We’d like to remind Forumites to please avoid political debate on the Forum.

This is to keep it a safe and useful space for MoneySaving discussions. Threads that are – or become – political in nature may be removed in line with the Forum’s rules. Thank you for your understanding.

📨 Have you signed up to the Forum's new Email Digest yet? Get a selection of trending threads sent straight to your inbox daily, weekly or monthly!

rebalancing my portfolio

13

Comments

  • Rollinghome
    Rollinghome Posts: 2,741 Forumite
    Part of the Furniture 1,000 Posts Name Dropper
    edited 12 November 2009 at 2:21PM
    Quote: But then you are a financial adviser and if your clients withdraw their investments you lose commision don't you?
    No.
    So they keep paying you your trail commission for the investment even when it's ended, do they? Tut, tut D. :rolleyes:

    Fair enough Dunston. You're in the business of persuading your prospects to invest whether you expect the markets to go up or down. No investment, no trail commission. And it's that confict of interest that's the problem with many IFAs.

    Let's hope the situation improves somewhat when the FSA forces through the changes it wants by 2012.


    .
  • xyy123
    xyy123 Posts: 61 Forumite
    jon3001 wrote: »
    I disagree. There's no reason to expect a diversified portfolio, especially one heavy in alternative assets such as commodity futures and property, to behave like any particular benchmark. So what's the value in benchmarking the whole porfolio against a narrow index? The purpose of the porfolio is balance risk and return; not target some arbitrary benchmark.

    If you want to assess individual assets within the porfolio against a benchmark (e.g. to see if a manager is adding value) that could be worthwhile. As a corollary you could produce a weighted benchmark of the underlying indexes. But if you're already doing it at the asset level then I can't see what value that holds for the private investor.
    jon3001 - You have completely misunderstood my post. A benchmark should not be used as measure of how good/bad your portfolio is doing. That is its secondary use. The point is to fill your portfolio with the assets contained within the benchmark. In other words, you may have a blended benchmark - 50% FTSE 100:50% UK gilts - which will fulfill your objectives in say 7 years, in which case your portfolio should most of the time remain like that, although you may attempt to deviate slightly in order to reach your benchmark sooner, by underweighting Gilts/Equities as you see fit. Or overweighting/underweighting within equities. I thought I was clear on this:
    xyy123 wrote: »
    Deciding on the correct asset allocation has to take this into account. Its not a question of correlation coeffecients. Its the opposite. You need the assets/constituents in the benchmark, in your portfolio, and hence exactly/closely related to whatever benchmark you choose so that your portfolio does roughly what the benchmark does.

    i.e. If your benchmark is FTSE 100 (not a fantastic benchmark but adequate), you should have enough FTSE 100 stock in your portfolio, such that the performance is broadly speaking the same as the benchmark - an index tracker does this perfectly (almost). But if you want to try and add a little value and attempt to beat the benchmark, then you need to make a/some decisions. Maybe you'll underweight financials. Maybe you'll include a little foreign exposure. But the more you deviate, the more risk you take on in not meeting the objective you set out to meet, which has to tie in with the selected benchmark.

    If the FTSE 100 is 25% financials, 20% energy, 15% consumer staples, then thats the approximate weighting you should have in your portfolio. Or maybe you think you can add value, so you go 20% financials. Its a benchmark risk because if you're wrong and financials do better than the benchmark average, your portfolio will lag and it will take longer to reach your goal.
    You don't specifiy how much growth of the portoflio is required. Clearly somone only wanting, say, 50% growth can achieve their goals with less shortfall risk by introducing a healthy amount of bonds and making other portfolio adjustments.
    True. But most people tend to underestimate how much money they need later in life. If they are positive they only need that tiny amount of growth, then of course they don't need equity, or even bond for that matter, exposure. They could probably achieve it in cash and gilts.
    To paraphrase:

    But the last 10-years in a diversified portfolio has been worse than the FTSE-100.
    [I doubt a decent one has but I'll just use your benchmark]
    I don't think you read my post correctly - I said the last 10 years in bonds beat equities...

