I was thinking of starting a thread about how to construct and manage a passive long-term investment portfolio. I think this topic is important to traders (or anyone with spare cash & income) but it seems to have very little discussion here on ET. I don't want to make a long series of posts if no one would want to read them, so I thought I would start a poll to gauge the level of interest. If enough people are interested, I'll give my thoughts on the subject, and invite contributions/questions - hopefully we can get a good thread going.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Passive long-term investment - who wants to discuss it?
- Thread starter Cutten
- Start date
- Status
- Not open for further replies.
By all means, lets do it.
I would love to hear your thoughts on the matter ... I think passive long-term portfolios should be well diversified and designed to catch the macro moves ... sounds kinda simple, but most folks aint to good at it.
I would love to hear your thoughts on the matter ... I think passive long-term portfolios should be well diversified and designed to catch the macro moves ... sounds kinda simple, but most folks aint to good at it.
Ok, so far there are only a few replies (not surprising for a Sunday) but all are at least partly positive, so I'll go ahead. I don't make any promises at regular updates on this thread, I'l just post as and when the mood takes me, but you are all welcome to chip in.
So, we're gonna talk passive long-term investment. Let's start by explaining what that means exactly:
1) It's investment. That means that any kind of technical analysis, trading, chart-reading, big betting and so on is out; adding to losers (within reason), buy & hold, following valuation instead of price is in. Forget your trader side when discussing this subject. Investment requires either a decent lump-sum, or regular contributions from income until your portfolio becomes a decent lump sum. Almost all able-bodied adults in the developed world have the ability to earn enough to invest. Even $200 per month is enough to make a meaningful difference by retirement age. Investment is pretty democratic in this respect - almost all of you reading this can retire with $1 mill+ if you are young enough (if you are 60 and broke, it's probably not gonna happen), can work, and have some personal discipline.
2) It's passive. Due to the terrible record of the universe of active managers and individual investors who use discretion/market-timing, I am going to recommend a purely passive approach based on low-cost, tax efficient exchange-traded funds (ETFs). Passive invesment has many benefits: i) it requires no market insight at all, other than understanding basic invesment concepts (which you will once you read the full thread) ii) it is low cost - ETFs have lower fees than active funds iii) it is tax efficient, ETFs incur lower capital gains than active funds, and even passive funds pay more in tax usually iv) it performs well - index-tracking ETFs outperform most active managers, and almost all individual investors, over the long-run v) it is low stress, and simple to do.
3) It's long-term. The average person cannot get rich quickly by sensible investing. High returns require either lots of luck, lots of risk, or truly special abilities (e.g. Buffett, Soros). Much as we'd like to be like them, the odds are against it. The beauty of investing though, is that over a long enough time period, any Joe Sixpack can become pretty rich by most standards. Long-term investing is possibly the only way that an average working stiff can practically *guarantee* to retire a millionaire. Seriously, I can think of almost nothing in life that is such a great sure thing for the typical guy or gal in the street. All it takes is a little work, some discipline, some patience and consistency.
4) It's conservative. A well-constructed investment portfolio has almost zero chance of massive loss of capital. As an example, a model portfolio I use as a reference would have lost at most 15% during the 2000-2003 bear market, and that had 50% in stocks. It's performance from Oct 2007 to the March 2008 lows was down around 9%, compared to down 20% for the S&P. If you get enough diversification, and use enough low or negative correlation assets, then a portfolio can be surprisingly robust. Many people are wedded to the idea of S&P or bust - which is IMO way too risky for most people; or something like 70% S&P, 30% bonds, which is better, but still a lot more risky than is necessary. I'll explain how you can easily beat cash returns, without taking all that much risk - 10-15% is the most you should lose from peak to trough during even a fairly severe bear market. A once per century perfect storm like the Great Depression would be worse, but even then you would lose no way near as much, and would vastly outperform almost all other stock/bond portfolios.
So, that's a quick description of the approach I'm going to take: passive, conservative, long-term investment. Now let's get into the details...
