If past doesn't predict future, why is index investing good?

Blubbs wrote: ↑I am about to start a couch potato TFSA index portfolio, and the common thing I hear from people promoting index funds is that you shouldn't buy managed funds because they can't consistently beat the market and that past performance doesn't predict future results.

My question is: doesn't the same logic apply at least somewhat to index investing? The arguments for index investing always point to past performance; perhaps over a longer period of time, but still past performance. Is it not conceivable that the historical returns from this methods are not going to repeat themselves at all? Perhaps stock markets worldwide will change their patterns.

Then, the deeper question that follows is: what metrics do we have other than the past for predicting performance? Or is everything basically just gambling? In that case, I may as well use my money to own a business or something I have more personal control over, so that my skills and expertise can become factors in success.

Thoughts?

The stock market has high returns when compared to other asset classes. So you want to be IN the market. Index strategies such as coach potato is a cheap and easy way to participate while being diversified. Stock markets in general has better returns because the cash flows are discounted more heavily compared to other asset classes.

Of course nothing is a sure thing but indexing is a better bet as 95% of active managers have consistently underperformed their benchmark index for as long as there has been active management. Also, because of the fees they charge they actually have to exceed their benchmark performance pre their 2%+/- fees to even equal the index.

So, yes, perhaps perhaps, but I'd rather choose the bet that has better odds, wouldn't you?

Everything is probablistic. Even newton's laws of motion just happen to predict the most likely path an object will take. Fact is indexing has demonstratebly more stable outcome than actively managed funds. We can sit here and speculate all day as to why that's the case. My guess: indexing involves the whole market, and the large number of participants averages out various random events and instabilities, giving you something close to the theoretical growth rate of the underlying processes.

My question is: doesn't the same logic apply at least somewhat to index investing? The arguments for index investing always point to past performance; perhaps over a longer period of time, but still past performance. Is it not conceivable that the historical returns from this methods are not going to repeat themselves at all? Perhaps stock markets worldwide will change their patterns.

Past performance isn't evidence of future performance. You're right. But indexing doesn't go on that ideological (as far as I'm aware). If the market goes down 20% a year for 30 years, yeah your retirement is ruined... but the bigger picture is that the country is destroyed and people are reduced to bashing in skulls for food... returns don't matter much in that scenario.

If you believe societies economies will function, people will still want to make money and business will profit, then you can enter the realm of investing.

I may as well use my money to own a business or something I have more personal control over, so that my skills and expertise can become factors in success.

Sure. If you want to run a business go for it and pour your heart into it. There are a lot of ways to make money. Only you'll know what's best for you. A business requires work from you, investing requires a few clicks of a mouse. Do what makes you happy.

The past actually is a solid predictor of the future when circumstances are very similar.

Indexes are very representative of the market as a whole. Unless you think the stock market is going to change considerably, then future returns should be similar to past returns.

I can think of a few reasons why future returns may differ from past returns:
1) baby boomers retiring. But this really only applies to North America. This would have a negative effect on the stock market.
2) productivity is increasing. This is happening globally, and it is a result of increased automation and technology. This increases corporate profitability, so it has a positive effect on the stock market.
3) globalization/development of major economies. This will increase profits of international companies and multinational companies, so it will have a positive effect on those companies. Furthermore, more money in those countries will mean greater investment in the stock market. The US market being the largest, most diversified, and among the most stable will probably be just as poised to benefit from this as the markets based in those developing economies.
4) decreased birth rates. This probably means that inflation and interest rates will not be as high as they were in the past. This goes back to point #1. Baby boomers' coming of age meant lots of buying, and that probably drove up inflation and interest rates, especially in the 80s when both were particularly high. Some of that buying was investments, and interest rates and the performance of stocks are correlated. This has a negative effect on the stock market.
5) rising economic inequality. Wealthy folks contribute proportionally more of their income and net worth than lesser economic classes into the stock market. This has a positive effect on the stock market.
6) You could go on and on, but I actually think #1-3 are by far the most important considerations.

All in all, I think these differences more or less cancel out, though I think it will be more important than ever before to diversify globally. Don't think S&P500; think MSCI World.

EDIT: Rising public debt (specifically in the US) will have an interesting effect on the markets too, and I think it's yet another reason to diversify globally. That's probably more of an issue 20+ years down the road though.

It's undoubtedly difficult to predict the future, but, unless you have a solid reason to believe that the stock market will take a directional turn, then I think it's best to assume returns will stay more or less the same. It's easy to be turned off if you look at just one factor and overlook others. If you were turned off of stocks due to the possibility of nuclear war in the 1950s through to the 1990s, then you'd have missed out on some good returns.

The main argument for an index fund is that they are guaranteed to get the average return, by definition, and generally do so with cheaper fees than managed funds.

All managed funds claim that they are smart enougj to beat the market, but who knows.

Whatever you do, don't churn your investments. The biggest cost to most people is changing their minds, paying transaction fees to get out of one investment and into another is usually just a loss of fees.