    Over longer time horizons (ten years or more), portfolios containing uncorrelated assets provide the greatest return and the least risk, usually outperforming any individual consistuent asset.

    I refer you again to this:
    Over longer time horizons (ten years or more), equities historically provide the greatest return, outperforming bonds in 98% of the 20-year rolling periods since 1926. Even at only ten years, equities beat bonds nearly 90% of the time.
    Equities are historically the highest yielding asset class. So, again, having an element of bonds or whatever in a long-term growth portfolio acts as a drag.
    You keep focusing on performance when for many people controlling volatility is important. A small amount of bonds can dramatically decrease volatility with only a small performance price.
    Again, you have either misunderstood or not fully read my post:
    Having any permanent element of bonds/cash then in a long-term growth portfolio is basically saying that you think the next ten year period will be one where equities underperform bonds. In other words, a 10-1 bet. The odds are therefore not in your favour.
    Unfortunately, people worry more about short-term volatility than their long-term objective. And, funnily enough, its only downside volatility people worry about. I myself love upside volatility. All a permanent allocation to fixed interest does in a long-term growth portfolio, without any needs for cashflow or income, is act as an anchor and restraint on the potential portfolio performance.

    Apart from the fact that, and I speak purely for myself here, I invest to make more money. Plain and simple. I don't want to lose it and I know how much risk I am willing to take. Having worked in the industry, I understand market risk better than most. Short-term volatility is only really an issue if you need the money soon. If you need the money soon, you shouldn't have it in a volatile asset class.
    What's the value in benchmarking cash funds you're about to spend? Most private investors would look for a decent interest rate and security and maybe other factors (ease of access, etc).
    There isn't much point, but it served to highlight my point. If average cash is yielding 4%, I want to do, say, 5%. I want to beat the average. And seeing as it is an average, it means I can.
  • jon3001
    jon3001 Posts: 890 Forumite
    xyy123 wrote: »
    I thought I was clear on this:

    i.e. If your benchmark is FTSE 100 (not a fantastic benchmark but adequate), you should have enough FTSE 100 stock in your portfolio, such that the performance is broadly speaking the same as the benchmark - an index tracker does this perfectly (almost). But if you want to try and add a little value and attempt to beat the benchmark, then you need to make a/some decisions. Maybe you'll underweight financials. Maybe you'll include a little foreign exposure. But the more you deviate, the more risk you take on in not meeting the objective you set out to meet, which has to tie in with the selected benchmark.

    If the FTSE 100 is 25% financials, 20% energy, 15% consumer staples, then thats the approximate weighting you should have in your portfolio. Or maybe you think you can add value, so you go 20% financials. Its a benchmark risk because if you're wrong and financials do better than the benchmark average, your portfolio will lag and it will take longer to reach your goal..

    In the scheme of things security selection (e.g. whether buy extra BP & Shell compared to HSBC and Barclays) has only a relatively minor role in portfolio returns. Most of the returns come from asset allocation (how much I put into UK Large-caps, small-caps, commodity futures, property, etc). A lot of energy spent focusing on benchmarks doesn't yield much compared to the bigger picture of blending uncorrelated assets.

    xyy123 wrote: »
    I don't think you read my post correctly - I said the last 10 years in bonds beat equities...

    Indeed you said that.

    xyy123 wrote: »
    I refer you again to this:

    Equities are historically the highest yielding asset class.

    Incorrect. Alternative assets such as commodity futures, property and private equity have all outperformed public equity over generational timescales. Which one will lead in the future noone knows. Hence the rationale behind a diversified portfolio consisting of uncorrelated assets.
  • xyy123
    xyy123 Posts: 61 Forumite
    jon3001 wrote: »
    In the scheme of things security selection (e.g. whether buy extra BP & Shell compared to HSBC and Barclays) has only a relatively minor role in portfolio returns. Most of the returns come from asset allocation (how much I put into UK Large-caps, small-caps, commodity futures, property, etc).
    Correct, except UK Large Cap/Small cap are sub-asset classes, not asset classes in their own right. I mentioned nothing about security selection.
    A lot of energy spent focusing on benchmarks doesn't yield much compared to the bigger picture of blending uncorrelated assets.
    Incorrect. I don't know how much clearer I can be here. A benchmark is a makeup of the underlying securities. The FTSE 100 is the 100 largest companies listed on the LSE by market capitalisation. If you want to follow the benchmark, you buy those 100 companies in the proportion they makeup the index (or an index tracker).

