Don’t dollar cost average on purpose

You have cash. Let’s say you have a nice lump of it. Maybe you inherited it, or won the lottery, or are new here and simply have it in your savings account. You decide you want to start investing in the stock market and make that money work for you. You learn you have two options:

  1. Invest it all right now.
  2. Invest only part of it now, investing the rest of it little by little over time.

Which is better? Well, option 1 is better, but let’s talk a little more about option 2.

Dollar Cost Averaging

Option 2 is essentially what’s known as dollar cost averaging. Instead of investing a lump sum all at once, you break it up into equal portions and invest these chunks at regular intervals–say, once a month. The idea behind DCA is that if the market goes down or is volatile in the near future, you’ll avoid investing all of your money right at a relative peak. And, of course, if there was a situation where we knew we were at a relative peak, this would work well. But unfortunately, that situation doesn’t happen. We simply can’t predict the future.

But let’s go through four possible scenarios and see when DCA would work. We’ll assume a time horizon of about 5 years. Over a 5 year period, there’s an 80 percent chance of positive stock market returns, so keep that in mind. Here’s what are four cases look like:

Option 1 is that we’re at a relative peak. Obvious DCA would be better in this case. But option 1 is exceedingly rare–it’s very uncommon that the market would just go straight down for 5 years.

Option 2: DCA still wins, because it’ll buy you more shares at lower prices in the middle years.

Option 3: DCA loses hard–you unknowingly buy many of your shares around a peak.

Option 4: DCA loses again–we were at a local minimum.

I don’t know whether option 2 is more common than option 3. Neither of them are that common, because it’s much more common for the market to go up over a 5 year period than be essentially flat. Option 4 is much more common than Option 1, because of the 20% of time the market goes down over a 5 year period, returns are very rarely more than negative single digit percentages annualized. All this to say that in the vast majority of cases (based on their probabilities of occurring), DCA is a worse choice.

Let’s briefly consider a shorter time frame of 1 year. Over a one year period, the market goes up about 2/3 of the time, so there’s a lot of room for error. Again we run into the same problem–we have no ability to predict when this will happen, and 2/3 is more likely than 1/3.

So, is DCA always bad? No, sometimes it’s great. The problem is we just don’t know when. So, if you have a lump sum, I suggest you play the odds and invest it all as soon as possible.

But what if the market crashes?!? Humans are driven by fear, so we may still not want to face this risk even though it’s the statistically worse choice. Luckily, unless we are currently retired, this risk is mitigated by the fact that we are already dollar cost averaging anyway. Every month as we make, save, and invest money, we are essentially DCA. JL Collins likes to say that the best case scenario after investing a lump sum is that the market continues to rise. But then second best scenario is that a large dip follows, so you can continue to purchase cheap shares with the money you save each subsequent month! So really, you’re in a win-win-win situation (options 2-4; and really option 1 as well given that the market always recovers).

So what are you waiting for? Go buy yourself some Vanguard total stock market index funds!

1 response to "Don’t dollar cost average on purpose"

  1. By: Alex Posted: December 29, 2018

    These hand drawn graphs are the best!

Leave a Reply

Your email address will not be published. Required fields are marked *