Swarup Mohanty's AI-driven simulation showed a 6% annual withdrawal could erode principal in just three flat years, prompting a shift to 4% and a cash buffer.
Swarup Mohanty, vice-chairman and CEO of Mirae Asset Investment Managers (India), ran his 15-year retirement plan through an AI stress test. The simulation delivered an uncomfortable result: a 6% annual withdrawal rate, long considered a safe rule of thumb, would quietly erode his principal if equity markets traded flat for just three consecutive years.
The math was straightforward. In a range-bound market, withdrawals are not funded by returns. They eat into the corpus itself. Mohanty, 55, called it a "fascinating learning" that exposed mathematical flaws he had overlooked despite more than a decade of planning. The event is not just a personal recalibration. It is a live-fire demonstration of sequence-of-returns risk, the hidden variable that can wreck retirement portfolios when the timing of withdrawals collides with a sideways market.
The simple read is that a 6% withdrawal rate is too high. The better market read is that static withdrawal rules ignore the distribution of returns. A 6% rule might work over a 30-year horizon if equities deliver 8-10% annualized. But that average conceals the danger: a few flat or negative years early in retirement, when withdrawals are being taken, permanently impair the capital base. The AI simulation forced Mohanty to confront that sequence risk directly.
He cut his withdrawal assumption to 4% and built a separate three-year cash buffer. The buffer is designed to cover living expenses during weak markets, avoiding the need to redeem long-term equity investments at depressed prices. This is not a theoretical exercise. It is a direct response to a stress test that showed a 6% withdrawal rate would begin consuming principal in year one of a flat market and accelerate the damage in years two and three.
For traders and investors tracking retirement portfolios, the implication is clear: the 4% rule, popularized by the Trinity Study, is not a conservative floor. It is a starting point that must be re-stressed for current valuations, inflation expectations, and the specific asset mix. Mohanty's AI tool did not just lower the number. It showed the exact market condition that would break the old plan.
Sequence risk is the danger that the order of investment returns, not just their average, determines portfolio survival. A retiree who experiences negative returns in the first few years of withdrawals faces a much higher failure risk than one who gets those same returns later. The mechanism is simple: selling assets at lower prices to fund withdrawals reduces the number of shares that can participate in any subsequent recovery.
Mohanty's simulation quantified this for a specific scenario: three flat years for equities. In that environment, a 6% withdrawal rate would force the sale of roughly 18% of the starting equity portfolio over three years, with no offsetting gains. If the market then recovers, the portfolio is smaller and the recovery must be proportionally larger just to get back to even. The AI output made the math visible, and it was worse than intuition suggested.
The buffer strategy addresses this by decoupling spending from market prices. A three-year cash or liquid fund reserve means the retiree can wait out a typical bear market without touching equities. The cost is the opportunity cost of holding cash, but the benefit is avoiding the permanent capital loss that forced selling creates. Mohanty's move to a 4% withdrawal rate, combined with the buffer, effectively lowers the required equity return and buys time.
The stress test triggered more than a withdrawal rate change. Mohanty's portfolio had already migrated from a 60:40 equity-debt structure five years ago to a 70:15:15 split–70% equity, 15% debt (including provident fund), and 15% alternates. The higher equity allocation reflects a conviction that India's growth story is broader than the Nifty 50, and that mid- and small-cap exposure is necessary to capture it. But it also raises the sequence-risk stakes.
His equity book is roughly 75% active and 25% passive. Within active funds, large-caps account for about 40%, small-caps for 10-15%, with the balance in mid-cap and thematic funds covering healthcare, banking, defence, chemicals, and capital market themes. The passive portion is not vanilla indexing; he uses ETFs for targeted thematic exposure. This structure performed well over the last year, delivering lower double-digit returns driven by mid- and small-cap strength and a timely silver position. Over five years, annualized returns are around 13.5-14%, ahead of his 11-12% target.
But the AI stress test forced a re-examination of the downside. A 70% equity portfolio in a flat market would still generate dividends, but the capital appreciation engine would stall. The 4% withdrawal rule, applied to a corpus that must last 30-plus years, now anchors the plan. Mohanty also stepped up monthly contributions during the final accumulation years, pushing toward a revised, higher corpus target.
