Indexing vs Individual Stocks

[scottish]

I suspect this agility is often a reason *not* to purchase individual stocks.   When a company's share price declines, you will be much more likely to sell your shares, i.e. buy high and sell low.

I find the whole index / individual stock debate difficult.

Index proponents say 'Just buy the index'!   Well which one?    If you're American buy the S&P 500.   If your in Oz, but the Australian index, and in Canada buy the Canadian index?

And then people view the index as the average of the stock market.   But it's not exactly the average of the stock market.   It's a basket of stocks selected by a committee of financiers whose objective is supposed to be to build something representative of the stock market.   Getting your company into a major index is worth a fortune to the owners.   So there's bound to be some influence peddling going on.

Comparing index investing to active fund investing is easy.   Mutual fund managers aren't any smarter than anyone else, after all, so if you're coughing up a management fee you're not going to do as well.  And hard data is easy to find - it's all online.

But comparing index investing to individual stocks is much harder.   There's no public data on the portfolios of individual stock investors.  It seems reasonable that most people won't/can't do the work to stay on top of a portfolio of individual stocks, and it's much easier to just buy index funds.   But clear data?  I didn't see any in the earlier links. 

Has anyone tried to compare a portfolio of individual stocks to the index?  I did, it's bloody hard!  You have to track dividends and stock splits and companies that were acquired and companies that went out of business.

[cjottawa]

While I'm sure most here lean towards indexing, I was wondering if there are others who also invest in individual stocks. Value Investing (the approach used by Buffett) seems to have a lot of overlap with FI so I wouldn't be surprised to see some here that search for stocks using that approach.

Personally, I'm 50% indexed and 50% in individual stocks and am trying to build up my cash position to take advantage of any attractive opportunities that present themselves in the future. I've been teaching myself finance, valuation, and accounting over the past year and it's been a really eye opening experience. If you stay disciplined and have a knack for it, I think you can do pretty well. Even if it's just a percent or two above what the S&P does, that could be hundreds of thousands of dollars in compounded gains down the line.

What do you look for with individual stocks, and have you had success?

First, apologies if I'm repeating anything that's already been written in thread; I haven't read all the replies.

Like you, I used to hold a bunch of individual dividend paying stocks and index funds. I realized all I was doing was "tilting" my portfolio heavily toward those companies; they were already held in the index. I've since sold off the individual stocks; coincidentally, some of them have been battered pretty heavily since then. No such stress with the index.

http://www.efficientfrontier.com/ef/900/15st.htm

Holding individual stocks seems like it just makes a portfolio significantly more risky while reducing risk-adjusted return.

...Index proponents say 'Just buy the index'!   Well which one?    If you're American buy the S&P 500.   If your in Oz, but the Australian index, and in Canada buy the Canadian index?

And then people view the index as the average of the stock market.   But it's not exactly the average of the stock market.   It's a basket of stocks selected by a committee of financiers whose objective is supposed to be to build something representative of the stock market.   Getting your company into a major index is worth a fortune to the owners.   So there's bound to be some influence peddling going on...

This is covered quite extensively in writing by Jack Bogle, Rick Ferri, and William Bernstein.

"Equity" and "fixed income" are asset class; you diversify across asset classes, and within them by holding multiple indexes. (S&P500, S&P/TSX, MSCI EAFE for equity, DEX Universe bond, if you're a Canadian). Japan is a cautionary tale to us all: many there had all their eggs in one, domestic basket when their market tanked (decades ago) and they haven't recovered. Diversify internationally.

As for who decides what goes into the index, that point is addressed quite well in books by William Bernstein (Four Pillars of Investing) and Jack Bogle. See this article:

http://www.investorhome.com/W5000.htm

From it:

Bogle compares the S&P 500 and total market returns back to 1926 using data from the Center for Research in Security Prices (CRSP). He concludes that for the full period the S&P500 return was 10.4% versus 10.2% for the Total Stock Market Index. Starting from 1930 both returned 9.9%, and recently (from 1998-2006) the Total Market Index outperformed the 500 by 1%.

There are added "frictional" costs of managing an additional 4500 stocks in a total stock market fund and those stocks only account for a small amount of market capitalization compared to the top 500. Go with a total stock market (e.g Wilshire 5000) fund if you like but the S&P500 is "good enough." (Perfect being the enemy of good)

[scottish]

thanks for reminding me about Bernstein's book.   I'll go read this again and see if I can check out their claims in detail.