So, we're gonna talk passive long-term investment. Let's start by explaining what that means exactly:
1) It's investment. That means that any kind of technical analysis, trading, chart-reading, big betting and so on is out; adding to losers (within reason), buy & hold, following valuation instead of price is in. Forget your trader side when discussing this subject. Investment requires either a decent lump-sum, or regular contributions from income until your portfolio becomes a decent lump sum. Almost all able-bodied adults in the developed world have the ability to earn enough to invest. Even $200 per month is enough to make a meaningful difference by retirement age. Investment is pretty democratic in this respect - almost all of you reading this can retire with $1 mill+ if you are young enough (if you are 60 and broke, it's probably not gonna happen), can work, and have some personal discipline.
2) It's passive. Due to the terrible record of the universe of active managers and individual investors who use discretion/market-timing, I am going to recommend a purely passive approach based on low-cost, tax efficient exchange-traded funds (ETFs). Passive invesment has many benefits: i) it requires no market insight at all, other than understanding basic invesment concepts (which you will once you read the full thread) ii) it is low cost - ETFs have lower fees than active funds iii) it is tax efficient, ETFs incur lower capital gains than active funds, and even passive funds pay more in tax usually iv) it performs well - index-tracking ETFs outperform most active managers, and almost all individual investors, over the long-run v) it is low stress, and simple to do.
3) It's long-term. The average person cannot get rich quickly by sensible investing. High returns require either lots of luck, lots of risk, or truly special abilities (e.g. Buffett, Soros). Much as we'd like to be like them, the odds are against it. The beauty of investing though, is that over a long enough time period, any Joe Sixpack can become pretty rich by most standards. Long-term investing is possibly the only way that an average working stiff can practically *guarantee* to retire a millionaire. Seriously, I can think of almost nothing in life that is such a great sure thing for the typical guy or gal in the street. All it takes is a little work, some discipline, some patience and consistency.
4) It's conservative. A well-constructed investment portfolio has almost zero chance of massive loss of capital. As an example, a model portfolio I use as a reference would have lost at most 15% during the 2000-2003 bear market, and that had 50% in stocks. It's performance from Oct 2007 to the March 2008 lows was down around 9%, compared to down 20% for the S&P. If you get enough diversification, and use enough low or negative correlation assets, then a portfolio can be surprisingly robust. Many people are wedded to the idea of S&P or bust - which is IMO way too risky for most people; or something like 70% S&P, 30% bonds, which is better, but still a lot more risky than is necessary. I'll explain how you can easily beat cash returns, without taking all that much risk - 10-15% is the most you should lose from peak to trough during even a fairly severe bear market. A once per century perfect storm like the Great Depression would be worse, but even then you would lose no way near as much, and would vastly outperform almost all other stock/bond portfolios.
So, that's a quick description of the approach I'm going to take: passive, conservative, long-term investment. Now let's get into the details...
Investment theory:
I'm going to start with some investment theory. Nothing too esoteric or ivory tower, but it's important to understand the general principles which govern investment assets and returns. It gives greater understanding, and gives you the confidence to stick with a portfolio and/or system when the market gives you one of its inevitable shocks to the system.
What makes something an investment asset?
Firstly, let's take the critical distinction between an investment asset, and a trading or speculative position. And investment asset is one that generates an *inherent long-run return*. A speculative or trading position (such as holding S&P futures for 2 hours) has a slightly negative inherent return (zero-sum, minus transactions costs). For example, bonds pay a coupon - you buy a 10 year treasury at a 5% yield, and it's pretty clear why you have an inherent return. T-bills have a near-certain return, at least over the short-run. Stocks have a less certain but still probable return, based on corporate profits and dividends. Real estate has a return based on inherent value (replacement cost of bricks & mortar, and land), and rental income. Trade commodities, such as sugar, wheat etc do not have inherent returns - they have storage costs, and a bunch of wheat sitting in a grain silo does not have any kind of economic profit accumulating each month. You profit form these trading or speculative assets purely by exploiting price changes or carry (in the rare cases it is positive), not from their inherent returns.