    As I say, the benchmark is your investment roadmap. Given amount to invest A, times expected rate of return, B, to the power of the number of years, n, will give you your terminal value, C, based on historic/projected returns. Of course its not exact, it never is. You find the benchmark/blend of benchmarks that will get you there with as little risk as possible. If your goal is to maximise the terminal value, you need to figure out what level of risk you are willing to accept. Your benchmark may be 100% equity, 100% gilts, 100% cash, it doesn't really matter.

    The point is you stick to your benchmark unless you're SURE its correct to deviate from it. E.g. Going from 100% equity to 100% is a massive benchmark risk if you're aiming for 10 year+ growth, because if you're wrong, you will have to buy in at a higher price and miss out on returns. The opposite is also true.
    Incorrect. Alternative assets such as commodity futures, property and private equity have all outperformed public equity over generational timescales. Which one will lead in the future noone knows. Hence the rationale behind a diversified portfolio consisting of uncorrelated assets.
    I'd like to see your evidence for this. Here's some of mine (please note tables 2 and 3 re: property):

    http://www.frontiercm.com/uploads/timehorizon.pdf

    As a side note I would also like to point out that many people who invest in property do not account for the ongoing costs like new guttering, new kitchens & bathrooms, redecorating, gardening, new driveways, home insurance, double glazing etc, to say nothing of heating, water, electric and council tax bills.

    I am also not convinced that commodity futures qualify as an asset class in their own right. What about futures in general? Or commodity options? Or swaps? How about credit default swaps or exotic bermuda swaptions? Are they asset classes? Also commodity futures have been around for about 30-odd years. I wouldn't have thought that was a long-enough timescale for relevant data. It would only mean 10 or so rolling-20 year periods or 20ish rolling-10 year periods.
  • turbobob
    turbobob Posts: 1,500 Forumite
    xyy123 wrote: »
    I am also not convinced that commodity futures qualify as an asset class in their own right. What about futures in general?

    Commodities are an asset class. I don't think futures could be considered an asset class in their own right - they are a derivative.
  • xyy123
    xyy123 Posts: 61 Forumite
    turbobob wrote: »
    Commodities are an asset class. I don't think futures could be considered an asset class in their own right - they are a derivative.
    Agreed. Commodities yes, commodity futures, no. Derivatives in general... arguable; their values are essentially derived from other asset classes so to say 'derivatives' is basically the average of all asset classes/instruments which is a bit pointless.
  • jon3001
    jon3001 Posts: 890 Forumite
    edited 13 November 2009 at 1:05AM
    xyy123 wrote: »
    I mentioned nothing about security selection
    xyy123 wrote: »
    If the FTSE 100 is 25% financials, 20% energy, 15% consumer staples, then thats the approximate weighting you should have in your portfolio. Or maybe you think you can add value, so you go 20% financials. Its a benchmark risk because if you're wrong and financials do better than the benchmark average, your portfolio will lag and it will take longer to reach your goal.

    By what mechanism are you proposing to deviate from the benchmark? When professional fund managers attempt to add value by over/underweighting sectors they engage in security selection.
    xyy123 wrote: »
    jon3001 wrote: »
    A lot of energy spent focusing on benchmarks doesn't yield much compared to the bigger picture of blending uncorrelated assets.
    Incorrect.

    I refer you to the study quoted by Dunstonh in this very thread:
    dunstonh wrote: »
    Brinson, Singer, Beebower (1991) found that the most dominant contributor to the variability of total portfolio returns is the asset allocation of the portfolio. 91.5% was down to asset allocation, 4.6% stock selection, 1.8% market timing and 2.1% - other.
    xyy123 wrote: »
    I don't know how much clearer I can be here. A benchmark is a makeup of the underlying securities. The FTSE 100 is the 100 largest companies listed on the LSE by market capitalisation. If you want to follow the benchmark, you buy those 100 companies in the proportion they makeup the index (or an index tracker).

