As a followup to yesterday’s post on how to make money in the stock market, I thought I’d expand out the numbers and include the full dot com debacle rather than cutting off the analysis in 2001. Thus, I pulled the split-adjusted closing value of the S&P 5000 for every day from July 1, 1982 through June 29, 2007. This gives us a full 25 years of data ending at the close of the second quarter of this year.
In all, there were 6, 308 trading days during the period under consideration. Here are the pertinent numbers:
July 1, 1982 closing value: 108.71
June 29, 2007 closing value: 1503.35
Compounding that out, the S&P 500 grew at an average rate of 10.7% per year.
Once I had the data in hand, I simply figured the daily increase/decrease across the entire period, translated that into a daily percentage change, and then sorted the data. The positive point changes accounted for by the 10, 20, and 30 best performing days (in terms of percent change) are as follows:
Top 10 days: 366.68
Top 20 days: 737.25
Top 30 days: 1012.02
Now, turning this back into the sorts of numbers that we talked about previously… If you had invested $10, 000 on July 1, 1982 then, assuming that you were in the market for the full period, or that you missed the top 10, 20, or 20 days, you would’ve had the following amount as of June 29, 2007:
In for full period: $138, 289.95 (10.7%)
Missed top 10 days: $104, 541.44 (9.4%)
Missed top 20 days: $70, 471.90 (7.5%)
Missed top 30 days: $45, 196.39 (5.2%)
(The parenthetical values are the compound annual return for each scenario.)
As you can see, missing out on the biggest days has a huge impact on your investment performance. What makes this is all the more sinister is when the biggest days happened. Looking solely at the top 10 days in terms of percentage return, 3 of the 10 best days overall occurred in the 10 days following Black Monday in 1987, whereas 4 more occurred during the tumultuous bursting of the dot com bubble, including 2 right as the marketing was bottoming and starting to head back up.
In other words, if you lost your nerve in the wake of Black Monday, you would’ve missed out on a major rebound that included daily gains of 9.3%, 5.3%, and 4.9%. By the time you realized the world wasn’t about to end, you would’ve missed out on a big portion of the recovery. In other words, you would’ve truly sold low and then, once you got your nerve back, bought high.
Granted, if you were able to foresee Black Monday and get out just ahead of it, you would’ve missed the pain entirely. But be honest with yourself. How confident are you that you can consistently see events like that coming and get out at the peak? And how confident are you that you would’ve called the bottom and gotten back in the next day?
Of course, this analysis ignores the fact that you may have been prescient enough to get out early enough during certain declines to avoid a portion of the losses. Nonetheless, it does highlight the risks associated with missing a market rebound.
Note: The numbers from yesterday were pulled from an article in Money Magazine, and I thus can’t vouch for their accuracy. I also don’t know how if they judged the best days based on point gains or percentage gains. I used the latter, as it’s a more accurate metric of relative performance.
nickel: I’m fan of rebalancing, and as you astutely acknowledge it’s inherently a form of market timing just without the active decision on when to do it. However for a calm and disciplined investor picking a random but systematic point in time for rebalancing is really no different than picking specific point in time. If you can’t market time then you’re no worse off, and if you can then you’re better off. Of course, most investors do not behave calmly and rationally. The market may be efficient but far from sensible.
dong: I think adjusting your asset allocation should be done (as necessary) on an ongoing basis rather than based on market performance. If you’re closing in on retirement, you should be moving toward a more conservative mix. That’s true regardless of how the market is performing.
On the other hand, regular rebalancing (say twice a year) gives you a systematic method for selling high and buying low. If a certain investment category or fund runs wild, it will be overrepresented relative to your target allocation. Thus, when you rebalance, you’ll be selling off a bit of something that has performed exceptionally well and buying into something that has performed less well.
I’m generally a fan of passive investing, but sometimes I think that it gets taken a little too far. I don’t mean to imply that you have, but rather there is some worth in attempting to time the market or pick individual stocks for that matter.
On timing, I think more important to anything else is the timing as it applies to your life plans. If you’re in the market and only time withdrawals by need you might be caught off guard. I think it’s best to have a plan. I mean if you know you’re going to retire in a few years, might it not make some sense to adjust your positions when you think you’re at a top?
This is certainly oversimplified, but I think it still makes the point. The point is, trying to time the market has a lot of people money and finding the exact bottom is impossible. Selling out on a downturn can make your returns go south in no time. Patience and risk tolerance are key.
Yes, of course this is an oversimplification. However, the point that good days are frequently directly after bad days is incredibly important. If you sell out on the way down, you will very frequently lock in your loss and miss the big bounce. You are assuming that you will sell *before* the drop, and then miss both the low and high performing days. But that’s not how investors behave. People get rattled when they incur losses, they sell, and then while licking their wounds, they miss the best part of the recovery. Once that happens, they feel better about the market and jump back in. Thus, they ended up selling low(er) and buying high(er).
Moreover, in a generally growing market (which ours has been over almost any reasonably long period) the ups outweigh the downs, so sitting out will (on average) lead to underperformance.
This type of analysis is flawed because, as one of the other commenters pointed out, it assumes that by staying out of the market one misses out only on the good days while still hitting ALL of the bad days. Of course, if you are out of the market you will miss those down days as well, so your average return would be substantially higher than the numbers you calculated.
In fact, I am guessing that bad days and good days show up at about the same frequency. Moreover, I imagine that good and bad days are probably clustered together – i.e. really good days closely preceding or following really bad days, as volatility in the market increases in times of uncertainty. If that is the case, and you take out the good and bad days in equal measure, I am guessing that your average return would still be close to the 10.7% average over the entire period.
Great analysis! This is quite a strong argument for not panicking 🙂
@raz: One problem with that approach is that mutual funds are priced at the close of the day. Thus, if you wait until the day after, you get the closing price after the potential immediate rebound. This is less of an issue for individual stocks or ETFs.
@Matt: Bear markets are defined as a 20% decrease in stocks. This period most definitely includes a serious bear market. In fact, that’s why I re-worked the numbers… To include the full effects of 2000-2002 bear market.
@raz:
How long do you leave it in for? The top 10 best days only amount to 1/5 of the gain over that period. The top 30 account for 2/3. Why not just leave your money in and get ALL of the returns?
You’re analyzing what is probably the best time period in our stock market’s history. We haven’t had a true bear market during that period, because Easy Al has always been there to bail out the speculators.
Well put this time. This is a much more complete argument. Of course it does make an argument for keeping your money on the sidelines until there is a bad day, and then jumping in the day after. If the best days come right after the worst, you could avoid the worst, and be fairly certain of being in on the best. Even if a bad day was followed by another bad day, history shows that the market goes up with time. This seems like a hard thing ot do, but it is really nothing more than following the philosophy of buy on bad news, sell on good news.
Wow! Excellent analysis. It is so amazing that in 25 years, even missing out on 10 days has a significant impact.
Another variation: Could you include DCA, to if that would smooth the variations out?
The unfortunate thing is that the worst days often seemingly come out of nowhere, making them virtually impossible to avoid for the typical investor. On top of that, the best days are frequently concentrated into the very early stages of a recovery, before most shaken investors get their nerve back. The thing that causes people to get out of the market is a scary drop, and they don’t get back in until things are looking bright. This means they’ll lock in a loss, and then often miss out on the strong positive days in the very early stages of the recovery. Thus, while it might be interesting to look at, I’m not sure it would be all that informative in the present context.
I think its been mentioned before, but I wonder what effect would missing the 10 etc worst days have had.