Investment assets have inherent long-run returns due to their natural economic characteristics. For example, a borrower must pay a premium to incentivise a lender to give money - no one would lend money for nothing. Thus, bonds will (almost) always pay yield. Stocks are claims on corporate assets and earnings streams, so as long as a corporation remains intact and profitable, they will accumulate money year over year. Real estate will grow as rents, incomes, and purchasing power rises. And so on.
Thus, investment portfolios should be made up of investment assets, for the most part. Options, trading assets, speculative vehicles and so on should generally be avoided. Your long-term portfolio return is generally going to come from steady accumulation of the inherent return of your portfolio assets, rather than from buying and selling in and out, making speculative bets, and the like.
Which brings us to the next subject - expected returns...
I'm going to start with some investment theory. Nothing too esoteric or ivory tower, but it's important to understand the general principles which govern investment assets and returns. It gives greater understanding, and gives you the confidence to stick with a portfolio and/or system when the market gives you one of its inevitable shocks to the system.
What makes something an investment asset?
Firstly, let's take the critical distinction between an investment asset, and a trading or speculative position. And investment asset is one that generates an *inherent long-run return*. A speculative or trading position (such as holding S&P futures for 2 hours) has a slightly negative inherent return (zero-sum, minus transactions costs). For example, bonds pay a coupon - you buy a 10 year treasury at a 5% yield, and it's pretty clear why you have an inherent return. T-bills have a near-certain return, at least over the short-run. Stocks have a less certain but still probable return, based on corporate profits and dividends. Real estate has a return based on inherent value (replacement cost of bricks & mortar, and land), and rental income. Trade commodities, such as sugar, wheat etc do not have inherent returns - they have storage costs, and a bunch of wheat sitting in a grain silo does not have any kind of economic profit accumulating each month. You profit form these trading or speculative assets purely by exploiting price changes or carry (in the rare cases it is positive), not from their inherent returns.
Investment assets have inherent long-run returns due to their natural economic characteristics. For example, a borrower must pay a premium to incentivise a lender to give money - no one would lend money for nothing. Thus, bonds will (almost) always pay yield. Stocks are claims on corporate assets and earnings streams, so as long as a corporation remains intact and profitable, they will accumulate money year over year. Real estate will grow as rents, incomes, and purchasing power rises. And so on.
Thus, investment portfolios should be made up of investment assets, for the most part. Options, trading assets, speculative vehicles and so on should generally be avoided. Your long-term portfolio return is generally going to come from steady accumulation of the inherent return of your portfolio assets, rather than from buying and selling in and out, making speculative bets, and the like.
Which brings us to the next subject - expected returns...
Risk and expected returns:
A fundamental tenet of investment theory is that returns are there to compensate investors for taking on risk. Intuitively this makes sense - if someone comes to you with a startup business, it is a lot more risky than if you lend money to the US government for 30 days. You would naturally require a much higher return to invest than you would expect from government T-bills.
Risk can be defined as the chance of loss. Conventionally it is represented by downside price volatility. Buffett and others have argued that downside volatility in the short-term is not true risk e.g. if I buy a box full of $10,000 in cash which will open in 3 years time, I suffer no permanent economic loss just because I can only find a bidder this year for $5000 as long as I am not compelled to sell (e.g. if I need the cash in a hurry). This is correct. However, for most investors who cannot be certain that their assessment of economic value is better than the market (i.e. 99% of us), market quotes are generally the best estimate of value, most of the time.
That's the theory. In practise, actual market returns over the long-run do seem closely correlated to risk. T-bills over the long run yield very little after inflation. Bonds yield more than t-bills but less than other riskier assets. Real estate returns more than bonds but less than equities. Large cap stocks return less over the long run than small cap stocks. Junk bonds yield more than AAA bonds, and so on. So, given that both theory and 100 years of market data suggests risk is correlated with long-run returns, I think it's fair to accept this as pretty much proven.