    You mean if one wants a constituent asset class within the portfolio to follow a benchmark then that's the course of action.

    xyy123 wrote: »
    As I say, the benchmark is your investment roadmap.

    Nope - I focus on the bigger picture of asset allocation.
    xyy123 wrote: »
    Given amount to invest A, times expected rate of return, B, to the power of the number of years, n, will give you your terminal value, C, based on historic/projected returns. Of course its not exact, it never is. You find the benchmark/blend of benchmarks that will get you there with as little risk as possible. If your goal is to maximise the terminal value, you need to figure out what level of risk you are willing to accept.

    Your benchmark may be 100% equity, 100% gilts, 100% cash, it doesn't really matter.

    The point is you stick to your benchmark unless you're SURE its correct to deviate from it. E.g. Going from 100% equity to 100% is a massive benchmark risk if you're aiming for 10 year+ growth, because if you're wrong, you will have to buy in at a higher price and miss out on returns. The opposite is also true.

    You seemed to be focused on 100% this or 100% that. My point is that the optimal portfolio (the one that yields the greatest return for a desired amount of risk) is not 100% of any one asset class. That's the one offering the most reliable means of reaching your investment goal.
    xyy123 wrote: »
    jon3001 wrote: »
    Alternative assets such as commodity futures, property and private equity have all outperformed public equity over generational timescales. Which one will lead in the future noone knows. Hence the rationale behind a diversified portfolio consisting of uncorrelated assets.

    I'd like to see your evidence for this. Here's some of mine (please note tables 2 and 3 re: property):

    http://www.frontiercm.com/uploads/timehorizon.pdf

    That's good evidence. Your table 2 shows that equities were not the highest yielding asset class (it was commodities). Your table 3 shows that the multi-asset class portfolio (the strategy I espouse in this thread) is the one with the greatest return per unit of volatility and therefore consistently good returns.

    In terms of property you can look outside of the domestic market (e.g. the USA) if you want to see examples of what I mention. I get my data from books such as those by Roger Gibson or David Darst.
    xyy123 wrote: »
    As a side note I would also like to point out that many people who invest in property do not account for the ongoing costs like new guttering, new kitchens & bathrooms, redecorating, gardening, new driveways, home insurance, double glazing etc, to say nothing of heating, water, electric and council tax bills.

    The investments I refer to are of a different nature:
    • capital projects e.g buying land and building a retail park on it.
    • commercial lets (e.g. office space)
    Retail investment funds are readily available covering these.
    xyy123 wrote: »
    I am also not convinced that commodity futures qualify as an asset class in their own right. What about futures in general? Or commodity options? Or swaps? How about credit default swaps or exotic bermuda swaptions? Are they asset classes? Also commodity futures have been around for about 30-odd years. I wouldn't have thought that was a long-enough timescale for relevant data. It would only mean 10 or so rolling-20 year periods or 20ish rolling-10 year periods.

    Depends on what definition you subscribe to but for the practical purposes of building a portfolio such semantics are unimportant. If you don't think the futures (with returns not replicable by the underlying commodity) are not an asset class in their own right then you have plenty of historic data for the spot prices. Personally, having seen the data over multiple economic cycles, I'm happy to go ahead rather than wait 50 years.
  • xyy123
    xyy123 Posts: 61 Forumite
    edited 13 November 2009 at 9:39AM
    Sigh. I don't really know what to say. You just don't get it.
    jon3001 wrote: »
    By what mechanism are you proposing to deviate from the benchmark? When professional fund managers attempt to add value by over/underweighting sectors they engage in security selection.
    Yes. Or you could use ETFs.

