentence after the word "evidence" makes your claim neither defensible nor refutable. In other words, not very helpful.

The paper you mentioned is not completely wrong: utility stocks DO have factor exposures that differ from the market, which is (obviously) what makes them a potential source of diversification.

The factor exposures aren't limited to value and interest rate sensitivity ("corporate bonds exposure" isn't a factor, and utilities actually have a negative exposure to the fixed income "credit" factor). Utility stocks (depending on the fund composition) typically also have exposure to momentum, quality, and low volatility factors.

Plus, a full factor regression for utility stocks typically has an r-squared of around 50% or so compared to >90% for most other index funds. This tells you that typical factor models aren't fully describing the diversification benefits of utility stocks: you can't simply replicate their addition by tuning up the value exposure.

With plenty of low-cost utility index funds available, replacing 25% to 30% (for example) of a TSM allocation with utilities will improve the diversification of the portfolio without incurring excessive fund expenses.
I copied your phrasing. Rather than being reasonable, I think it would be more accurate to state that there is no evidence at all.

All sectors have factor exposure that differs from the market. That is not a sufficient condition to over-weight a certain sector. All sectors have R^2 < 100%, that is not a sufficient condition to over-weight a single sector. If you think average factor exposure and low R^2 is a sufficient condition for a tilt, over-weight apple.

It is possible to replicate all the benefits of the utility sector with the appropriate small/value/bond funds unless the utilities sector contains statistically significant positive alpha. There is no evidence this sector contains such a thing.

Active tilts require extraordinary evidence.
User avatar
Forester
Posts: 1606
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Utilities - Does anyone own some?

Post by Forester »

It's possible that some sectors or industries are inherently better than others on a risk-adjusted basis. Personally instead of Utilities I would buy a Min vol or Low vol fund, to avoid a concentrated bet.
Colorado Guy
Posts: 94
Joined: Sun Nov 05, 2017 1:57 pm

Re: Utilities - Does anyone own some?

Post by Colorado Guy »

CyclingDuo wrote: Fri Jun 12, 2020 1:13 pm
92irish wrote: Fri Jun 12, 2020 12:50 pm About a month ago, I purchased some VPU (Vanguard Utilities ETF) for my retirement account. I had a bond mature and with rates so incredible low, I chose to increase my equity exposure with some utilities rather than reinvesting in bonds. I know it is not a bond substitute per se, but they do create a high level of income with the possibility of slow appreciation. We will see how it turns out over the years, but with utility stock prices in a bit of COVID dip and bond yields at ridiculous lows I like my chances.
Utility companies for gas, electric, water, and other forms of power often operate with government regulatory protection that acts as a barrier to entry in a market, which shields companies from competition.

Retirees, conservative investors, and other investors more interested in income gravitate towards utilities.

Stock dividends from utility companies often prove to outyield other fixed-income investments and have less volatility than other equities.
.....(partial quote)

CyclingDuo


Interesting. The "Abandon Bonds" strategy thread is currently active. While different alternative investments to bonds are being discussed (pros and cons), the alternative of dividend paying stocks with low volatility were not. I used to hold a significant % of specific utility stock, but was advised at the time that this was not a wise investment as it was concentrated on one utility. I have not considered VPU as an addition to some of my existing bond allocation, but now wondering if it is a reasonable action due to both low volatility and dividends.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

Uncorrelated wrote: Sun Jun 14, 2020 9:46 am Rather than being reasonable, I think it would be more accurate to state that there is no evidence at all.
I think you're being combative where such an approach is unwarranted, and building strawman arguments to do it.

Among the GICS sectors, utilities are the least well explained by traditional factor models (and not by a little bit). Thus, there's no way a utility "tilt" can be fully replicated with traditional factor exposures.

Among the GICS sectors, utilities have the lowest correlation with a the total stock market with similar variance. Thus, any attempt to diversify a large-cap US equity portfolio that DOESN'T overweight utilities is going to be less effective than it could be.

In any event, you need not rely on theory alone. If there is low-cost US equity ETF that, when combined with a total US stock market fund, results in a MORE diversified portfolio than a 75/25 portfolio of VTI and VPU I'd love to see it.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
tibbitts
Posts: 12237
Joined: Tue Feb 27, 2007 6:50 pm

Re: Utilities - Does anyone own some?

