Stress Testing Your Allocation
Stress Testing Your Allocation
A portfolio you cannot hold through a 40% drawdown is not safe. Stress testing means taking the allocation you think you want—say, 80/20 stocks and bonds—and running it backward through the five worst crashes of the past century. The outcome shows you whether your allocation matches your actual tolerance, or whether you'll panic and sell at exactly the wrong moment.
Key takeaways
- Run your allocation through 1929, 1973, 2000, 2008, and 2020 to see peak-to-trough losses.
- A 60/40 portfolio lost around 24% in 2008; an 80/20 lost roughly 45%.
- The largest drawdown you would have accepted is a hard ceiling on equity exposure.
- Historical stress tests are backward-looking but reveal psychological limits more honestly than questionnaires.
- Build a spreadsheet or use Morningstar data to model the outcomes; don't skip this step.
What stress testing answers
Most people asked "Can you handle a 30% loss?" answer yes. Then 2022 happens—a mere 18% decline in a 60/40 portfolio—and they're checking prices every 15 minutes, reading panic articles, and seriously considering selling. This gap between the answer on a form and the answer under pressure is the reason stress testing exists.
You cannot know your true tolerance in advance. But you can model what would have happened if you'd held your current (or proposed) allocation through the five worst episodes in financial history. The point is not to predict the future. The point is to train your intuition by seeing real numbers in a real context.
The five historical stress-test scenarios
1929–1932: The Great Depression. A 100% stock portfolio (before bond alternatives were readily accessible) fell about 89% from peak to trough. Returns took 25 years to recover. This is extreme; most modern portfolios include bonds, which helped less then because government bonds had fixed rates and fell too. But a pure stock portfolio in 1929 was liquidated at losses of 85%+.
1973–1974: The Oil Crisis. Stocks fell 48%. A 60/40 portfolio (60% stocks in funds like VTSAX, 40% bonds in BND) lost about 20% from peak. A 70/30 lost roughly 26%. This was brutal partly because inflation roared ahead 12% per year, so real losses were worse, and bond returns were poor. Many retirees suffered heavily.
2000–2002: The Dot-Com Crash. The S&P 500 fell 49% (2000–2002). The NASDAQ fell 78%. But large-cap value stocks held up better. A 60/40 portfolio lost around 11% peak to trough; an 80/20 lost around 20%. International stocks and bonds helped absorb the shock. This crash taught investors that diversification works, even if imperfectly.
2008–2009: The Financial Crisis. Stocks (VTI, VTSAX) fell 57% from peak. A 60/40 portfolio lost about 24% peak to trough. An 80/20 portfolio lost around 45%. Many people who said they could handle 40% losses were shocked by 45%; several panic-sold near the bottom in March 2009. This is the stress test that breaks most allocations and most investors.
2020: The COVID Crash. A sharp but brief selloff: stocks fell 34% in five weeks (February–March 2020). A 60/40 portfolio lost about 16% peak to trough. An 80/20 lost about 26%. Recovery took four months. This crash was less painful than 2008 because it was swift and the policy response immediate. Still, holding through it required conviction.
How to run the stress test
Use historical fund returns or index data. Morningstar, FactSet, and Bloomberg all provide them. If you don't have institutional access, Yahoo Finance and FRED (Federal Reserve Economic Data) offer free historical prices for major indexes. You need:
-
Stock return: For a U.S. portfolio, use VTI (Vanguard Total Stock Market) or SPY (S&P 500). For global, use VXUS (international) and VTSAX (U.S.). Look up the price on the peak date and the price on the trough date.
-
Bond return: Use BND (Vanguard Total Bond Market) or AGG (iShares Core U.S. Aggregate Bond). Bond returns during 2008 were positive or near-flat (they rose in value as risk-off sentiment drove demand). In 1973, they were hurt less than stocks but did fall.
-
Calculate your blended loss: If your allocation is 70% stocks and 30% bonds, and stocks fell 40% while bonds fell 5%, your portfolio loss is 0.70 × (–0.40) + 0.30 × (–0.05) = –0.285 or –28.5%.
-
Make it real: This is crucial. Don't just write "–28.5%." Picture the number. In 2020, a $1 million portfolio at 70/30 would have fallen to $715,000 in five weeks. Could you hold that without panic-selling or checking daily? Most people cannot hold a 70/30 in a sudden crash.
What the numbers typically show
A 40/60 portfolio (40% stocks, 60% bonds) weathered 2008 with losses around 12–14%. Most people can stomach that. If you started at $500,000, you fell to $430,000–$440,000. Painful, but not catastrophic.
A 50/50 portfolio lost around 18% in 2008. A $500,000 portfolio fell to $410,000. More unsettling for most people.
A 60/40 portfolio lost around 24% in 2008. $500,000 fell to $380,000. Many people felt panic here.
A 70/30 portfolio lost around 31–32% in 2008. $500,000 fell to $340,000–$345,000. This is where panic-selling accelerated in real time.
An 80/20 portfolio lost around 40–45% in 2008. $500,000 fell to $275,000–$300,000. Severe enough that some investors abandoned the plan.
A 90/10 portfolio lost around 50% in 2008. $500,000 fell to $250,000. Only long-horizon, high-income investors should touch this.
The emotional dimension
The stress test reveals your ceiling by accident. Most people don't have a theoretical tolerance of 45%; they have a real tolerance of about 25–30%. When they model running an 80/20 through 2008 and see a 45% loss, they have two honest choices: reduce to 60/40 (accepting lower returns), or accept that they will panic when it happens.
Some people genuinely can hold 80/20 through a 45% crash because they have decades left to recover, they have steady income, and they've internalized that losses are temporary. Those people should run the test, see the 45%, nod, and move forward. Others should reduce equity. Both answers are correct; the stress test simply clarifies which one applies to you.
Common shortcuts and pitfalls
Pitfall: Using only bond funds that performed well in 2008. Some bond funds spiked in value during the 2008 crisis (long-duration Treasuries). Others fell. Using only the winners gives a false sense of security. Use a broad bond index like BND or the Bloomberg Aggregate Bond Index.
Pitfall: Assuming 2020 will be typical. The COVID crash was short and sharp but followed by a 60%+ rally in one year. Most crashes are longer and slower. Stress-test against 2008 or 2000, where recovery took 4–5 years, not 4 months.
Pitfall: Ignoring correlation changes. In extreme stress, correlations approach 1.0 (everything falls together). Your usual 0.3 or 0.4 correlation between stocks and bonds may not hold. Model a scenario where stocks fall 40% and bonds fall 10%, not one where bonds mysteriously rise.
Shortcut: Online calculators. Many financial advisors offer "stress test calculators" that show pretty charts. Use them for intuition, but verify the historical data. Some are accurate; others underestimate losses.
Building conviction from stress testing
The goal is not to find the "safe" allocation (there is none). The goal is to find your allocation—the one you can actually hold through a crash. If that's 50/50, good. If it's 70/30, good. If it's 30/70 because you sleep poorly and have low income, that's also good.
Run the test. See the worst outcome. Ask yourself: "If I had $500,000 and it fell to $280,000 in six months, would I still hold, or would I sell?" Your honest answer is your allocation ceiling. Work backward from there.
Related concepts
Next
Historical crashes teach us that tolerance is not static. The next article explores two thought experiments—2008 and 2020—where your actual decisions and emotions under real market stress reveal what you can truly hold.