    Asset Class: Equities
    Sub-Asset Classes: Sector, Geographic, Size, Growth/Value etc.
    Security Selection: BP or Royal Dutch

    If FTSE 100 is 25% Financials, you could simply have a 25% FTSE Financials ETF - no security selection involved. Or if you thought you should underweight/overweight them, you do it appropriately.
    I refer you to the study quoted by Dunstonh in this very thread:
    I am aware of the study. Enough to know it was originally actually by Brinson, Hood and Beebower, in 1986. I have referred to it myself in other threads.
    Nope - I focus on the bigger picture of asset allocation.
    Do you not understand that deciding on a benchmark IS deciding on asset allocation? i.e. what your default portfolio, by asset class, should look like over time to get you where you want to go? All the benchmark does, as I have said, is act as the map. So once you have decided on it, you don't want assets that are non-correlated to it, since it defeats the point. I suggest you actually go and read Determinants of Portfolio Performance, metioned above - be in the right asset class at the right time, not 6 different asset classes, only 1 of which can be correct.
    You seemed to be focused on 100% this or 100% that. My point is that the optimal portfolio (the one that yields the greatest return for a desired amount of risk) is not 100% of any one asset class. That's the one offering the most reliable means of reaching your investment goal.
    I never said it was. Although for 10 year+ time horizons and less than 2 year time horizons, it is. Static portfolios do not work.

    I give up trying to make you understand.
  • jon3001
    jon3001 Posts: 890 Forumite
    xyy123 wrote: »
    Do you not understand that deciding on a benchmark IS deciding on asset allocation?.

    Only if your portfolio consists of a single asset. Note that the reverse isn't true: deciding on an asset allocation is not deciding on a benchmark. Asset allocation comes first for MAC investors.
    xyy123 wrote: »
    All the benchmark does, as I have said, is act as the map. So once you have decided on it, you don't want assets that are non-correlated to it, since it defeats the point. I suggest you actually go and read Determinants of Portfolio Performance, metioned above - be in the right asset class at the right time, not 6 different asset classes, only 1 of which can be correct.

    It's well documented that professionals are rarely successful in timing entry/exit between markets.
    xyy123 wrote: »
    I never said it was. Although for 10 year+ time horizons and less than 2 year time horizons, it is. Static portfolios do not work.

    And yet the MAC portfolio in your table 3 gave good investment returns over a 17-year period with relatively low volatility. How is that not working?
    xyy123 wrote: »
    I give up trying to make you understand.

    I understand that building portfolios constisting of uncorrelated assets is a perfectly viable investment strategy.
  • xyy123
    xyy123 Posts: 61 Forumite
    jon3001 wrote: »
    Only if your portfolio consists of a single asset. Note that the reverse isn't true: deciding on an asset allocation is not deciding on a benchmark. Asset allocation comes first for MAC investors.
    Blended benchmarks are perfectly satisfactory, if the reason for selecting them is the right one.

    i.e. 70% equity: 30% gilt or 50:50 or 40:60 or 50% equity:40%gilt:10%cash.

    Selecting them to avoid short-term volatility is pointless as it ignores the original goal of investing.
    It's well documented that professionals are rarely successful in timing entry/exit between markets.
    Hence remaining true to the benchmark and deviating only when you're sure it is the right thing to do. Some investment/money managers have a documented track record of market timing, but yes, most are woeful.
    And yet the MAC portfolio in your table 3 gave good investment returns over a 17-year period with relatively low volatility. How is that not working?
    Bear in mind the end date for the analysis was at the near-lows of the worst bear market since the Depression, having been a historically awful performance in equities for the last 10 years.
    I understand that building portfolios constisting of uncorrelated assets is a perfectly viable investment strategy.
    Its a lazy way of constructing a portfolio, in the name of preventing losses rather than actually making money.
This discussion has been closed.
Meet your Ambassadors

🚀 Getting Started

Hi new member!

Our Getting Started Guide will help you get the most out of the Forum

Categories

  • All Categories
  • 352.2K Banking & Borrowing
  • 253.6K Reduce Debt & Boost Income
  • 454.3K Spending & Discounts
  • 245.3K Work, Benefits & Business
  • 600.9K Mortgages, Homes & Bills
  • 177.5K Life & Family
  • 259.1K Travel & Transport
  • 1.5M Hobbies & Leisure
  • 16K Discuss & Feedback
  • 37.7K Read-Only Boards

Is this how you want to be seen?

We see you are using a default avatar. It takes only a few seconds to pick a picture.