Post by tibbitts »

james22 wrote: Sun Jun 14, 2020 9:45 am Buy $100,000, sell $50,000+ the next day?
That would work but I don't have another $50k - it would be too complicated.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

Colorado Guy wrote: Sun Jun 14, 2020 10:13 am Interesting. The "Abandon Bonds" strategy thread is currently active. While different alternative investments to bonds are being discussed (pros and cons), the alternative of dividend paying stocks with low volatility were not. I used to hold a significant % of specific utility stock, but was advised at the time that this was not a wise investment as it was concentrated on one utility. I have not considered VPU as an addition to some of my existing bond allocation, but now wondering if it is a reasonable action due to both low volatility and dividends.
Even low-volatility stocks are not, IMHO, generally a suitable substitute for bonds in a portfolio.

Although utility stocks are good diversifiers, they are still going to behave more like stocks than bonds. People sometimes call utility stocks "bond proxies" or "bond surrogates", but such language is misleading.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Colorado Guy
Posts: 94
Joined: Sun Nov 05, 2017 1:57 pm

Re: Utilities - Does anyone own some?

Post by Colorado Guy »

vineviz wrote: Sun Jun 14, 2020 10:45 am Even low-volatility stocks are not, IMHO, generally a suitable substitute for bonds in a portfolio.

Although utility stocks are good diversifiers, they are still going to behave more like stocks than bonds. People sometimes call utility stocks "bond proxies" or "bond surrogates", but such language is misleading.
Good food for thought. Do you actually use, for your stock allocation, 75/25 portfolio of VTI and VPU, or was that just an example of diversification? If you do use 75/25 VTI/VPU, what has been your experience?
langlands
Posts: 598
Joined: Wed Apr 03, 2019 10:05 pm

Re: Utilities - Does anyone own some?

Post by langlands »

What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
User avatar
Uncorrelated
Posts: 1100
Joined: Sun Oct 13, 2019 3:16 pm

Re: Utilities - Does anyone own some?

Post by Uncorrelated »

vineviz wrote: Sun Jun 14, 2020 10:29 am
Uncorrelated wrote: Sun Jun 14, 2020 9:46 am Rather than being reasonable, I think it would be more accurate to state that there is no evidence at all.
I think you're being combative where such an approach is unwarranted, and building strawman arguments to do it.

Among the GICS sectors, utilities are the least well explained by traditional factor models (and not by a little bit). Thus, there's no way a utility "tilt" can be fully replicated with traditional factor exposures.

Among the GICS sectors, utilities have the lowest correlation with a the total stock market with similar variance. Thus, any attempt to diversify a large-cap US equity portfolio that DOESN'T overweight utilities is going to be less effective than it could be.

In any event, you need not rely on theory alone. If there is low-cost US equity ETF that, when combined with a total US stock market fund, results in a MORE diversified portfolio than a 75/25 portfolio of VTI and VPU I'd love to see it.
I said that all the benefits of a utilities tilt can be replicated with a factor tilt. There is no evidence that an utilities tilt contains benefits that are not already explained by traditional factor models. Low R^2 is not indicative of better diversification. Unless combined with statistically significant alpha, it's generally indicative of concentration.

Apple is less well-explained by factor models than utilities. Which objective metrics have you used to determine that an utilities tilt results in better diversification than an apple tilt. How do other GICS sectors fare in comparison? Is the performance of the utilities sector statistically significant compared to other sectors?
User avatar
Uncorrelated
Posts: 1100
Joined: Sun Oct 13, 2019 3:16 pm

Re: Utilities - Does anyone own some?

Post by Uncorrelated »

langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Running Markowitz portfolio optimization will give the wrong answer in this situation. For example, suppose that you have 3 asset classes to choose from: Stocks, bonds and hedge funds. Hedge funds invest passively in stocks and bonds and add some expense ratio. In this situation markowitz portfolio optimization will tell you that the optimal portfolio is a combination of all 3 asset classes. But if you run a factor regression, then you will see that the hedge fund asset class is 100% redundant.

In general, valid investments must contain independent positive risk factors. There is no evidence the utilities sector contains such things. Over-weighting utilities is like entering a casino. Yes you will add uncorrelated sources of risk, but that is not helpful in itself...
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

Uncorrelated wrote: Sun Jun 14, 2020 11:36 am
I said that all the benefits of a utilities tilt can be replicated with a factor tilt.
Yes, you said that but it's not correct.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
langlands
Posts: 598
Joined: Wed Apr 03, 2019 10:05 pm

Re: Utilities - Does anyone own some?

