Of all the investing principles I think everybody should follow, this is the most important. Without discipline, you are virtually guaranteed to lose money.

The stock market is the only "store" I know of that when people see an amazing sale, they run away in panic. In fact, they return their items to the store at a loss!

"It is only a loss if you sell"---one of the sayings over at the Boglehead forums. When the stock market crashes, do not panic and sell. You only serve to lock in your losses. In fact, if you are still in the wealth building process, you should buy more stocks! This is in line with the principle of rebalancing (discussed later).

There is no better story that I have found to illustrate this point than that of Bob, the world's worst market timer. Long story short, Bob starts saving at 1970. However, he is incredibly unlucky and decides to invest all the money he's saved up right before a large market crash (the crashes of 1972, '87, '99, and '07). In the end, he invested $184,000 (these are nominal, not inflation adjusted (real) dollars) over his lifetime. And despite his terrible luck, he still ended up with $1,100,000 (nominal) by the time he retired in 2013. The key to all this is that he never sold a single share after a market crash.

Now granted, to have an all stock portfolio right up until retirement is incredibly risky, and not recommended (the author of that post doesn't recommend it either). It's just for illustrative purposes.

If you invest with passive index funds in a portfolio suited for your risk tolerance, and don't panic during market crashes, you will do just fine. And more than likely, better than fine.

To aid you with this, I highly suggest you write an investment policy statement.

Investment Policy Statement (IPS) - A statement explaining your plan for investing. It should explain what to do during a market downturn (ie, do not panic and sell), your desired allocation of stocks and bonds, what funds you will use to accomplish your investment goals, how frequently you will invest, and your rebalancing scheme.

Before you start investing, you should write an IPS. Or, if you are already investing, write one now (unless the market has crashed at the moment). The single biggest reason to write an IPS is to guide you during the market downturns. Hence, you should write it when the market is not in turmoil.

It is only effective if you write it intelligently and actually follow your IPS to the letter. So make sure you give it some thought when you write it.

And there may be times when you feel like it needs revision. Which is fine. Sometimes life situations require that. But a change to your IPS should not be taken lightly. Give it careful thought. And only change your IPS when the stock market hasn't just crashed, as you do not want to be make changes to your IPS in a panic.

This disicpline should extend to beyond major market crashes, but also minor dips in the market. The best way to avoid urges to sell is to ignore the day to day prices of the stock market. If there is a major crash, which triggers your IPS thresholds for rebalalncing, then it will show up in all major news channels and you'll be aware of that. Otherwise, do not concern yourself with stock prices, as they can only tempt you to sell.

Know your time horizon

It is important to keep in mind your time horizon while investing. If you are saving for a short term goal, such as a down payment on a house, you should not invest those savings in the stock market. Any investment goal within about ten years (there's some debate on where this threshold should be) involves a time horizon that is simply too short for stock investments. Stocks always recover eventually after a crash, but there is a pretty good chance that it will not happen fast enough for a 10 year investment goal.

In the case of short term goals, you should use no and low risk investments, such as FDIC insured bank acconts and CDs, and US Treasury bonds (perhaps municipal bonds also, but I wouldn't recommend coprorate bonds).

For the most common long term goal, retirement, you need to make sure that you are investing for the end of retirement. The "typical" investment advice geared for people who want to retire around 60 is to save up enough money for a 30 year retirement. In my case, to be financially independent by 40, I need enough money for an approximately 50 year retirement. There are some people who save for an "infinite" retirement (I'm still learning about this, and will definitely write a post on it). The idea behind that is to have a large enough portfolio that it will throw off enough dividends and interest income to fund your lifestyle forever. This is probably what I'll end up doing.

Assess your risk profile

As I've expressed before, investing and not investing both involve risk. You face risk one way or another, and you should choose a stock/bond allocation in accordance with your risk profile. There are many questionnaires you can find online that can help you with this. Or, one common starting point is to hold your age in bonds - ie, if you are 29 you should hold 71% of your investment portfoilo in stocks and 29% in bonds.

As for me, I decided to hold 10 less than my age in bonds, because I'm pretty sure that rule of thumb is fairly old, and people are living longer these days. Also, I expect to earn far more once I graduate, so I can weather large losses to my portfolio fairly easily. But at the end of the day, what matters is your ability to sleep well at night. Pick an allocation where you would still be comfortable knowing that your stocks could drop by 50% or more.


Once you've decided on a stock/bond allocation, you should make sure your porfolio doesn't drift too far from your desired ratio. This means that during a stock market crash, you should sell some of your bonds and buy some stocks. This also has the side effect of selling high and buying low. This is necessary to make sure your portfolio reflects your risk profile - in a bull market, it is possible for your stock allocation to get far too large.

There are quite a few ways to rebalance: at fixed time intervals, at allocation thresholds, and at both fixed time intervals and allocation thresholds.

  • Fixed time interval rebalancing - You rebalance at regular intervals, whether that's monthly, quarterly, yearly, etc. There's no hard and fast rule on how frequently you should rebalance if you choose this method, but I probably wouldn't recommend rebalancing less frequently than a year.
  • Threshold rebalancing - You rebalance only when your portfolio allocation deviates beyond a certain point. So for example, if your threshold is ±5% (a common one and the one I recommend), and your stock allocation is 71% stocks, then you only rebalance when your stocks go above 76% or below 66%. For small allocations, such as holding only 12% in bonds (more common for younger investors), a 5 percentage point shift would be a very large shift in the relative value of the bonds. Hence, one method to address this is to also set a 25% shift of the allocation itself--so in this scenario, the threshold for rebalancing bonds is ±3%. However, you shouldn't let this 25% rule override your 5% threshold for larger allocations.
  • Time and threshold rebalancing - You check your portfolio at your desired time interval, but only rebalance if the rebalancing threshold has been met.
Advantages Disadvantages
Fixed Interval Very simple to implement
  • If the time interval is too short, it has the potential to cause you to miss out on (qualified) dividends and long term capital gains tax rates (to be explained in a future post)
  • If the time interval is too long, your portfolio may become out of balance, and the stock market could crash before you get a chance to rebalance. Hence, you may take on more or less risk than you intended
Threshold Ensures your portfolio accurately reflects your risk tolerance
  • Can require you to monitor your portfolio more closely. However, you can set up alerts for when your investments experience a large drop or rise in value, so this shouldn't require active monitoring.
  • While this is not likely, it can cause you to trade more frequently. Some brokerages such as Vanguard have a policy against this, but at least with Vanguard there are some easy workarounds.
Time and threshold Gets most of the advantages of each
  • Your portfolio will not reflect your desired risk profile as accurately as with threshold rebalalncing

I think the threshold and time and threshold methods are the best - at the end of the day, the point of rebalancing is to keep your portfolio aligned with your risk profile. And fixed interval rebalancing is either not good at that at long intervals, or can cause other problems at short intervals. However, this isn't to say that it is a bad method and you shoudln't use it. It's certainly better than not rebalancing at all.