Index funds - how do they actually work?

[kyanamerinas]

I was explaining index tracker funds to my partner the other day and he asked a question that stumped me, so i'm hoping you can help!

On hearing about index funds he said 'but surely that's almost guaranteed to make money so why doesn't everyone do it? there must be a catch'. He asked whether you are locked into the companies which match the index when you invest.
for example: say we were investing £1000 in the ftse 100 now (April 14). If in May 14, 1 or more of the companies which formed the ftse 100 when we invested suddenly crashed for some reason, would we continue to be invested in them? not looking at trading actively, just tracker funds.

i guess if this happened the fund would sell the lost companies and buy the new ones, which would result in a loss of money in the short term due to the cost difference of the two companies' shares. but i assume that shares continue to be bought and sold to match the index. is this right?

sorry if it seems a stupid question! rather nervous about investing and trying to get our heads properly round it!

[matchewed]

If a company falls outside of the parameters of the index they are sold.

[warfreak2]

Stocks are held in proportions defined by the index they are tracking. If one stock crashes, it might be sold, or other stocks might be sold to buy more of it - or maybe neither. For example, the S&P 500 index weights stocks by the market capitalisation, so if a stock price rises or falls, but nothing else changes, its market capitalisation rises or falls in proportion, and thus the holding in an S&P 500 index tracker fund should remain in proportion to market capitalisation, so none needs to be sold or bought. Meanwhile, in the Russell 1000 index, if a stock price falls so that that company is no longer one of the 1000 highest-ranked stocks in the Russell 3000 index, a Russell 1000 index tracker fund would sell all of it.

[marty998]

The actual mechanics of constructing an index portfolio are quite simple in theory. Whilst not practical, you could do it yourself. Lets say you have $10m to invest.

If a stock makes up 5% of an index, you will need to buy $500k worth of that stock. If a stock is 0.25% you buy $25,000 worth of shares. Just carry on in this manner until you buy every stock in the index.

Your portfolio revalues in line with the change in value of all the stocks in the index. There's no further buying and selling required. However, if you wish to add to your portfolio, you need to buy all stocks in proportion to the number you currently hold in all them. Should a stock crash, it will be replaced, but not necessarily straight away.

See attached excel doc for a quick and hopefully not too complicated example of building an index portfolio.

Regarding your specific question, indexes are typically "rebalanced" every quarter. You do actually lose, because if a very large stock crashes it gets replaced eventually by one that is coming in at the tail end, all else being equal, so the overall market capitalisation of the index is still smaller. You may wonder why re-balancing doesn't occur more frequently (even daily perhaps) but you can easily see that index funds would have to buy and sell stock daily on companies that are on the margin of inclusion/exclusion. It's going to fall apart quite quickly and trading in those stocks will be very volatile, thus causing more index problems.

This actually presents a bit of an opportunity for a canny investor to trade stocks on the margins because as soon as a stock is added to/removed from an index, ETFs and other tracking funds have no choice but to buy & sell quite heavily in those stocks shortly after. You can take advantage if you really know what you are doing.

[kyleaaa]

There's no catch with index funds. People invest elsewhere because they think they can do better (a few will, the vast majority won't).

[zurich78]

I'd say the only catch, if there is one, is that there is more limited growth potential with index funds.  I mean, index funds aren't likely to double in a year.  With a stock, that can certainly be the case.  Heck, some stocks will double and then some in an hour or day!  But they can also crash in a short period of time as well.

So the catch, from my POV, is just that you're trading the potential for massive gains for an increased likelihood of actual growth.

[TreeTired]

'but surely that's almost guaranteed to make money so why doesn't everyone do it? there must be a catch'.

I don't understand how a person can even ask that question.   Stock MARKETS,  and stock indices, can go down.

My favorite pretend do it yourself index portfolio is the Dow index.   Due to the way the index is calculated all you have to do is buy 100 shares of each dow stock.  (30 stocks in the index).    Never did it myself -  even with a discount broker you would pay $8 x 30 = $240 in commissions just to construct your index portfolio. 

[beltim]

'but surely that's almost guaranteed to make money so why doesn't everyone do it? there must be a catch'.

I don't understand how a person can even ask that question.   Stock MARKETS,  and stock indices, can go down.

My favorite pretend do it yourself index portfolio is the Dow index.   Due to the way the index is calculated all you have to do is buy 100 shares of each dow stock.  (30 stocks in the index).    Never did it myself -  even with a discount broker you would pay $8 x 30 = $240 in commissions just to construct your index portfolio.

I love this example.  That would result in a total cash outlay of about $255,000.  Your $240 in commissions would be about 0.094%.  And then you'd never have to pay any more fees!  Much lower cost indexing than even the lowest cost Vanguard fund.

Of course, it requires $250k to get there.

[marty998]

Could cost you only $25k if you buy 10 shares in each company to get the same effect.

Price weighted indices have always struck me as a bit odd compared to market cap ones.

[nereo]

'but surely that's almost guaranteed to make money so why doesn't everyone do it? there must be a catch'.

I don't understand how a person can even ask that question.   Stock MARKETS,  and stock indices, can go down.

My favorite pretend do it yourself index portfolio is the Dow index.   Due to the way the index is calculated all you have to do is buy 100 shares of each dow stock.  (30 stocks in the index).    Never did it myself -  even with a discount broker you would pay $8 x 30 = $240 in commissions just to construct your index portfolio.

Here's a thought experiment for you (and something i've been wondering about for quite some time...) 
Picking the stocks that will consistently beat the market is hard.  But is the converse different or any easier?
Take the dow index as an example.  How hard would it be for someone to successfully construct a "Dow+ index" - 25/30 of the DOW stocks.  Instead of picking the stocks primed to do the best, you would eliminate 5 that you'd expect to have sub-par performance.  Do that, and you beat the index.  Of course there are always surprises and turnarounds, but in a way I *think* it's easier to pick a company that is going no where than to pick a company that will beat everyone else.

hmm....

[hodedofome]

Interactive Brokers does commissions for $.005 a share with a minimum of $1 a trade. If you don't have a very large account, then it's something to look into for super low cost investing.

[hodedofome]

'but surely that's almost guaranteed to make money so why doesn't everyone do it? there must be a catch'.

I don't understand how a person can even ask that question.   Stock MARKETS,  and stock indices, can go down.

My favorite pretend do it yourself index portfolio is the Dow index.   Due to the way the index is calculated all you have to do is buy 100 shares of each dow stock.  (30 stocks in the index).    Never did it myself -  even with a discount broker you would pay $8 x 30 = $240 in commissions just to construct your index portfolio.

Here's a thought experiment for you (and something i've been wondering about for quite some time...) 
Picking the stocks that will consistently beat the market is hard.  But is the converse different or any easier?
Take the dow index as an example.  How hard would it be for someone to successfully construct a "Dow+ index" - 25/30 of the DOW stocks.  Instead of picking the stocks primed to do the best, you would eliminate 5 that you'd expect to have sub-par performance.  Do that, and you beat the index.  Of course there are always surprises and turnarounds, but in a way I *think* it's easier to pick a company that is going no where than to pick a company that will beat everyone else.

hmm....

That's the idea behind most of the 'smart beta'/RAFI indexes. Look into SYLD, they eliminate the shareholder value destroyers as an example. You still have to determine if the strategy can outperform the fees...