Thus, for an investor, to earn good long-run returns, you are going to have to take some risk. You will not accumulate a large sum in 10-20 years by parking your assets in cash, money market, or 2 year treasuries. Even long-term bonds will not return that much in most cases (occasionally they might e.g. if you bought in the early 1980s at 15% yields). Thus, high return assets should be sought out to maximise long-run returns.
The flip side is that these high return assets, experience and theory suggest, will have higher risk. The Dow Jones fell 89% from 1929 to 1932, and 50% in 1973-74. The S&P 500 fell almost 50% from 2000-2002. The nasdaq fell around 80% over the same period. Recently, the S&P fell 20% in a mere 3 months. Real estate, often considered a conservative asset class, has been hammered in the last year or so. Even bonds got hit hard in the inflationary 1970s.
So - as investors we generally want high return assets. However, at the same time we want to reduce risk. These are inherently conflicting goals. Is there some way to reach the holy grail of good returns with moderate risk? Luckily there is a way to at least get closer to that goal. Not only that, but the even better news is that one can even *profit* from the risk and volatility of investment assets.
How to reduce risk, without reducing returns just as much? By using one of the few free lunches in the markets - diversification. Read on to find out more...
A fundamental tenet of investment theory is that returns are there to compensate investors for taking on risk. Intuitively this makes sense - if someone comes to you with a startup business, it is a lot more risky than if you lend money to the US government for 30 days. You would naturally require a much higher return to invest than you would expect from government T-bills.
Risk can be defined as the chance of loss. Conventionally it is represented by downside price volatility. Buffett and others have argued that downside volatility in the short-term is not true risk e.g. if I buy a box full of $10,000 in cash which will open in 3 years time, I suffer no permanent economic loss just because I can only find a bidder this year for $5000 as long as I am not compelled to sell (e.g. if I need the cash in a hurry). This is correct. However, for most investors who cannot be certain that their assessment of economic value is better than the market (i.e. 99% of us), market quotes are generally the best estimate of value, most of the time.
That's the theory. In practise, actual market returns over the long-run do seem closely correlated to risk. T-bills over the long run yield very little after inflation. Bonds yield more than t-bills but less than other riskier assets. Real estate returns more than bonds but less than equities. Large cap stocks return less over the long run than small cap stocks. Junk bonds yield more than AAA bonds, and so on. So, given that both theory and 100 years of market data suggests risk is correlated with long-run returns, I think it's fair to accept this as pretty much proven.
Thus, for an investor, to earn good long-run returns, you are going to have to take some risk. You will not accumulate a large sum in 10-20 years by parking your assets in cash, money market, or 2 year treasuries. Even long-term bonds will not return that much in most cases (occasionally they might e.g. if you bought in the early 1980s at 15% yields). Thus, high return assets should be sought out to maximise long-run returns.
The flip side is that these high return assets, experience and theory suggest, will have higher risk. The Dow Jones fell 89% from 1929 to 1932, and 50% in 1973-74. The S&P 500 fell almost 50% from 2000-2002. The nasdaq fell around 80% over the same period. Recently, the S&P fell 20% in a mere 3 months. Real estate, often considered a conservative asset class, has been hammered in the last year or so. Even bonds got hit hard in the inflationary 1970s.
So - as investors we generally want high return assets. However, at the same time we want to reduce risk. These are inherently conflicting goals. Is there some way to reach the holy grail of good returns with moderate risk? Luckily there is a way to at least get closer to that goal. Not only that, but the even better news is that one can even *profit* from the risk and volatility of investment assets.
How to reduce risk, without reducing returns just as much? By using one of the few free lunches in the markets - diversification. Read on to find out more...
Quote from gkishot:
Question: you don't believe that commodities appreciate inherently over the long term ?
I believe over the long-term they will broadly rise in line with inflation plus a modest risk premium, minus their carrying/storage charges, other things being equal. The historical data I have seen seems to bear that out. Gold over the long-term underperforms US bonds, stocks, real estate, and even t-bills. It basically keeps its value but does not provide a surplus return above inflation.