[adamwoods137]

Here's an interesting question I have for the indexers.  Is it possible for an individual investor to consistently do (what I mean here is to have an expected value) worse than the market before transaction costs?

Second, who is likely to do better: someone who buys an index fund or someone who buys 50 to 500 stocks at random in that index and only sits on them for the period?

[waltworks]

Is this a serious question? Individual investors can, and have, lost everything! It is certainly possible to end up with all sorts of results if you're going to pick stocks. In general, you'll probably do worse than indexing, but you might get lucky/be incredibly good and do better. The longer you do it, the worse your odds of beating the index with luck get, of course.

If you buy 50 stocks at random, you are buying a mini-index. The bigger the pot of stocks, the more closely it will track the entire market. With only 50, you'd have quite a bit more volatility (and risk) but the expected result, assuming you bought them randomly, would be the same as the market as a whole, not counting fees and taxes, of course.

If what you are asking is if you can create your own index, the answer is "sure!" but it's going to cost a lot more and take a lot more time/effort. There is probably an index product out there that has what you want, just start poring over the Vanguard funds.

-W

Here's an interesting question I have for the indexers.  Is it possible for an individual investor to consistently do (what I mean here is to have an expected value) worse than the market before transaction costs?

Second, who is likely to do better: someone who buys an index fund or someone who buys 50 to 500 stocks at random in that index and only sits on them for the period?

[beltim]

If you buy 50 stocks at random, you are buying a mini-index. The bigger the pot of stocks, the more closely it will track the entire market. With only 50, you'd have quite a bit more volatility (and risk) but the expected result, assuming you bought them randomly, would be the same as the market as a whole, not counting fees and taxes, of course.

Some studies argue that you can get ~90% of the benefits of diversification with as small a portfolio as 12-18 stocks:
http://news.morningstar.com/classroom2/course.asp?docId=145385&page=4&CN
Others say you need 30:
https://www.wiso.uni-hamburg.de/fileadmin/sozialoekonomie/bwl/bassen/Lehre/International_Finance_I/Lectures/20080506_Number_of_stocks_in_a_diversified_portfolio.pdf
or even 100:
http://www.aaii.com/journal/article/how-many-stocks-do-you-need-to-be-diversified-.touch

[beltim]

To answer Adam's question though, we'd need to know the distribution of stock returns within an index, over a given holding period.  For example, if the distribution of returns of S&P 500 components is right skewed, then a relatively small number of components are responsible for a disproportionate fraction of the index return, and a random sample of the components would be more likely to trail the index.  In contrast, if the distribution is left skewed, then a random sample would be more likely to beat the index.  In each case, though, the average of all such portfolios would equal the index return.

[adamwoods137]

Is this a serious question? Individual investors can, and have, lost everything! It is certainly possible to end up with all sorts of results if you're going to pick stocks. In general, you'll probably do worse than indexing, but you might get lucky/be incredibly good and do better. The longer you do it, the worse your odds of beating the index with luck get, of course.

How can you square that with the efficient market hypothesis?  If that were true, couldn't I outperform by just finding an individual investor and shorting what he does while going long an index?

If you buy 50 stocks at random, you are buying a mini-index. The bigger the pot of stocks, the more closely it will track the entire market. With only 50, you'd have quite a bit more volatility (and risk) but the expected result, assuming you bought them randomly, would be the same as the market as a whole, not counting fees and taxes, of course.

If what you are asking is if you can create your own index, the answer is "sure!" but it's going to cost a lot more and take a lot more time/effort. There is probably an index product out there that has what you want, just start poring over the Vanguard funds.

-W

According to the Efficient Market Hypothesis the person who bought and sat would probably OUTperform the index.  The marginal benefits of diversification (volatility reduction) drop quite substantially after the first 20 equities or something.  Transaction costs, taxes, and index frontrunning would lower the relative returns of the buy and hold portfolio.  For an example see ING Corporate Leaders Trust Fund Series B.  Now its possible that this mutual fund was simply lucky for 80 years, but buying 30 stocks and doing literally nothing outperformed the S&P 500 since the inception. $10000 invested in the trust in November 1935 paying an expense ratio of 0.46%(!) would have turned into $32 Million.   The S&P 500 with dividends reinvested (and NO FEES) would have only resulted in 29 Million.  This despite the fact that the 30 stocks selected were "Corporate Leaders", large cap stocks that had the small cap premium work against them over the period. 

