Dollar-Cost Averaging vs. Lump-Sum Investing After a Crash: What the Data Says

At some point after a crash, you end up holding a lump sum you have to decide what to do with — proceeds from selling what was left of a position, a tax refund, a settlement, or just the cash you moved to the sidelines when things got ugly and haven’t put back to work. The question that follows is almost always the same one: put it all in at once, or spread it out over time and buy in pieces?

This is a different decision than the one covered in Down 50%? Here’s the Actual Plan, which is about what to do with a position you’re already holding. It’s also different from Dead Cat Bounce or Real Recovery, which is about timing — whether a bounce is real before you act on it at all. This is about mechanics: once you’ve decided you want exposure again, how do you actually deploy the money. Get the timing question right and still botch this one, and you can turn a good decision into a worse outcome than it needed to be.

What the actual research says — and why it’s more useful than most people assume

Vanguard has studied this question directly, more than once. Its original 2012 paper, “Dollar-Cost Averaging Just Means Taking Risk Later,” compared a lump-sum investment against a 12-month dollar-cost-averaging schedule across historical return data in the US, UK, and Australia, and found lump-sum investing came out ahead roughly 68% of the time, by an average of about 2.3 percentage points over the deployment year in the US data. A follow-up 2023 Vanguard study by researchers Megan Finlay and Josef Zorn, using the MSCI World Index from 1976 through 2022, found the same pattern: lump sum beat a three-month cost-averaging schedule in a 100% equity portfolio by an average of 2.2%, again winning in roughly two-thirds of the historical periods tested.

Read at face value, that looks like a clean verdict for lump sum. But the “roughly two-thirds of the time” framing is really an average across every kind of market environment in the dataset — mostly ordinary years where prices drifted upward, which is what markets do most of the time over long stretches. That’s not the environment you’re actually deciding in. You’re deciding after a crash, in a market that has already dropped sharply and may or may not be done. The other third of Vanguard’s own data — the periods where dollar-cost averaging won — clusters specifically around sustained downturns and the recoveries that followed them. Starting a lump-sum investment in early 2008, right before the financial crisis, meant absorbing the full decline before any recovery began. A dollar-cost-average schedule starting at the same point kept buying at progressively lower prices through the fall, which meant a lower average cost basis heading into the recovery. The same basic pattern shows up around the 2000-2002 decline and the sharp 2020 pandemic drop.

None of this means dollar-cost averaging is now guaranteed to win just because you’re recovering from a loss — nobody, including Vanguard, can tell you in advance whether the specific market you’re re-entering has finished falling or has further to go. What it means is that the “lump sum usually wins” finding, on its own, is close to irrelevant to your actual decision. You’re not choosing between the two strategies in a randomly selected year. You’re choosing in the specific subset of conditions where the historical edge is genuinely closer to a coin flip, or tilts the other way.

The comparison

	Lump Sum	Dollar-Cost Averaging
Historical win rate (all periods, Vanguard data)	~68% of the time	~32% of the time
Performance if the market keeps rising from here	Best case — full amount captures the entire move	Worse — later purchases cost more than the first
Performance if the market keeps falling from here	Worst case — full amount takes the full additional drawdown	Better — later purchases average down the cost basis
Emotional load	High at the moment of the trade, low afterward	Low at any single moment, but drawn out over the whole schedule
Best suited to	Someone confident the worst is over, or investing on a long horizon where short-term timing washes out	Someone genuinely uncertain whether the bottom is in, or still shaken enough that a single all-in decision feels paralyzing
Tax/recordkeeping complexity	One transaction, one cost-basis lot	Multiple purchases, multiple lots to track — see FIFO, LIFO, HIFO
Risk if you abandon the plan halfway	Not applicable — the decision is made in one move	Real — stopping partway through because the market moved against the partial position gives you the worst of both: some money exposed to further downside, the rest still sitting in cash missing the recovery

The psychological math, not just the statistical math

There’s a second reason this decision matters beyond the raw expected-value comparison, and it’s specific to this site’s audience. If you just took a large loss, the emotional cost of being wrong twice in a row — going all-in on the “recovery” only to watch it drop further — is not symmetrical with the emotional cost of a slower, staged re-entry that underperforms a hypothetical lump sum by a couple of percentage points. Loss aversion means a second loss on top of a first one doesn’t just hurt twice as much; for a lot of people it’s the point where they stop investing altogether, which is a far worse long-run outcome than modestly underperforming an all-at-once entry that happened to work out.

