PMS Performance Metrics: How to Judge Returns, Risk and Benchmarks
PMS performance in India follows one reporting standard. Every portfolio manager must show each strategy’s time-weighted return (TWRR) next to one benchmark, after all fees and expenses. The same figures go to SEBI and to APMI, the industry body for portfolio managers, every month.
That standard makes returns comparable. It does not answer the questions an investor in portfolio management services cares about. Did the strategy beat its own benchmark over three and five years, or only in the last twelve months? How far did it fall in a bad year, and how long did the recovery take? Did the extra return come from skill or from extra risk?
The SEBI rules and APMI data below are current at 17 October 2026. APMI’s latest table runs to 30 September 2026. The worked example uses illustrative numbers and says so.
To judge PMS performance, start with TWRR against the strategy’s own benchmark over three years or more. Check that figure on APMI’s website. Then look at the worst fall and the recovery time. Use Sharpe, Sortino, alpha and the information ratio last, always for the same period.
In this article8 sections
Where PMS performance is published, and what the rules require
Three places carry official numbers. APMI publishes a table for every investment approach. You can filter it by category (equity, debt, hybrid or multi asset) and by service type. It shows returns from one month to five years and since inception, next to the benchmark. SEBI’s Portfolio Manager Monthly Report holds each manager’s own monthly filing. The manager’s factsheet repeats the figures and adds holdings and commentary.
Two SEBI circulars set the rules behind those numbers. Since 1 April 2023, a manager that quotes a strategy’s returns must show its TWRR beside the benchmark’s trailing return (SEBI circular of 16 December 2022). SEBI’s guidelines of February 2020 already required performance net of all fees and all expenses. Cash and liquid-fund holdings count in the return.
The same rules limit what a manager can show. A manager may not quote model-portfolio returns or the record of hand-picked clients. Marketing material must compare the strategy with its benchmark and with other managers in the same category. It must also state that SEBI has not verified the figures. Website numbers must match what the manager files with SEBI.
| Source | What it shows | Use it to |
|---|---|---|
| APMI investment-approach table | TWRR and AUM for each strategy, one month to five years and since inception, with benchmark returns | Check a quoted figure and compare strategies in one category on the same date |
| SEBI Portfolio Manager Monthly Report | Each manager’s monthly filing | Confirm the manager’s figures at source |
| Manager’s factsheet | TWRR against the benchmark, holdings, sector mix, commentary | See what the strategy owns and why |
| Your own PMS statement | XIRR on your money, plus the minimum, maximum and median XIRR across all clients in that strategy | See how your account did against other clients |
Information as of October 2026. Sources: SEBI circulars of 13 February 2020 and 16 December 2022; APMI investment-approach performance page (data as on 30 September 2026).
TWRR, XIRR and your own PMS performance
TWRR measures the strategy. It removes the effect of when clients added or withdrew money, so two strategies can sit side by side on the same terms. It still includes market movement, so a strong TWRR in a rising market says little about skill on its own.
XIRR measures your account. Suppose you invested ₹50 lakh, SEBI’s minimum for a PMS, and added ₹25 lakh after a market fall. Your XIRR will differ from the strategy’s TWRR, and SEBI requires your statement to show both. It also gives the lowest, highest and median XIRR across all clients in that strategy. That spread shows how much timing moved individual results.
Benchmarks used to judge PMS performance
SEBI asked APMI to name up to three benchmarks for each of four categories, and each strategy picks one of them. APMI’s list from March 2023 reads as follows.
| Category | Benchmarks a strategy can choose |
|---|---|
| Equity | Nifty 50 TRI; BSE 500 TRI; MSEI SX 40 TRI |
| Debt | Nifty Medium to Long Duration Debt Index; a CRISIL credit index (33% AA+/AA, 17% AA-, 50% A+/A/A-, 2 to 2.5-year duration); CRISIL Composite Bond Fund Index |
| Hybrid | Nifty 50 Hybrid Composite Debt 50:50 Index; 50% BSE 500 and 50% BSE Arbitrage Rate Index; CRISIL Hybrid 50+50 Moderate Index |
| Multi asset | Three blends of equity, arbitrage, debt and other asset indices; check the exact mix in the disclosure document |
Information as of October 2026. Sources: APMI benchmark annexure (31 March 2023) and APMI performance page. TRI means total return index, which adds dividends back.