Buy what suits your beliefs and then stick with it through thick and thin. Don't trade out when the market declines or flip flop between investment strategies.

Blubbs wrote: ↑The arguments for index investing always point to past performance; perhaps over a longer period of time, but still past performance.

I have not heard of any arguments in favour of index investing that have anything to do with past performance.

The arguments in favour of index investing are all mathematical, either plain arithmetic or basic probability.

Can you point out some arguments in favour of index investing that are based on past performance?

The arithmetic argument is that the average return of all actively managed dollars is the average market return, which is by definition the return of the index. Therefore, the expected return of an active investment is equal to the expected return of the index, not counting fees. The argument in favour of the index is that the fees (which are disclosed ahead of time) are almost always lower than for the active investment.

The probabilistic argument is that since indexing provides diversification, it gives you the same expected return with a lower risk. Here's an example. Say you have ten fair coins. For each $1 you bet on a coin, you get $3 if it comes up heads, or $0 if it comes up tails. You have $10. If you bet $10 all on one coin, your expected payout is $15 (50%*$0 + 50%*$30), but you have a 50% chance of ending up with less than the $10 you started with. If you bet $1 on each of the 10 coins, your expected payout is also $15, but the chance of ending up with less than your original $10 is only 17% (using the binomial distribution). If you diversify over more than 10 coins, your risk goes down even more while your expected payout stays the same.

Of course, if you "know" that one coin will always come up heads (using past performance or moving averages or technical analysis or crystal balls), you should forget indexing and just bet everything on that coin and get $30. But in order for that to work, you need to know for sure, and you need to know something that nobody else knows. If other people knew that the coin always pays out $3, they would drive up the price of that coin so it would cost $3 to bet on it instead of $1, and you wouldn't make any money by betting on it.

The past doesn't predict future isn't always 100% true. To quote the wealthy barber:

[QUOTE]"If you were assigned to select the best QB in the NFL but didn't follow football, where would you start? You'd look at statistics, of course. Who is the highest-rated passer? Who's the most consistent? Has he been good for just one year, or has he proven he can do it ear after year? When times are tough and his team is struggling, does he maintain a high level of play or drop off badly? Once you had studied the statistics, you would read some articles and talk to people in the know. What are the experts' opinions? Do they jibe with your findings? It's no different when you're looking for a good mutual fund. Assess past records. They don't guarantee future results, but they sure as heck are a good indication of the manager's abilities. What's the five-year average return? Ten-year? Fifteen-year? Does the fund perform consistently, or is it way up one year and way down the next? How does it fare during the bad times? Some funds have excellent records during tough times.[/QUOTE]

theBasher91 wrote: ↑The past doesn't predict future isn't always 100% true. To quote the wealthy barber:

https://math.temple.edu/~paulos/scam.html

The problem with predicting future returns based on an actively managed fund's past performance is that that fund changes over time, and it changes in unpredictable ways. A lot of this is done simply to match or beat past returns, a nearly impossible task in some conditions. In making those changes, past circumstances are not the same as current/future circumstances, so past returns are not likely to be predictive of future returns.

An index changes very slowly over time and does so in predictable ways. It changes, yes, but it changes in the very same ways that it changed in the past. As a result, the future performance should be very similar for that reason; any variability would be a result of extrinsic factors, i.e. broad market and economic changes.

Blubbs wrote: ↑My question is: doesn't the same logic apply at least somewhat to index investing? The arguments for index investing always point to past performance; perhaps over a longer period of time, but still past performance. Is it not conceivable that the historical returns from this methods are not going to repeat themselves at all? Perhaps stock markets worldwide will change their patterns.

Then, the deeper question that follows is: what metrics do we have other than the past for predicting performance? Or is everything basically just gambling? In that case, I may as well use my money to own a business or something I have more personal control over, so that my skills and expertise can become factors in success.

You've got it a little mixed up.

Index investing isn't really based on past performance indicating future performance. Index investing is based on the idea that the market, over time, will always go up. This is a pretty safe bet, because if the market ends up decreasing over time we're looking at a catastrophic collapse to the world economy. Since nothing matters in that case (aside from having a well stocked bunker with lots of dehydrated food) there's no point wasting time worrying about it and we should focus on the alternative: that the market will increase over time.

Going forward with that assumption, we now have to choose where to invest our money.

Option 1: Index investing. This will pretty much mimic the market's average returns, minus a small MER (<1%).

Option 2: Mutual funds. This will pretty much mimic the market's average returns, minus a hefty MER (>1%). This is where the caveat past performance doesn't guarantee future returns comes in. It's definitely possible that some mutual funds are exceptionally well managed and will consistently beat the market average returns, but just because a fund made some good guesses in the last 10 years doesn't mean they'll keep making good guesses for the next 10 years.

Option 3: Buy individual stocks. You can do this if you'd like, but on average you'll be looking at market average returns and there's a chance your entire portfolio goes to near 0 due to poor stock picking and poor diversification. I wouldn't recommend this under any circumstance, but I don't take these kind of risks with my retirement money, so YMMV.