Dollar-Cost Averaging vs Lump Sum: What the Evidence Says

Someone hands you $30,000. An inheritance, a bonus, the proceeds of a house sale. It sits in your savings account and every day you do not invest it feels like a day you might be about to make a mistake in one direction or the other. That is the only version of this question that is real, and there are two honest answers to it: the one with the higher expected return, and the one you will actually go through with.

The question only exists when you already have the cash

$500 leaves your pay each month and lands in a retirement account. You are dollar-cost averaging already. Nobody chose it: you can only invest money that has been paid to you, so it arrives in instalments and buys whatever the price is that morning.

Lump sum means committing money you already hold. All of it, at once, into your target mix. Averaging in splits that same sum into equal pieces, and you buy on a schedule, whatever the market is doing on each date.

Six months, $500 a month, and what a fixed amount buys

Here is the mechanism that gives averaging its reputation. Six monthly purchases of $500. The fund’s price falls by half and climbs all the way back.

Month	Price per share	Amount invested	Shares bought
1	$50.00	$500	10.000
2	$40.00	$500	12.500
3	$25.00	$500	20.000
4	$32.00	$500	15.625
5	$40.00	$500	12.500
6	$50.00	$500	10.000
Total		$3,000	80.625

Two numbers fall out of that table. The gap between them is the whole idea.

The simple average of the six prices is (50 + 40 + 25 + 32 + 40 + 50) / 6 = $39.50. Your actual average cost is 3,000 / 80.625 = $37.21 a share. You paid $2.29 less than the average price you saw, and you did nothing clever to achieve it: the $500 simply bought 20 shares in the cheap month and 10 in the expensive ones.

At the month six price of $50, the position is worth 80.625 x 50 = $4,031.25 on $3,000 invested, a gain of 34.4 percent, in a fund whose price finished exactly where it began, while the lump sum comparison for those same six months is bleak: $3,000 at $50 buys 60 shares, worth 60 x 50 = $3,000 at the end. Flat.

Now reverse the price path. Markets do that too. Same $500 a month, same six months, price climbing from $50 to $62.

Month	Price	Shares bought
1	$50.00	10.000
2	$52.00	9.615
3	$55.00	9.091
4	$58.00	8.621
5	$60.00	8.333
6	$62.00	8.065
Total		53.725

Average cost is 3,000 / 53.725 = $55.84 and the position is worth 53.725 x 62 = $3,330.94. The lump sum investor bought 60 shares at $50 and holds 60 x 62 = $3,720. That is $389 more from the same $3,000.

Averaging in wins when the price falls after you start. It loses when the price rises. That is the entire trade, and you do not get to know in advance which path you are on, so test your own price paths in the dollar-cost averaging calculator.

What the evidence says, and what an average hides

Markets have risen over more periods than they have fallen. Cash waiting on the sidelines usually misses returns it could have earned. Vanguard research published in 2012, comparing immediate investment against averaging in over rolling twelve month periods in US, UK and Australian data, found immediate investment finished ahead roughly two thirds of the time.

Read the other third carefully. In those periods, averaging in meant a smaller loss, and for a number of people it meant the difference between staying invested and giving up on investing altogether. That does not appear anywhere in an average.

There is a second reason the arithmetic favours moving sooner. Cash earns a short-term interest rate while it waits, and over long horizons that rate has generally trailed the return on a stock and bond portfolio, so every month you delay you are deliberately holding the lower returning asset. Across six months on $30,000 that is a small price for calm. Across three years it stops being small.

Approach	Expected outcome	The bad scenario	Fits
Invest it all now	Higher on average, historically	Buying the week before a large decline	Money you will not need for a decade
Average in over 6 to 12 months	Lower on average	A steady climb you bought into slowly	Cash you are frightened to commit
Half now, half on a schedule	Between the two	Between the two	Most people with a windfall

Choosing, and then writing the dates down

If you decide to average in, the schedule needs to be short enough to finish and long enough to be worth doing. Six to twelve months covers nearly every case.

Work it in dollars. That $30,000 over six months is $5,000 on the same date each month. If the market falls 20 percent in month two, you still have $25,000 to deploy at better prices and the rare experience of watching a decline do something useful for you. If it rises 10 percent instead, the cost of your caution is the return given up on the uninvested balance, which comes to a few hundred dollars. Nobody is ruined by that.

The strongest argument for averaging in has nothing to do with returns, and I want to state it plainly because most writing on this subject pretends otherwise. Someone who invests $30,000 on a Tuesday and watches it fall 18 percent by Christmas may sell at the bottom and sit in cash for three years, while someone committing $5,000 a month has four more purchases coming and a reason to want the lower price. The second investor has the worse expected return and a much better chance of still being invested in five years, and five years is what actually pays you.

The part that matters more than either answer

Whatever you do with a windfall, automate the ongoing money, because a standing transfer on payday into your chosen funds removes the monthly decision, and the monthly decision is where most of the damage happens.

Two practical notes. Set the transfer for the day after payday so it never bounces, and make sure it buys the funds that implement your target mix from asset allocation, because uninvested cash left in the settlement account has a talent for staying uninvested for years.

Those contributions also do rebalancing work for free. Directing each month’s money into whichever sleeve is behind keeps you near target without selling anything, which is the cheap method in portfolio rebalancing.

Where averaging in fails

It does nothing for you in a market that falls and stays down. Every purchase is underwater, and a lower average cost is just a better price on an investment that is still losing money, which is the version of a bad year that averaging cannot help with.

It cannot fix a bad portfolio either. Putting $500 a month into one concentrated position is a diversification problem on a direct debit, which is the argument for reading diversification before you automate anything.

And it costs a little more in a taxable account when it means many small purchases of exchange-traded funds, each one paying the bid/ask spread. Mutual funds price once a day and handle small automatic amounts more cheaply. That difference is covered in ETF versus mutual fund.

Do one thing this week: set up a single automatic transfer for an amount you will not miss, and if you are sitting on cash, write the investment dates in your calendar before you close the laptop. The rule of 72 calculator will show you how long that money takes to double at various returns, which is usually enough motivation to stop deliberating.