I say this simply to point out that individual investors can reliably beat the market.  They just need to lose the passwords to their sharebuilder account after their initial purchases.  When people suggest that they are buying large dividend payers to hold (something I don't do), this is precisely what they *intend* to do.  The S&P 500's turnover is something like 3%.  IF the folks who pursue these sort of buy and hold strategies stick to their strategy, their annual turnover will be substantially less.  You might be inclined to claim that such "active" investors might be tempted to sell their losers and thereby screw up this plan.  This is certainly possible, but its not fair to say that someone who buys an index won't do something similar. 

Follow up question.  Don't hedge fund managers outperform their indexes *before* fees?  Hedge fund manager fee's are typically huge!  Doesn't this disprove the efficient market hypothesis?  A typical hedge fund manager fee structure is 2&20.  If the market does 10%, to market perform after fees the hedge fund manager has to do 15% (!).  If skill and hard work are adding 5 points to returns how do you square that with EMH? 

Here's an interesting question I have for the indexers.  Is it possible for an individual investor to consistently do (what I mean here is to have an expected value) worse than the market before transaction costs?

Second, who is likely to do better: someone who buys an index fund or someone who buys 50 to 500 stocks at random in that index and only sits on them for the period?

[Wolf359]

There was a study that found that Jim Cramer's picks are neither good nor bad. If he's an expert and he can't beat the indexes then how would you expect to do any better?

Anyone that knows anything doesn't consider Cramer an expert, he's a tv personality.

Before you dismiss Jim Cramer, let me quote him:

"You want a cheap, low-cost index fund that mirrors the market as a whole, one that mimics the S&P 500. You have a vehicle that will let you participate in the strength of the market without spending the time picking individual stocks. This may sound like a really simple solution, but don't overthink it. The whole point of putting your money in the fund is to save you from time and effort to manage your own portfolio of stocks."

If you want to watch it to see it in context, go here: http://www.cnbc.com/id/101782773
August 22, 2014 episode on CNBC. 

Cramer is a fan of index funds!

[adamwoods137]

To answer Adam's question though, we'd need to know the distribution of stock returns within an index, over a given holding period.  For example, if the distribution of returns of S&P 500 components is right skewed, then a relatively small number of components are responsible for a disproportionate fraction of the index return, and a random sample of the components would be more likely to trail the index.  In contrast, if the distribution is left skewed, then a random sample would be more likely to beat the index.  In each case, though, the average of all such portfolios would equal the index return.

This is a really good point. I am assuming that market returns are normally distributed.  But again I'm primarily concerned with the expected value of the hypothetical portfolio, rather than the median value.  Maybe the median is more important.

[FastStache]

You guys do realize that indexing = aiming for average return. This means you get to look down on 50% of the investors and think you are doing well while forgetting the other 50% who beat your returns handily.

If you're looking at one particular year, sure.  But over the long term, you're beating 50% of all invested dollars this year, then beating 50% of all invested dollars next year, and the year after that...every single year for the rest of your life.

If you're trying to beat this, even one bad year can make it almost impossible to catch up. Losing 50% one year, means you need to gain 100% next year just to break even. If you do 10% better than the index one year, then do 10% worse than the index the next year, you will still end up with less money than simply choosing the index.

It seems counterintuitive, but the opposite has the same result. Do 10% worse than the index one year, then 10% better than the index the next year, and you are still behind the index.

When looking at the long term, you're beating significantly more than 50%, you're beating almost everybody. The compounding effects of bad years make it incredibly difficult to beat the index long-term.

The math is very flawed here as you completely ignore compounding interest in the previous years.

[waltworks]

I think hedge fund managers generally underperform the index by a small amount but that's not a great comparison - the original premise of hedge funds was slightly lower returns but protection against catastrophic crashes (hence "hedge"). People seem to now use the term to mean any actively managed fund.

If you are talking just mutual fund managers, the evidence is mixed (lots of academic literature is available) but overall no, they don't beat the market once you account for survivor bias.

-W