Wall Street Journal columnist Jason Zweig wrote a puzzling article Why Buying on the Dips Isn’t All It’s Cracked Up to Be. The gist of it is that buying on the dips doesn’t work.
I find the article puzzling because it has a weird definition for buying on the dips. It defines a dip as a relatively large one-day drop. If the market reaches an all-time high and the very next day it drops say 2% or 5%, it’s a dip. The article goes on to show that, in the last 10 years, holding cash waiting for those "dips" didn’t beat investing a lump sum on day one.
Of course it wouldn’t. I don’t have access to the raw data used in the backtesting, but I would venture to guess that a large percentage of the one-day dips happened toward the end of the study period. Although prices were lower than those on the previous day, they were still still higher than several years before. That’s why buying on the dips, as defined by the article, didn’t work.
But that’s not what I understand buying on the dips means. To me, buying on the dips is a shorthand for buying more on the dips. Also, a dip isn’t just a one-day drop. A dip is a low price point relative to a recent high.
Under my definition, which I think is more common than the one-day drop definition, you start out with the same portfolio as buy-and-hold, say 60% in stocks 40% in bonds. If stocks drop say 20% from a recent high, that’s a dip: you sell bonds to buy more stocks. When stocks recover, you sell stocks and return to the neutral position.
In this setup, you always have as much in stocks as in buy-and-hold. Sometimes you have more, but you never have less. As long as stocks recover from the drop, you will have a higher return than buy-and-hold.
Since we know it takes a good 20% drop to move the portfolio weight for stocks by 5 percentage points, buying more on dips is basically band-triggered rebalancing. As Allan Roth showed in his blog post How to Play This Volatile Market, this would’ve worked very well in the last 10 years.
Buying (more) on the dips is still all it’s cracked up to be.