Post by langlands »

Uncorrelated wrote: Sun Jun 14, 2020 11:46 am
langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Running Markowitz portfolio optimization will give the wrong answer in this situation. For example, suppose that you have 3 asset classes to choose from: Stocks, bonds and hedge funds. Hedge funds invest passively in stocks and bonds and add some expense ratio. In this situation markowitz portfolio optimization will tell you that the optimal portfolio is a combination of all 3 asset classes. But if you run a factor regression, then you will see that the hedge fund asset class is 100% redundant.

In general, valid investments must contain independent positive risk factors. There is no evidence the utilities sector contains such things. Over-weighting utilities is like entering a casino. Yes you will add uncorrelated sources of risk, but that is not helpful in itself...
I don't understand. Just because an asset class is redundant doesn't mean it will cause problems in the portfolio optimization. What will happen in your scenario is that because the hedge fund has the expense ratio but otherwise just replicates stocks and bonds, the optimizer will put 0 weight on the hedge fund. When you compute the rate of return of VOO and VPU, of course feel free to add in the expense ratio for both funds and any other costs you feel are necessary to the rate of return calculation.
jeep5ter
Posts: 64
Joined: Mon Jan 07, 2013 5:50 pm

Re: Utilities - Does anyone own some?

Post by jeep5ter »

Housedoc wrote: Sat Jun 13, 2020 7:25 am I have some Southern Company and Duke shares purchased long time ago before I began index funds. I switched to dividend in cash from reinventment at retirement. Nice to treat ourselves quarterly to some fun.
Same here.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Mean-variance analysis won't shed any light on the question of diversification, since diversification isn't what mean-variance optimization solves for.

The most diversified combination of two assets is approximated by solving for risk parity, though, and this is easy to do with just two numbers: the volatilities (standard deviation) of the two assets. To get the maximum diversification combination simply weight the two assets by the inverse of their volatility. This is not 100% sound in this case because the socks in VPU are also in VTI, which technically violates one of the theoretical assumptions underlying the approach. However, the small allocation of utilities in VTI (just 3.2% currently) means that this violation doesn't have much effect on the calculated result.

Presumably your actual portfolio has more than just two assets, so this solution isn't generally pragmatic.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Uncorrelated
Posts: 1100
Joined: Sun Oct 13, 2019 3:16 pm

Re: Utilities - Does anyone own some?

Post by Uncorrelated »

langlands wrote: Sun Jun 14, 2020 12:07 pm
Uncorrelated wrote: Sun Jun 14, 2020 11:46 am
langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Running Markowitz portfolio optimization will give the wrong answer in this situation. For example, suppose that you have 3 asset classes to choose from: Stocks, bonds and hedge funds. Hedge funds invest passively in stocks and bonds and add some expense ratio. In this situation markowitz portfolio optimization will tell you that the optimal portfolio is a combination of all 3 asset classes. But if you run a factor regression, then you will see that the hedge fund asset class is 100% redundant.

In general, valid investments must contain independent positive risk factors. There is no evidence the utilities sector contains such things. Over-weighting utilities is like entering a casino. Yes you will add uncorrelated sources of risk, but that is not helpful in itself...
I don't understand. Just because an asset class is redundant doesn't mean it will cause problems in the portfolio optimization. What will happen in your scenario is that because the hedge fund has the expense ratio but otherwise just replicates stocks and bonds, the optimizer will put 0 weight on hedge funds. When you compute the rate of return of VOO and VPU, of course feel free to add in the expense ratio for both funds and any other costs you feel are necessary to the rate of return calculation.
Suppose that:
Stocks: 6% return, 10% std
Bonds: 6% return, 10% std
stocks and bonds uncorrelated
hedge funds = stocks * 50% + bonds * 50%

If we run markowitz portfolio optimization, the optimal allocation is 37% stocks, 37% bonds, 26% hedge fund with a total standard deviation of 6.6%. This result is wrong: since there are only two truly independent sources of risk in our example, it should be impossible to obtain a standard deviation below 7.1%.

Now replace "hedge fund" with "a tiny slice of the stock market" and you see why I require extraordinary amounts of evidence for the claim "X improves diversification".
langlands
Posts: 598
Joined: Wed Apr 03, 2019 10:05 pm

Re: Utilities - Does anyone own some?

