My company just switched to a new 401(k) administrator in the U.S., so a lot of people will be thinking about their portfolio allocations around now. I decided to put my advice on this blog (rather than the internal blog) because a lot of my friends tend to ask me for investment advice periodically.
A 401(k) is a tax-advantaged retirement savings account. Put another way, it is a government subsidy; the more money you put into your account, the bigger the subsidy. (As an aside, it is another one of those subsidies that helps the rich more than the poor. Rich people can afford to put the maximum into their accounts, and they have higher tax rates, both of which increase their subsidies; poor people can't save much for retirement, and they have lower tax rates, both of which decrease their subsidies. People who think government subsidies go only to the poor and minorities - generally Republicans - are dead wrong for hundreds of similar reasons.)
401(k) plans are taking over from traditional pension plans (where your retirement benefit is an annual percentage of your final salary, or something like that) for one simple reason: they are better for employers. They are cheaper (essentially free if the company doesn't provide a match, like mine) and less risky (because your costs are completely predictable). There is a strong case to be made that our country would be better off with the old pension system (see this
Fresh Air episode), but that's what most of us are stuck with.
401(k)s place the responsibility on the employee to, first, make the choice to participate and, second, manage his or her money by deciding which mutual fund(s) to put that it in. On the first point, you should save absolutely as much as you possibly can. Remember, the more money you save, the more money the government is giving you for free. End of story.
On the second point, which is the one I usually get questions about, there are only two concepts you need to understand: indexing and asset allocation.
An index fund is designed to give you the same returns as some market index, most often the S&P 500, which follows some large segment of the market as a whole. In its opposite, an actively managed fund, a fund manager tries to guess at which stocks or other securities will perform better than the market as a whole and buy those. You should never invest in an actively managed fund.
There are mountains of evidence that index funds are better than actively managed funds. First, in any given year, a large majority of actively managed funds do worse than the appropriate indices. But then people think that they should find the fund managers who do well and entrust them with their money. But the fund managers who do well are just lucky - they are the equivalent of people throwing darts at the stock pages of the Wall Street Journal and buying the stocks they hit. Of course some of them will beat the index by luck, but you have no reason to believe they will beat it next time.
But, people say, what about fund XYZ that has beaten the index 10 years in a row? Well, out of 1,024 funds, each with a 50% chance of beating the index each year, 1 of them will beat it 10 years in a row. And there are many more than 1,024 funds out there, and the relevant sample is far larger than that, because fund companies tend to shut down the funds that don't do well in their initial years. In fact,
statistical analysis demonstrates that virtually all apparent ability to beat the market can be written off as pure random variation.
Many people refuse to believe this. To them, stock picking should be like any other endeavor - some people are good and some are bad - so the whole argument makes no sense. But there is a reason why the statistics show what they do, and that is the theory of efficient markets. Stock picking is trading on information - you buy a stock because you have information that makes you think the price will go up. In an efficient market where information flows freely, however, all public information is already reflected in the stock price. That is, by the time you find out that some pharmaceutical company had its drug approved by the FDA, thousands of other people already know, and they have already bid the stock price up to fairly reflect the value of that approval. So when you buy the stock, you have no advantage, and it could go down just as easily as it could go up.
The stock market was not always efficient, which is why stock picking used to work for some people. And there are markets that are not efficient, like the market for timber forests. But all the markets that all the mutual funds in your 401(k) invest in are efficient, because there are thousands of investment banks and hedge funds with very large computers trading instantaneously whenever new information becomes available. There is a lot of math that makes the same argument, but this is why stock picking doesn't work. It just creates transaction costs (every time anyone buys or sells a stock he loses a small amount of money) and increases taxes, and you also have to pay the salary of the stock picker, which is why most mutual funds do worse than indices.
(Yes, there are still ways to beat indices. You can take on more risk, which is basically what hedge funds and investment banks did for the last 10 years, but that's not something you should do. You can also trade on insider information, because it isn't priced into the stock price yet, but that's illegal.)
So all your money should go into index funds. In fact, I think it's negligent that any company - including mine - even offers any other kind of fund in its 401(k) plan, but that's beyond my control.
Asset allocation is a more complex topic. Different asset classes (US stocks, international stocks, US corporate bonds, US Treasury bonds, commodities, etc.) have different risk/return profiles; in general, the higher the expected return, the higher the risk (the variance around that expected return). Unlike with indexing, there is no mathematical proof that one allocation is better than another. The important thing is to pick one and stick with it, because otherwise you could fall into the trap of buying more of your funds that are doing well (otherwise known as buying high) and selling your funds that are doing poorly (otherwise known as selling low). A common recommendation is 70% stocks and 30% bonds, or maybe 50% US stocks, 20% international stocks, and 30% bonds, and that's probably a good enough recommendation for these purposes. (Mine is 38% US stocks, 25% international stocks, 11% real estate investment trusts (a kind of stock fund), 13% traditional US bonds, and 13% US inflation-indexed bonds, not counting cash.) But just pick one and stick with it, meaning every six months or so you should trade a little bit of money between your funds to get back to your original allocation. (As things appreciate or depreciate at different rates, your allocation will change slightly.)
And finally, do not withdraw money from your 401(k) and pay a penalty if you can possibly avoid it. That's like giving your entire subsidy back to the government, and then giving them an additional 10% of your money on top.
For people at my company, the only funds you should even consider putting money in are:
- Spartan U.S. Equity Index Fund (tracks S&P 500 Index)
- Spartan Extended Market Index Fund (tracks Wilshire 4500, essentially all US stocks other than the S&P 500)
- Spartan International Index Fund (tracks MSCI EAFE Index, essentially most stocks outside North America)
- Fidelity U.S. Bond Index Fund (tracks Lehman Brothers Aggregate Bond Index)
The Fidelity Freedom funds are designed to change your asset allocation as you approach retirement, but you should avoid them, since they invest in actively managed funds (and they have the high expenses to prove it). Vanguard has a similar set of funds that invest in underlying index funds (duh!) ... but Fidelity, like any company, is looking out for the interests of its shareholders, not your interests. Which is a point you should remember for virtually the entire financial services industry.
