Every crash looks different while it is happening. Afterwards the question is always the same.
What did you do?
The past seven months have given investors in Asia and the US plenty of material, and the tariff shock of the past two weeks is still unfolding as we write. This is not a post about predicting what comes next. It is about the one document that determines what you do when it comes.
A market shock tests the plan you wrote before it, not the plan you wish you had.
Three shocks, three lessons
24 to 30 September 2024: China's stimulus rally. After the central bank's stimulus package on 24 September, Shanghai's main index rose 4.15% that day. By 30 September the CSI 300 had risen 8.5% in a single session and more than 25% over five days. Investors who chased it were reportedly met with sharp volatility afterwards. Lesson: a sentiment spike is a timing test. You cannot tell in advance whether a rise is the start of something or the end of a move.
18 March 2025: Indonesia halts trading. The Jakarta Composite fell more than 5%, and the exchange halted trading for about 30 minutes, after sharp falls in two heavily weighted stocks. Lesson: in a concentrated market, a few names can move the whole index. A halt is a reminder that you may not be able to sell when you want to.
2 to 9 April 2025: the tariff shock. After the US announced sweeping tariffs, the S&P 500 fell more than 12% in seven days. On 9 April, a pause on parts of the plan produced a 9.5% one-day gain. Taiwan's index fell a record 9.7% on 7 April, and Indonesia's closed down about 8.5% on 8 April, its lowest since June 2021. Lesson: the best days came right after the worst. Selling on the 8th meant missing the 9th, and nobody could have known on the 8th.
What the plan has to say
The standard answer is the Investment Policy Statement (IPS): a written document that fixes the objectives and constraints of the portfolio, and is kept separate from market views. A useful IPS for an individual has two halves.
Objectives.
- Return objective: what the money needs to earn to meet its goals.
- Risk objective: two parts. Ability to take risk is arithmetic: time horizon, income, spending needs, other assets. Willingness is psychological: how much loss you can bear without abandoning the plan. The lower of the two governs.
Constraints:
- Liquidity (cash needs)
- Time horizon
- Taxes
- Legal and regulatory
- Unique circumstances
Then come the rules that actually guide behavior in a crisis:
- Strategic allocation: the target mix and acceptable ranges for each asset class.
- Rebalancing rule: the trigger (a calendar date or a band, such as 5 percentage points from target) and the process.
- Limits: the maximum for any single position, sector or country.
- What you do not do: "No leverage." "No selling because of a headline." "No new positions outside this list."
Why a rule beats a resolution
Behavioral finance explains why the rules must be written down.
Loss aversion. A loss feels roughly twice as bad as an equal gain feels good. In a drawdown, the urge to stop the pain is a systematic bias, not a signal.
The disposition effect. Investors sell winners too early and hold losers too long. A rebalancing rule forces the reverse, selling what has run up and buying what has fallen.
Recency and herding. What just happened feels like what will happen next, especially if everyone around you agrees.
A pre-committed rebalancing band is a mechanical way to buy after a fall and trim after a rise, which is what the biases make hardest to do. It is not a forecast. It is a way of keeping the allocation you chose.
A worked example
Take a portfolio of 60% equities and 40% bonds with a 5-point rebalancing band. Equities fall 25% and bonds are flat. The portfolio is now 45 in equities and 40 in bonds, so the equity weight has drifted to about 53% (45 of 85). That breaches the 5-point band. The rule says to sell 6 of bonds and buy 6 of equities to return to 60/40. It requires buying near the low.
Whether prices fall further next week is unknowable. The rule does not depend on knowing. It depends on having decided, in advance, that the equity weight you wanted was 60%.
Practical implications
- Write the IPS in one page. Objectives, constraints, target ranges, rebalancing rule, and the list of things you will not do.
- Measure willingness honestly. Ask what drawdown you would have to see in a year before you wanted to sell everything. Use the smaller of that and your ability to bear loss.
- Choose bands or dates, not moods. Rebalance quarterly or when an asset class drifts by a set amount.
- Pre-decide the leverage answer. The Taiwan margin-loan numbers show who gets forced out first. If you borrow to invest, write down the loan-to-value at which you de-risk.
- Review after, not during. After a shock, check whether the plan worked. Change it in calm, not in the middle.
Update, October 2026: what the next 18 months added
The tariff episode resolved faster than most expected. According to the Bank for International Settlements, the S&P 500 recovered its losses by the end of April 2025 and passed its previous peak in just under 20 weeks. The lesson held: the investors who had a rule to follow were not the ones forced to decide in the worst week.
Three more shocks followed, each testing a different part of the plan.
27 to 30 January 2026: Indonesia and MSCI. The Jakarta Composite lost more than 11.5% in two days, with three trading halts, after the index provider flagged concerns about transparency and free float. Lesson: some shocks are about the plumbing of the market rather than the companies in it. An index fund cannot avoid them.
17 July 2026: Taiwan's record point drop. The main index fell 2,953.71 points, or 6.47%, in a day, with the largest company down more than 7% after a margin warning. Lesson: with a single company so large, a sector view is a position. Leverage turns it into a deadline.
16 to 17 September 2026: the Fed raises. The US Federal Reserve raised rates to a 3.75% to 4.00% range, the first increase since 2023 according to coverage of the decision, after August inflation of 3.4% year on year (reported). Cash yields and borrowing costs both repriced. Lesson: a change in the rate regime tests whether your plan was built for one regime or for several.
What we could not confirm
Several market-level figures are from secondary sources: the Indonesian one-day close on 8 April 2025, the Fed's September 2026 decision and the 3.4% inflation figure (from news summaries, not the Bureau of Labor Statistics), and some Chinese rally figures. The body of this post reflects what was known on 15 April 2025; later developments are in the update section. The rebalancing example is an illustration, not a recommendation. This is education, not investment advice.
Sources
- Bank for International Settlements, Quarterly Review, September 2025
- Taipei Times, record fall on 7 April 2025; Xinhua, Indonesia trading halt and Binus Finance on 8 April 2025
- Asia Times and Daily Sabah on the September 2024 China rally
- Malay Mail, IDX chief quits after market plunge (30 January 2026)
- Focus Taiwan, 17 July 2026 close
- Federal Reserve, implementation note, 16 September 2026