Dollar Cost vs. Lump Sum Investing: Where Dollar Cost Averaging Fails.

Dollar Cost Averaging (DCA) is touted by some financial planners as the solution to all of investing’s problems.  By continuing to invest money at a regular interval, you buy more shares when prices are low and more shares when prices are high.  Additionally, dollar cost averaging fits the general schedule of how people normally get paid – every two weeks you get your paycheck, and you also automatically invest in your 401(k), for example.

That’s great… for predictable streams of income.  This article will *not* try to convince you to stop your normal recurring investments.  However, dollar cost averaging meets its match when introduced to a windfall.  When you have extra funds, you shouldn’t tiptoe into the market, you should dive right in with a lump sum investment.  Don’t believe me?  Let me convince you…

Your Normal Disclaimer

I have lots of risk tolerance, and a long time horizon.  This meshes well with stock investments, and taking on additional risk in my investment strategies.  Your situation may be similar, but remember you are unique.  You should consult a financial planner if you have any doubts about implementing an investment strategy.  I am not a financial planner, just a guy armed with some spreadsheets.  This is not advice, and of course, past performance does not guarantee future results.  Carry on…

The S&P 500 Index

I am using the S&P 500 index, a basket of 500 domestic stocks, as a representation of the total stock market in America.  You may argue for or against it; I find that since it is a commonly used indicator of the health of the abstruse  ‘stock market’ on any one day it is good for our purposes.  This article will use the S&P 500′s data for its arguments, and assume that the investor actually holds the index components.  This is likely impossible for the average investor.  If you want to invest in an index, find an index mutual fund or ETF like SPY.

The 'Absolute' and the 'Price Only' Return of the S&P 500, 1989 - 2009
The ‘Absolute’ and the ‘Price Only’ Return of the S&P 500, 1989 – 2009

Above you can see a chart of the real price (in blue) that the S&P 500 closed at on the dates listed on the x-axis.  The chart covers the 20 year period from 1989 to 2009.  I also added (in red) the absolute price, a made up value which includes the dividends paid in the price of the stock.  This reflects the real return of the index, if you held the component stocks.  The data components of this chart can be found in this S&P provided spreadsheet.  The S&P’s price increase averaged (in this article, ‘average’ means ‘geometric mean‘) around 6.074% annually, and the absolute return was 8.426% annually.  The additional return provided by the dividends was therefore important to include in our data. Below is this data smoothed for your consideration.

S&P 500 Returns, Smoothed Data
S&P 500 Returns, Smoothed Data, 1989 – 2009

Basically, the smoothed data is stocks with their volatility stripped out.  Since volatility is a good substitute for risk, (they are often used interchangeably, even though volatility talks of the past.  Perhaps I’ll discuss risk in the future?) this is fine to look at academically.  Of course, if the volatility of stocks was this timid, return would have to fall in turn.  However, it’s important to establish a trend.

Drift Rate

Understanding that historic performance of a basket of stocks has little bearing on the future, let’s assume that stocks will, in general, go up.  Data since 1871 seems to bear this out, even though there has been little movement in the last 10 years.  We can thus talk about the drift rate of stocks – a term from.  I will assume that an investor receives a windfall of $10,000 for this exercise, on the last day of the market close one year. Let’s also assume (so many assumptions!) that dividends are smoothed over 2 week intervals as well, to make this math a little easier.  For a look at non-smoothed results, see this calculator.

For the lump sum investor, they will buy $10,000 worth of ‘S&P 500′.  At the end of this fictional year, they will have $10,842.60.  The dollar cost averager spreads his purchases out over the entirety of the year, 2 weeks at a time.  Each 2 week period has an average gain of 0.31163%.  His investment, at the end of a year, is only worth $10,431.96.  How is this so?  Look at this graph, which shows the amount invested, and its worth, for each 2 week period.

Lump Sum vs. DCA, 2 Week Intervals, 1 Year Duration

Lump Sum vs. DCA, 2 Week Intervals, 1 Year Duration

How to Use This Data

If you agree with these statements: “Market Timing is Hard, if not Impossible“, and “Stocks Will Generally Increase“, you should invest your windfalls (that you were planning to invest, anyway) using the lump sum method.  In general, stocks will move up and to the right year after year.  Armed with this knowledge, you will avoid missing this (theoretical but not guaranteed) 0.31163% biweekly return.

There are ways to massage the data to have dollar cost averaging beat lump sum investing for information of this sort.  Once volatility is added back to stock returns, some of these situations become apparent.  However, lump sum investing will generally beat dollar cost averaging in the long run.  Consider your options and risk tolerance, but also consider historical results when you sit down to invest your next windfall (and talk to your financial advisor).  Good luck!

Tweet about this on Twitter1Share on Facebook0Share on Google+0Share on Reddit0Share on StumbleUpon0Email this to someone

Trackbacks

Leave a Reply

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