Post by langlands »

vineviz wrote: Sun Jun 14, 2020 12:18 pm
langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Mean-variance analysis won't shed any light on the question of diversification, since diversification isn't what mean-variance optimization solves for.

The most diversified combination of two assets is approximated by solving for risk parity, though, and this is easy to do with just two numbers: the volatilities (standard deviation) of the two assets. To get the maximum diversification combination simply weight the two assets by the inverse of their volatility. This is not 100% sound in this case because the socks in VPU are also in VTI, which technically violates one of the theoretical assumptions underlying the approach. However, the small allocation of utilities in VTI (just 3.2% currently) means that this violation doesn't have much effect on the calculated result.

Presumably your actual portfolio has more than just two assets, so this solution isn't generally pragmatic.
Mean-variance analysis is the quantitative answer to the vague notion of "diversification," a word that is thrown around very often but rarely defined precisely. The idea is simply to find the mix of assets that result in a portfolio with the highest risk-adjusted return. It is literally the only thing that matters if you assume only quadratic risks (e.g. if you assume normally distributed returns for all assets), which is a pretty good place to start.

Risk parity is just mean-variance analysis where you assume a diagonal covariance matrix, i.e. you assume the correlation between all assets is 0.

Of course your portfolio contains more than two assets. It might be a good start though to figure out whether VPU would help diversify VOO. Start from the simplest case and add complexity as needed.
langlands
Posts: 598
Joined: Wed Apr 03, 2019 10:05 pm

Re: Utilities - Does anyone own some?

Post by langlands »

Uncorrelated wrote: Sun Jun 14, 2020 12:28 pm
langlands wrote: Sun Jun 14, 2020 12:07 pm
Uncorrelated wrote: Sun Jun 14, 2020 11:46 am
langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Running Markowitz portfolio optimization will give the wrong answer in this situation. For example, suppose that you have 3 asset classes to choose from: Stocks, bonds and hedge funds. Hedge funds invest passively in stocks and bonds and add some expense ratio. In this situation markowitz portfolio optimization will tell you that the optimal portfolio is a combination of all 3 asset classes. But if you run a factor regression, then you will see that the hedge fund asset class is 100% redundant.

In general, valid investments must contain independent positive risk factors. There is no evidence the utilities sector contains such things. Over-weighting utilities is like entering a casino. Yes you will add uncorrelated sources of risk, but that is not helpful in itself...
I don't understand. Just because an asset class is redundant doesn't mean it will cause problems in the portfolio optimization. What will happen in your scenario is that because the hedge fund has the expense ratio but otherwise just replicates stocks and bonds, the optimizer will put 0 weight on hedge funds. When you compute the rate of return of VOO and VPU, of course feel free to add in the expense ratio for both funds and any other costs you feel are necessary to the rate of return calculation.
Suppose that:
Stocks: 6% return, 10% std
Bonds: 6% return, 10% std
stocks and bonds uncorrelated
hedge funds = stocks * 50% + bonds * 50%

If we run markowitz portfolio optimization, the optimal allocation is 37% stocks, 37% bonds, 26% hedge fund with a total standard deviation of 6.6%. This result is wrong: since there are only two truly independent sources of risk in our example, it should be impossible to obtain a standard deviation below 7.1%.

Now replace "hedge fund" with "a tiny slice of the stock market" and you see why I require extraordinary amounts of evidence for the claim "X improves diversification".
The standard deviation of the 37%, 37%, 26% portfolio is not 6.6%, but 7.1% exactly as you would expect. How did you get 6.6%?
james22
Posts: 1706
Joined: Tue Aug 21, 2007 2:22 pm

Re: Utilities - Does anyone own some?

Post by james22 »

Uncorrelated wrote: Sun Jun 14, 2020 11:36 am
vineviz wrote: Sun Jun 14, 2020 10:29 amAmong the GICS sectors, utilities are the least well explained by traditional factor models (and not by a little bit). Thus, there's no way a utility "tilt" can be fully replicated with traditional factor exposures.
I said that all the benefits of a utilities tilt can be replicated with a factor tilt. There is no evidence that an utilities tilt contains benefits that are not already explained by traditional factor models. Low R^2 is not indicative of better diversification. Unless combined with statistically significant alpha, it's generally indicative of concentration.