Diversification:
Portfolio theory shows that diversification between assets that are not perfectly correlated will increase returns for a given level of risk, or will reduce risk for a given level of returns. For example, let's take two stock markets that return 10% over the long-run. If they are not perfectly correlated, a portfolio of 50% long stockmarket A and 50% long stockmarket B will return the same and have less risk than a portfolio of 100% long A and 0% long B. I won't go into the detailed theory, but this can be proven mathematically, and is confirmed by actual data from the markets over the last century or so. This is the same principle casinos use with roulette tables. A dozen tables risking $1k max per spin will have much less risk than 1 stable risking $12k per spin. This principle works basically because in some years one asset will be up while the other is down. It also provides valuable disaster insurance in cases like the Great Depression, 9/11, WWII etc. It is obvious in common sense terms too - a portfolio of 20 stocks is far less likely to hit disaster than if you have everything in one stock.
Now, there are some even more desireable qualities of diversification. For example, a portfolio of low risk assets plus a small amount of high risk but <100% correlated assets not only produces a higher return than a pure low-risk asset portfolio, but it actually has *lower risk*. That's right - a 100% bond portfolio has lower return AND higher risk than a 90% bond, 10% stock portfolio. Widows and orphans, even if they want to purely minimise risk, should have at least some stock allocation. This may sound counterintuitive, but both finance theory/maths, and empirical data from the last 80 years or so back it up. The reason is simply that stocks are not 100% correlated to bonds - some years when bonds get killed, stocks will do well. Having lower risk and higher returns is a genuine free lunch. That's the power of diversification.
Another quality of diversification is that the more different asset classes you add, the stronger its effect. A stock/bond portfolio will have more risk than a stock/bond/real estate portfolio, for the same expected long-run return. This means that as investors we ought to seek out as many uncorrelated, positive-return investment asset classes as possible, provided they are suitable for passive long-run investment.
This brings us onto the topic of suitable investment asset classes...
Portfolio theory shows that diversification between assets that are not perfectly correlated will increase returns for a given level of risk, or will reduce risk for a given level of returns. For example, let's take two stock markets that return 10% over the long-run. If they are not perfectly correlated, a portfolio of 50% long stockmarket A and 50% long stockmarket B will return the same and have less risk than a portfolio of 100% long A and 0% long B. I won't go into the detailed theory, but this can be proven mathematically, and is confirmed by actual data from the markets over the last century or so. This is the same principle casinos use with roulette tables. A dozen tables risking $1k max per spin will have much less risk than 1 stable risking $12k per spin. This principle works basically because in some years one asset will be up while the other is down. It also provides valuable disaster insurance in cases like the Great Depression, 9/11, WWII etc. It is obvious in common sense terms too - a portfolio of 20 stocks is far less likely to hit disaster than if you have everything in one stock.
Now, there are some even more desireable qualities of diversification. For example, a portfolio of low risk assets plus a small amount of high risk but <100% correlated assets not only produces a higher return than a pure low-risk asset portfolio, but it actually has *lower risk*. That's right - a 100% bond portfolio has lower return AND higher risk than a 90% bond, 10% stock portfolio. Widows and orphans, even if they want to purely minimise risk, should have at least some stock allocation. This may sound counterintuitive, but both finance theory/maths, and empirical data from the last 80 years or so back it up. The reason is simply that stocks are not 100% correlated to bonds - some years when bonds get killed, stocks will do well. Having lower risk and higher returns is a genuine free lunch. That's the power of diversification.
Another quality of diversification is that the more different asset classes you add, the stronger its effect. A stock/bond portfolio will have more risk than a stock/bond/real estate portfolio, for the same expected long-run return. This means that as investors we ought to seek out as many uncorrelated, positive-return investment asset classes as possible, provided they are suitable for passive long-run investment.
This brings us onto the topic of suitable investment asset classes...
- Status
- Not open for further replies.