Two details change how you read the comparison. All three equity choices are broad or large-company indices, so a small-cap strategy still reports against one of them. Say small caps rallied and a small-cap PMS beat the Nifty 50 TRI. That PMS performance says nothing about whether it beat small caps. Compare its holdings with the benchmark’s, and ask what a small-cap index did over the same dates. Debt strategies have their own credit and duration benchmarks, set out in the debt PMS guide.
A benchmark choice also sticks. Before switching category or benchmark, a manager must let existing clients exit without an exit load. It then cannot use the earlier record in its reporting. If a strategy’s reported history starts later than its launch, ask whether the category or benchmark changed.
Reading 1-year, 3-year, 5-year and since-inception PMS performance
Each period answers a different question. The 1-year figure mostly reflects the latest market phase, so many strategies in a category can look alike after a strong year. Three-year and five-year figures are more likely to include a correction, and say more about the process. Since-inception returns depend on the start date: a strategy launched just before a rally can show a high number for years.
A few checks keep the comparison fair:
- Compare strategies on the same as-on date. APMI updates every month, and a factsheet one month older can show a different 1-year figure.
- Read the footnote to see whether periods over one year show annualised returns.
- Treat “NA” in APMI’s 5-year column as information: the strategy has no five-year record yet.
- Remember that trailing returns move with the end date. A 3-year return measured from a market low looks very different from one measured from a high.
Buying on the strength of the latest 12 months is the pattern behind chasing recent PMS returns. The reported figures also sit before your own capital gains tax. In a PMS the shares stay in your demat account, so the manager’s sales can create taxable gains for you. A tax professional can confirm how that applies.
Drawdown and recovery: what a CAGR hides
Maximum drawdown is the largest fall from a peak to a later low. Recovery arithmetic makes it matter: the deeper the fall, the larger the gain required just to get back to the old peak.
| Fall from peak | Gain required to recover |
|---|---|
| 10% | 11.1% |
| 20% | 25.0% |
| 30% | 42.9% |
| 40% | 66.7% |
| 50% | 100.0% |
Arithmetic: gain required = fall ÷ (1 – fall).
Ask for two numbers next to any CAGR: the worst drawdown in the record and the months the strategy took to regain its previous peak. Compare both with the benchmark’s fall over the same dates. A strategy that fell 35% while its benchmark fell 25% took more risk than the index, whatever its long-run return.
Risk-adjusted PMS performance: Sharpe, Sortino, alpha, beta and information ratio
Each ratio divides return by a different idea of risk. None of them works across different periods, so compare strategies only over the same dates and the same risk-free rate.
| Measure | What it tells you | How to read it |
|---|---|---|
| Standard deviation | How widely returns swing around the average | Higher means a bumpier ride |
| Sharpe ratio | (Return – risk-free rate) ÷ standard deviation | Return earned for each unit of total volatility |
| Sortino ratio | (Return – risk-free rate) ÷ downside deviation | A second view for investors who care most about losses |
| Beta | Sensitivity to the benchmark, from past returns | Beta 1.2: historically, when the market moved 10%, the PMS tended to move about 12% |
| Jensen’s alpha | Return above what the strategy’s beta predicted | One good year says little about skill |
| Information ratio | (Return – benchmark return) ÷ tracking error | Rule of thumb: above 0.5 suggests reasonably consistent outperformance, and above 1 is rare |
| R-squared | How closely returns follow the benchmark, 0 to 100 | Below 70: moves quite differently from the benchmark (an active strategy, or a benchmark that may not fit) |
Take a PMS that returned 14% with a 6% risk-free rate and 10% standard deviation. Its Sharpe ratio is (14 – 6) ÷ 10 = 0.8. A common rule of thumb reads a Sharpe above 1 as very good, 0.5 to 1 as acceptable and below 0.5 as weak. Treat it as a rule of thumb, not a regulatory standard, because the period and the risk-free rate change the result.
Two equity strategies with the same 12% return can still differ on Sortino. The one with smaller falls in its down months has the higher Sortino ratio. Our investing terms glossary defines Sharpe, Sortino, alpha, beta and the other risk measures in plain words.
