The Emergency Fund: It’s Still Useless!

May 26, 2021

One of my earliest blog posts back in 2016 detailed my thought process for skipping the emergency fund. Back then, while we were still accumulating and saving for early retirement, our entire seven-figure equity index fund portfolio served as our emergency fund. We never kept more than $1,000, maybe $2,000 in a checking account. We’d simply used the credit card float and/or the Home Equity Line of Credit (HELOC) if any larger expenses came up that exceeded my monthly cash flow, e.g., car and home repairs, medical bills, etc.

So, again, the idea is that if you are a practitioner of the FIRE movement and you save and invest aggressively just put all your money in an equity index fund and be done. It’s not crazy at all to simply invest your emergency fund in stocks! And I repeat it again in case people misunderstand (intentionally or unintentionally) my point here, I most definitely advocate stashing a large pile of money. I simply advocate for moving all that money into an investment with high expected returns, ideally equities, instead of letting the money languish in a money market account at 0.03% interest. Please refrain from quoting the strawman sob stories about people who couldn’t afford their roof or car repair because they live paycheck to paycheck. 

But a lot has happened since. We’ve had the 2020 recession and bear market and massive equity market volatility. Many financial experts, bloggers, and podcasters started spreading the “you need x months worth of expenses stashed away in a savings/money market account” mantra again. Have I changed my opinion? Heck no! Quite the opposite! The 2020 recession is a perfect example of the lunacy of the emergency fund invested in a money market account. Let me explain why…

The uniqueness of the 2020 Recession and Bear Market

The 2020 recession likely lasted only 2 months: February to April, though the NBER hasn’t officially announced the end of the recession yet. And the Bear Market was short, too! If we look at daily data, it lasted only 33 days (Feb 19 to March 23) and using month-end S&P 500 readings, the Bear Market lasted just three months: from the December 2019 peak to the March 2020 trough.

The recovery was also relatively swift. In contrast to the previous two Bear Markets, we’ve already achieved a new all-time high after only 7 months, even after accounting for inflation. That same task took 65 months after the 2007 market peak and 153 months after the dot-com bubble burst. And yes, it’s not a typo: after the dot-com crash the stock market never even recovered until after(!) the Global Financial Crisis, if we adjust the returns for inflation. And, yes, I am factoring in dividends along the way! See my old post from 2019 about the history of bear markets. So, in a nutshell, we went through the 2020 Bear Market in 10x or even 20x “fast-forward” mode!

Cumulative returns after the 2000, 2007, and 2019 market peaks. We went through the 2020 Bear Market in “fast-forward”!

So, this recession should bode well for the folks following my 2016 advice of simply investing the emergency fund in the stock market. Let’s extend the simulation exercise I did in 2018 to include the new data to run this until April 2021. Specifically, I simulate portfolio values after a 36-month contribution period to see whether the final portfolio value does better in a money market account (growing at the 3-month T-Bill rate) or an S&P 500 index fund (e.g. FXAIX or FNILX).

I assume that both the contributions and the final portfolio value are CPI-adjusted. In the chart below, I plot that final real portfolio value. Indeed, the emergency fund invested in the S&P 500 did slightly worse in one single month in 2020. If you were unlucky enough and your car broke down in March 2020, the money market fund would have done slightly better. The drop in March 2020 was bad enough to wipe out the 36 months of gains. Ouch!

But the stock market did significantly better during all the other months post-2010. So unless your home and your car have a tendency of breaking down when the stock market is down, you would have been better off with the emergency fund invested in stocks. On average, the stock market outperformed the money market investment by about over $3 and over 10%. Over 75% of the time you would have done better with the equity portfolio.

Final portfolio value after 36 months of $1 contributions. Side note: Notice that the money market save withdrawal amounts drop below $36 occasionally. That’s possible because there were several periods of negative real interest rates!

What about 12 months of withdrawals?

We don’t want to restrict our attention to the one-time large expenditures (e.g., major car/home repairs or medical bills). An equally appropriate purpose of the emergency fund is the hedge against a job loss. In that scenario, you would likely not withdraw the funds as one large lump sum as I assumed in the simulations above (and my 2018 post). Rather, you’d withdraw regular amounts to make up for a lost paycheck. So, let’s assume now that the 36 months of contributions of $1 each are then withdrawn in 12 equal increments over the subsequent 12 months, each at the beginning of the month. Analogously to my Safe Withdrawal Rate simulations, I solve for the level withdrawal amount that precisely exhausts the emergency fund over the subsequent 12 months. See Part 8, the mathematical appendix, for the details on how that’s done in a simple matrix. Alternatively, I could have specified a fixed withdrawal amount and then simulate the two portfolios and compare the final portfolio values. Qualitatively, with very similar results.

So, here are the sustainable withdrawal amounts from the emergency fund, see the chart below. The time stamp on the y-axis is the month of the last contribution (which implies that the final x-axis date is April 2020, because we need 12 additional monthly return observations going forward!). Quite intriguingly, the stock market investment did better even during the entire 2020 event! That’s because you had 36 months of strong returns before the pandemic and then only a very brief Sequence Risk problem during 2020, followed by an impressive rebound. Thus, even if you had lost your job in March 2020, the equity-based emergency fund would have been the better option when withdrawing over a 12-month stretch!

But just to be sure, I certainly concede this: there is a consistent underperformance of the stock market relative to the money market if you need your money during a recession, 1991 and 2020 being the only exceptions. If your job security is indeed strongly correlated with the economy one could still justify a less aggressive approach. But for all others, you’re doing better with the stock market in three-quarters of the historical simulation periods!