This is a legitimate reason to weight the decision differently than a purely statistical read of Vanguard’s numbers would suggest — not because the math changes, but because the cost of the bad outcome isn’t the same for someone re-entering after a loss as it is for an investor deploying a routine windfall in an otherwise calm market. Staged re-entry is, among other things, a way of managing your own probability of quitting, not just your expected return.

A worked example

Say you’re sitting on $30,000 after selling out of a position that lost half its value, and the broader market has since dropped another 15% and is showing the first signs of stabilizing — not confirmed, just showing signs.

Lump sum: You put the full $30,000 in today. If the market has actually bottomed and rises 20% over the next year, you’ve captured the full move — you’re at $36,000. If it drops another 15% first before recovering, you’re down to $25,500 before any recovery even starts, and you have to sit through that additional drawdown having just committed the entire sum.

Six-month dollar-cost average: You commit $5,000 a month for six months. If the market rises steadily from today, your average entry price is higher than the lump-sum investor’s, and you underperform — a real cost, not a hypothetical one. If the market drops another 15% over the first two or three months before turning up, your later purchases land at the lower prices, your average cost basis ends up better than the lump-sum investor’s single entry point, and the emotional experience of watching $30,000 immediately lose value overnight never happens, because only a fraction of it was exposed at any given moment.

Neither outcome is guaranteed, and that’s the actual point: the honest answer to “which one is better” is “it depends on what the market does after you decide,” which is exactly the thing you can’t know in advance. What you can decide in advance is which failure mode you can tolerate — modestly underperforming a rising market, or taking a full-size hit to a still-falling one.

A framework for choosing

Ask these questions before picking either approach:

• Can you name a specific reason to believe the worst is over, beyond “it feels like it should be by now”? If yes, that leans toward lump sum. If your honest answer is “I don’t know, I just want back in,” that uncertainty is itself information favoring a staged approach.
• Could you tolerate watching the full amount drop further immediately after committing it, without it triggering the same panic-sell impulse that may have contributed to the original loss? If the answer is genuinely no, a lump sum isn’t the right tool for you regardless of what the statistics say, because a strategy you abandon under stress doesn’t get to keep its theoretical expected value.
• Are you deploying new capital you’re only now getting to, or moving money that’s been sitting in cash for a while already? The longer money has already sat out of the market, the more the “get exposure back sooner rather than later” argument for lump sum strengthens, since every month in cash is a month of missed compounding, not just a month of avoided risk.
• Can you actually commit to a schedule and follow it, including on days the market is falling and every instinct says stop? A dollar-cost-average plan you’ll likely abandon partway through isn’t really a dollar-cost-average plan — it’s a worse, accidental version of both strategies at once.

FAQ

If lump sum wins most of the time, why would anyone dollar-cost average back in? Because “most of the time” is an average across all starting points, including plenty of ordinary years where the market just drifted upward. It says little about your specific starting point, which is a market that already fell sharply and may not be done falling. Vanguard’s own comparisons show dollar-cost averaging closing the gap or coming out ahead specifically around sustained downturns and their recoveries — 2008, the 2000-2002 slide, and the 2020 crash are the commonly cited examples — because a schedule that keeps buying through a continued decline lowers your average cost basis instead of locking it in at a single, possibly still-elevated price.

How long should a re-entry schedule actually run? There’s no universal number, but most of the research comparing the two approaches uses schedules between three and twelve months, and that range is a reasonable starting point for a recovery re-entry too. Shorter than that and you’ve mostly just recreated a lump sum with extra steps; longer than that and you’re carrying cash-drag risk (sitting out of a rally that started while you were still deploying) for an extended period. Pick a length you can commit to in advance and actually follow, rather than one you’ll be tempted to abandon the first time the market moves against your partial position.

Does this apply to a lump sum I’m sitting on, or also to money I save and add each month? It’s really a question about the lump sum specifically. If you’re already investing a portion of every paycheck as it arrives, you’re functionally dollar-cost averaging by default — there’s no separate decision to make there. This comparison matters when you’re holding a discrete pile of cash all at once (proceeds from selling a position, a tax refund, a settlement, savings you pulled to the sidelines during the crash) and have to decide whether to put all of it to work in one transaction or spread the entry out.

What if I genuinely can’t decide and just keep sitting in cash instead? That’s the one option the research is unambiguous about: it’s very likely the worst of the three. Both lump-sum and dollar-cost-average strategies get your money into the market; sitting in cash indefinitely while you wait for certainty guarantees you miss the return on all of it, and “wait until I’m sure” after a crash tends to quietly extend for months once the fear that drove the indecision doesn’t fully resolve on any fixed schedule. If you can’t choose between the two approaches in this article, defaulting to a moderate dollar-cost-average schedule (say, six months) is a reasonable way to make a decision instead of making no decision at all.