Worked example: two strategies, one benchmark
The numbers below are illustrative, not real fund data. Two equity strategies ran over the same five years against the same benchmark, with a 6% risk-free rate.
| Annualised figure | Strategy A | Strategy B | Benchmark |
|---|---|---|---|
| Return | 16% | 15% | 12% |
| Standard deviation | 20% | 14% | 15% |
| Downside deviation | 13% | 8% | not needed |
| Beta | 1.20 | 0.85 | 1.00 |
| Tracking error | 8% | 5% | not applicable |
| Maximum drawdown | -35% | -20% | -25% |
| Measure | Strategy A | Strategy B |
|---|---|---|
| Sharpe ratio | (16 – 6) ÷ 20 = 0.50 | (15 – 6) ÷ 14 = 0.64 |
| Sortino ratio | 10 ÷ 13 = 0.77 | 9 ÷ 8 = 1.13 |
| Jensen’s alpha | 16 – (6 + 1.2 × 6) = 2.8% | 15 – (6 + 0.85 × 6) = 3.9% |
| Information ratio | (16 – 12) ÷ 8 = 0.50 | (15 – 12) ÷ 5 = 0.60 |
| Gain required to recover the worst fall | 53.8% | 25.0% |
Illustrative numbers. The benchmark’s own Sharpe ratio is (12 – 6) ÷ 15 = 0.40.
Strategy A leads on headline return by one percentage point. Per unit of risk, Strategy B earned more on every measure. It has the higher Sharpe and Sortino, more alpha once its lower beta counts, and a steadier lead over the benchmark. A fell further than the benchmark at its worst; B fell less. Both beat the benchmark’s Sharpe of 0.40. Whether an investor could sit through a 35% fall is a question no ratio answers.
Consistency: rolling returns and capture ratios
A point-to-point return depends on two dates. Rolling returns repeat the calculation for every window in the record, for example each 3-year period ending in each month. They show how often the strategy beat its benchmark and by how much. A record that beat the benchmark in most rolling windows tells a different story from one carried by a single strong stretch. If the factsheet does not show rolling returns, ask the manager for them.
Capture ratios split the record into up and down markets. If the market rises 10% and the PMS rises 12%, upside capture is 120%. If the market falls 10% and the PMS falls 6%, downside capture is 60%, so the strategy took less of the fall. Sundaram’s September 2026 product notes, for example, show since-launch downside capture of 75.1 to 109.7 across its four Sundaram PMS equity strategies. SEBI’s monthly reporting format also asks for each strategy’s portfolio turnover over one month and one year. Higher turnover means more trading costs, which already sit inside the net return, and more frequent taxable sales for you.
What a PMS factsheet shows, and what to ask for
When a factsheet shows PMS performance, SEBI’s rules require three things. These are the TWRR next to the selected benchmark, a comparison with other managers in the same category, and a note that SEBI has not verified the figures. Several things a decision needs sit outside it.
| Usually on the factsheet | Ask for separately |
|---|---|
| TWRR against the benchmark for several periods, with an as-on date | Your own XIRR, if you already invest |
| Holdings and sector or market-cap mix on one date | Worst drawdown and months to recover |
| Strategy description and the manager’s commentary | Rolling returns against the benchmark |
| Risk ratios, in some factsheets | The period and risk-free rate behind each ratio |
| The benchmark name and the inception date | What the fee would have cost on your amount |
Reported TWRR is net of fees, but your own cost depends on the fee option you pick, the hurdle and the high-water mark. The PMS fee structure explainer covers how those work, and the PMS fee calculator puts them in rupees.
What investors ask about PMS performance
What is a good Sharpe ratio for a PMS?
Above 1 reads as very good on the common rule of thumb, and below 0.5 as weak. Compare it with the benchmark’s Sharpe over the same period and with the same risk-free rate. Without those two details, the PMS performance figure means little.
Is PMS performance shown before or after fees?
After. SEBI requires portfolio managers to report performance net of all fees and all expenses. Your own return can still differ from the strategy’s because of when you invested and the fee option you chose.
Why does my PMS return differ from the factsheet?
The factsheet shows the strategy’s TWRR, while your statement shows your XIRR, which depends on the timing of your money. The minimum, maximum and median XIRR on your statement show where your account sits among other clients.
What makes a good PMS?
No single figure decides it. Investors usually look for a record across at least one full market cycle against the strategy’s own benchmark. They also look for falls they could hold through, a process that has stayed the same, and costs they understand. The PMS selection checklist sets these out.
How often should PMS performance be reviewed?
Managers report every month, so APMI’s figures change monthly. A quarterly review against the benchmark catches problems without reacting to one month’s noise, while judging the manager takes three years or more.
Can a PMS with high returns still be risky?
Yes. High returns can come with high volatility, concentrated positions or deep drawdowns. The worst fall and the recovery time belong next to the return.
A PMS return means little until you know its benchmark, its period and the falls along the way. In the example, Strategy A’s extra point of return came with a fall that needed a 53.8% gain to repair, against 25% for Strategy B.