Emergency fund with 36 contributions and 12 subsequent equal withdrawals: The stock market did much better in 2020 than the money market!

How about aggressive savers with a 50% savings rate?

For the average FIRE enthusiast, the accumulation phase should be much shorter than 36 months. If you have a 50% savings rate you need to replace only 50% of your income, so a rough estimate would be to accumulate for 12 months to get roughly 12 months’ worth of expenditures, assuming a zero real return. How does that look in the historical simulations? Let’s see the chart below. Even in that scenario, the stock market safe withdrawal amounts never dip below the money market for the entire post-housing-crisis era. So, it’s certainly true that recessions are correlated with the stock market underperformance. But overall, the stock market will beat the money market more than 75% of the time.

12 contributions followed by 12 withdrawals: the stock market still beats the money market in 2020!

Just for the record: Do not keep 100% equities if…

  • You have retired already. I recommend keeping at least a 20-25% allocation to safe assets to hedge against Sequence Risk. Though, my research on the post-retirement glidepath shows that you can indeed shift back into 100% equities again, eventually. But notice that most financial experts would prefer bonds over a money market account because 1) the yields are slightly higher, at least most of the time, and 2) there is usually a better diversification potential with bonds because interest rates go down during most recessions and thus you get a duration effect in the bond portfolio.
  • You are nearing retirement. Traditional Target Date Funds would move you out of an aggressive asset allocation already decades before your planned retirement date. I find that way too conservative. But 2-5 years before retirement might be a good time to revisit the asset allocation and probably shift into a more cautious portfolio. See my post on pre-retirement glidepaths from earlier this year!
  • You have a strong and predictable need for taking money out of your portfolio during a recession because your business and/or employment opportunities dry up during economic downturns.

When to self-insure when to buy insurance…

Update August 21, 2021: This new section was inspired by the discussion with readers Bruce Varney and ANDYG42 in the comments section below. Thanks for the comments!

The whole discussion on whether to use a money market account to save for emergencies vs. your Fidelity/Vanguard/Schwab equity index, fund boils down to the same intuition as for the question of whether to buy insurance vs. self-insure. My rule has always been:

  • Small to moderate losses: Self insure. I don’t buy extended warranties!
  • Large to catastrophic losses: Buy insurance! We have car liability insurance, health insurance, life insurance, etc.

Where does the emergency fund (self-insure) vs. money market account (buy insurance) example fall? I would put this squarely into the “small to moderate losses” category. Yeah, it’s possible that your equity-based emergency fund is worth only $7,000 or $8,000 relative to the $10,000 money market emergency fund. But a $2k-$3k loss is something I can swallow in exchange for significant average outperformance.

On the other hand, going with 100% equities in retirement could lead to a rapid depletion of your portfolio, which I would consider a catastrophic event. In retirement, you want to diversify your equity risk, even if that implies giving up some of the expected returns. Buy that insurance!


I thought I had written everything I needed to write about emergency funds. But today’s post was a good exercise because we could test my old theory with new data. I’m the first to concede that investing the emergency fund in stocks will sometimes backfire. But that (unconditional) probability was only 25% in historical simulations.

Also notice that I made it hard for the equity portfolio to beat the money market account. Specifically, I assumed that the emergency happens right after you are done accumulating your EF. Much more likely would be a situation where you finished accumulating your EF target and then the money just lingers there for another 1 to 5 years before you actually need it. With the additional time and runway for the equity portfolio to take off, you would find it even easier for the stock portfolio to beat the money market! (that said, the money market proponents will point to the possibility that you can shop around for higher rates. But that sounds like a lot of work. And keep in mind that interest income is less tax-efficient than the equity capital gains and dividends and the possibility of tax-loss harvesting! So, it’s probably a wash!)

What if you still have doubts? First of all, don’t fall for this myopic fallacy where you try to avoid the stock market loss over one single 36-month window. Your life is longer than that. Over your entire adult financial life, you are bound to have multiple financial emergencies. Sometimes you do better with the money market, most of the time you do better with equities. If you average the pros and cons over multiple instances you diversify that small risk more and more and you are even more likely to come out ahead with the equity investment.

And if you’re still not convinced, that’s fine too. If you sleep better with a five-figure amount sitting in a money market account, then so be it. I’m just a blogger, pointing out what would have been a better approach historically. I’m here to give good advice, not feel-good advice! But I won’t judge you if you ignore my ramblings! 🙂

Also, make sure you check out the posts on this topic:

Thanks for stopping by today! Looking forward to your comments!

Title picture credit:

179 thoughts on “The Emergency Fund: It’s Still Useless!

  1. Hi ERN, I appreciate you making your case for a $0 emergency fund. On a related topic, I’m wondering what you think about setting aside funds for large projects.

    For example, I need ~$50K for new siding/windows. It’s not extremely time sensitive but should be done within the next couple of years. In my case, it’s more money than I would cashflow (after living expenses paid) in a year. Assuming no LOCs available, how far in advance of anticipated payment should one sell equities to pay for something like this? My savings have always fit neatly into retirement accounts, so I would need to pull some roth contributions or do a 401k loan to pay for this.

    1. Great question.
      If you have some flexibility in the exact timing, sure, go ahead and play in the stock market. And sell when you get the bill.
      Any shortfall, fund that with future savings.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.