The 4% rule is not a guarantee. It is a probability-based guideline derived from US historical data. In an Indian context, with different inflation dynamics and equity return patterns, the safe withdrawal rate may be lower or higher depending on the period. Mohanty's AI simulation did not rely on historical averages alone. It stress-tested specific sequences, including the flat-market scenario that traditional Monte Carlo simulations might underweight.
The three-year buffer is the operational layer. It means that even if equities go nowhere for 36 months, the retiree does not sell a single share. The buffer is replenished during good years, creating a mechanical rule: when equity returns exceed the withdrawal rate, the excess refills the cash bucket. When returns fall short, the buffer absorbs the spending. This is a form of dynamic asset allocation that responds to market conditions without requiring market timing.
For someone tracking retirement risk, the next concrete marker is the buffer size relative to projected expenses. Mohanty's buffer is three years. An investor with a higher equity allocation or a longer expected retirement might need four or five. The AI simulation can be rerun with updated assumptions about inflation, longevity, and market returns. The key is that the plan is now stress-tested against a specific failure mode, not just a historical average.
Mohanty's experience is a case study in why static retirement plans fail. The 6% withdrawal rate was not reckless. It was a common assumption that worked in backtests over many periods. But it failed under a plausible, not extreme, scenario: three flat years. That scenario is not a black swan. The Nifty 50 has experienced multiple three-year periods of near-zero total returns, most recently from 2010 to 2013.
The broader lesson is that retirement portfolios need a stress test that models sequence risk explicitly. A simple Monte Carlo simulation that assumes normally distributed returns may not capture the clustering of low returns. Mohanty's AI tool appears to have used a more tailored approach, perhaps incorporating regime-switching or block bootstrapping, to surface the specific condition that would break the plan.
For traders and investors managing long-term portfolios, the actionable steps are: first, stress-test the withdrawal rate against a three-year flat equity market; second, build a cash buffer sized to cover expenses during that window; third, use asset allocation as the primary risk control, not market timing. Mohanty's rule is mechanical: if equity corrects 10%, the next investment goes into equity to restore the target allocation. If equity becomes overweight, fresh money shifts to debt or alternates. This discipline was tested during the COVID crash, when he deployed capital aggressively after a 30-35% drawdown, materially improving long-term returns.
AlphaScala's own scoring system rates UPS (United Parcel Service Inc.) at 56/100, a moderate score that reflects a mature business with stable cash flows. A stock like UPS might find a place in a retirement portfolio's dividend-income sleeve, but the real risk is not in any single holding. It is in the withdrawal architecture. The AI stress test that caught Mohanty's attention is a reminder that even a well-constructed portfolio can fail if the distribution plan is not engineered for the worst sequences.
Mohanty's insurance framework reinforces the point. He maintains ₹3.75 crore in family health cover and ₹12 crore in term insurance tied to a ₹3 crore home loan. The logic: plan for a longer life, insure for a worse one, and let asset allocation do the rest. The AI simulation added a new layer: plan for a flat market, and build a buffer so that a bad decade does not force bad decisions.
The risk event here is not a market crash. It is the quiet realization that a widely used withdrawal rule fails under a common market condition. The exposure is every retirement portfolio that assumes a static withdrawal rate without a sequence-risk buffer. The timeline is immediate for anyone within five years of retirement. The affected assets are equity mutual funds, ETFs, and any long-term portfolio that must fund regular withdrawals. What would reduce the risk is a lower withdrawal rate, a dedicated cash buffer, and a mechanical rebalancing discipline. What would make it worse is a prolonged period of low equity returns coinciding with the early years of retirement.
Mohanty's AI-driven recalibration is a template. It shows that the right question is not "What is the safe withdrawal rate?" but "Under what specific market conditions does my plan fail?" Answering that question requires a stress test, not a backtest. And the answer, as Mohanty discovered, may be uncomfortable enough to force a change.
Drafted by the AlphaScala research model and grounded in primary market data – live prices, fundamentals, SEC filings, hedge-fund holdings, and insider activity. Each story is checked against AlphaScala publishing rules before release. Educational coverage, not personalized advice.