Apple is less well-explained by factor models than utilities. Which objective metrics have you used to determine that an utilities tilt results in better diversification than an apple tilt. How do other GICS sectors fare in comparison? Is the performance of the utilities sector statistically significant compared to other sectors?
We've covered this before:

viewtopic.php?f=10&t=275899&start=100#p4673540
User avatar
Uncorrelated
Posts: 1100
Joined: Sun Oct 13, 2019 3:16 pm

Re: Utilities - Does anyone own some?

Post by Uncorrelated »

langlands wrote: Sun Jun 14, 2020 12:37 pm
Uncorrelated wrote: Sun Jun 14, 2020 12:28 pm
langlands wrote: Sun Jun 14, 2020 12:07 pm
Uncorrelated wrote: Sun Jun 14, 2020 11:46 am
langlands wrote: Sun Jun 14, 2020 11:33 am What does the Markowitz mean variance portfolio say about the ideal mix of VOO and VPU? We just need 6 numbers: the rate of return of VOO, the rate of return of VPU, and the 2x2 covariance matrix of VOO and VPU. I'm not very good at pulling up these kind of numbers, but I'm sure vineviz or Uncorrelated could do so easily. I suggest using the max length possible for the rates of return (since rate of return is so noisy) and a 1 year window for the covariance matrix (since it can be estimated much more precisely).
Running Markowitz portfolio optimization will give the wrong answer in this situation. For example, suppose that you have 3 asset classes to choose from: Stocks, bonds and hedge funds. Hedge funds invest passively in stocks and bonds and add some expense ratio. In this situation markowitz portfolio optimization will tell you that the optimal portfolio is a combination of all 3 asset classes. But if you run a factor regression, then you will see that the hedge fund asset class is 100% redundant.

In general, valid investments must contain independent positive risk factors. There is no evidence the utilities sector contains such things. Over-weighting utilities is like entering a casino. Yes you will add uncorrelated sources of risk, but that is not helpful in itself...
I don't understand. Just because an asset class is redundant doesn't mean it will cause problems in the portfolio optimization. What will happen in your scenario is that because the hedge fund has the expense ratio but otherwise just replicates stocks and bonds, the optimizer will put 0 weight on hedge funds. When you compute the rate of return of VOO and VPU, of course feel free to add in the expense ratio for both funds and any other costs you feel are necessary to the rate of return calculation.
Suppose that:
Stocks: 6% return, 10% std
Bonds: 6% return, 10% std
stocks and bonds uncorrelated
hedge funds = stocks * 50% + bonds * 50%

If we run markowitz portfolio optimization, the optimal allocation is 37% stocks, 37% bonds, 26% hedge fund with a total standard deviation of 6.6%. This result is wrong: since there are only two truly independent sources of risk in our example, it should be impossible to obtain a standard deviation below 7.1%.

Now replace "hedge fund" with "a tiny slice of the stock market" and you see why I require extraordinary amounts of evidence for the claim "X improves diversification".
The standard deviation of the 37%, 37%, 26% portfolio is not 6.6%, but 7.1% exactly as you would expect. How did you get 6.6%?
You're right, I made a calculation error.

AFAIK it still breaks in slightly less trivial examples where we define hedge funds as a small subset of stocks.
User avatar
vineviz
Posts: 8001
Joined: Tue May 15, 2018 1:55 pm

Re: Utilities - Does anyone own some?

Post by vineviz »

langlands wrote: Sun Jun 14, 2020 12:33 pmMean-variance analysis is the quantitative answer to the vague notion of "diversification," a word that is thrown around very often but rarely defined precisely. The idea is simply to find the mix of assets that result in a portfolio with the highest risk-adjusted return. It is literally the only thing that matters if you assume only quadratic risks (e.g. if you assume normally distributed returns for all assets), which is a pretty good place to start.
Mean-variance analysis is only tangentially related to diversification if you make one clearly erroneous assumption, which is that "risk" is synonymous with "variance". Even with that assumption, the relationship is only tangential because mean-variance optimization is an attempt to solve for two portfolio variables (expected return and expected variance) whereas diversification is concerned with only one of those two variables (i.e. variance).

Thus, a higher risk-adjusted return is NOT necessarily indicative of greater diversification no matter what utility function one assumes.

For a 2-asset problem, the maximum diversification solution and risk parity solution both happen to be identical to the maximum Sharpe-ratio solution under the (not obviously supportable) assumption that both assets have the same Sharpe ratio. If you believe in a two-factor CAPM model that assumption might be defensible. If you believe we live in a world in which there are more than two sources of systematic risk, the assumption of equal Sharpe ratios is a dubious one.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Post Reply