ToolsSIP check
Every SIP number, in full
The tables behind the SIP reference page: every one of the 141 portfolios from Apr 2005, after costs and tax, the popular refinements tested in detail, the check against published SIP studies, and the first five predictions, graded. Start with the reference page; come here for a number it does not show.
Written for SIP investors and the advisers who explain SIPs to them. Data to 8 Oct 2026.
Every portfolio
Each portfolio was run as Rs 10,000 a month, bought at the first session of each month, held for a fixed number of years and then sold in full, after fund costs and tax (with the annual equity exemption on, since a small investor is who it is for). Every start month from Apr 2005 is one window. Windows a month apart share almost all their months, so read the independent periods before reading any percentile.
What a 3-, 5- and 10-year SIP returned
After-tax XIRR across start months from Apr 2005, Rs 10,000 a month, per cent a year. Equity indices, then gold and G-sec, then mixes.
How to read it. The dot is the median window, the bar the middle half of windows. The shaded column is the same instalments in fixed deposits. On the right, the separate periods behind each row; hollow rows rest on a single one.
What a 3-, 5- and 10-year SIP returned. 3 horizons, 15 bands. Nifty 50: median +12.2%, 7 periods; Nifty 100: median +12.7%, 7 periods; Nifty 500: median +13.8%, 7 periods; Nifty LargeMidcap 250: median +16.1%, 7 periods; Nifty Midcap 150: median +19.3%, 7 periods; Nifty Smallcap 250: median +17.9%, 7 periods; Gold: median +9.8%, 7 periods; G-sec, 5-year: median +6.1%, 7 periods; 60/40: median +10.8%, 7 periods; 60/20/20: median +11.4%, 7 periods; Equal thirds: median +9.5%, 7 periods; Four assets, equal quarters: median +13.0%, 7 periods; Four assets, 80% equity: median +15.4%, 7 periods; Four assets, 60% equity: median +13.3%, 7 periods; Four assets, 40% equity: median +11.2%, 7 periods. Fixed deposit: median +4.9%.
What tax took from a ten-year SIP
Median ten-year SIP XIRR, before and after tax, per cent a year
How to read it. Each row runs from the before-tax median (grey) to the after-tax median. Gold lost most to tax (0.6 points) and Nifty 50 least (0.1). The tax rules by date are in the method note.
What tax took from a ten-year SIP. Range chart of 15 items. Before tax: highest Nifty Midcap 150 at 17.4%, lowest G-sec, 5-year at 7.8%.
Historical research on index portfolios, not investment advice. Returns are after fund costs; 'after tax' applies Indian capital-gains rules by date for a 30% slab investor and sells everything at the end. Most factor, mid-cap and small-cap index history before each index's launch date is back-tested by NSE. One 21-year sample, domestic assets only. The fixed-deposit comparison uses banks' 1-3 year deposit rate (RBI Handbook Table 74 to July 2010); from August 2010 the 364-day T-bill yield stands in until RBI's weekly series is loaded, interest taxed each year at the 30% slab plus cess.
The numbers: five years
After-tax XIRR across start months, per cent a year. 'Below FD' and 'below inflation' are shares of overlapping windows, not probabilities. Click a name for that portfolio's own SIP chart.
| Portfolio | Windows | Independent periods | 5th pct | Median | 95th pct | Worst (start) | Below FD | Below inflation | More |
|---|---|---|---|---|---|---|---|---|---|
| Nifty Midcap 150 | 199 | 4 | 3.6% | 17.9% | 28.0% | −8.1% (2015-04) | 10% | 17% | |
| Four-factor blend | 199 | 4 | 6.2% | 16.2% | 25.4% | −4.9% (2015-04) | 3% | 4% | |
| Nifty Smallcap 250 | 199 | 4 | −2.9% | 16.7% | 30.2% | −17.8% (2015-04) | 25% | 28% | |
| Nifty 500 | 199 | 4 | 4.0% | 13.4% | 20.0% | −6.1% (2015-04) | 11% | 20% | |
| Nifty 50 | 199 | 4 | 4.3% | 12.2% | 17.6% | −4.7% (2015-04) | 8% | 19% | |
| 60/20/20 | 199 | 4 | 7.4% | 11.4% | 16.6% | 1.1% (2015-04) | 2% | 11% | |
| 60/40 | 199 | 4 | 5.5% | 11.2% | 14.9% | 0.6% (2015-04) | 6% | 20% | |
| Equal thirds | 199 | 4 | 6.6% | 10.5% | 15.8% | 4.5% (2015-04) | 1% | 8% | |
| All Weather (Indian approximation) | 199 | 4 | 6.6% | 8.9% | 12.3% | 4.2% (2015-04) | 1% | 21% | |
| Dual momentum | 199 | 4 | 3.2% | 8.9% | 18.9% | 1.5% (2014-05) | 14% | 21% | |
| Permanent portfolio | 199 | 4 | 6.2% | 8.8% | 13.1% | 4.5% (2015-04) | 0% | 15% | |
| GTAA (10-month trend) | 199 | 4 | 4.3% | 8.0% | 13.8% | 3.3% (2014-05) | 13% | 17% |
The numbers: ten years
As above, ten-year SIPs.
| Portfolio | Windows | Independent periods | 5th pct | Median | 95th pct | Worst (start) | Below FD | Below inflation | More |
|---|---|---|---|---|---|---|---|---|---|
| Nifty Midcap 150 | 139 | 2 | 11.5% | 16.9% | 19.9% | 5.2% (2010-04) | 0% | 1% | |
| Four-factor blend | 139 | 2 | 10.2% | 15.3% | 18.2% | 5.6% (2010-04) | 0% | 0% | |
| Nifty Smallcap 250 | 139 | 2 | 6.7% | 14.5% | 19.0% | −1.1% (2010-04) | 3% | 4% | |
| Nifty 500 | 139 | 2 | 9.1% | 12.8% | 15.6% | 3.4% (2010-04) | 1% | 1% | |
| Nifty 50 | 139 | 2 | 8.5% | 12.0% | 14.5% | 3.4% (2010-04) | 1% | 1% | |
| 60/20/20 | 139 | 2 | 9.0% | 11.1% | 13.7% | 5.6% (2010-04) | 0% | 0% | |
| 60/40 | 139 | 2 | 9.0% | 10.9% | 12.5% | 5.9% (2010-04) | 0% | 0% | |
| Equal thirds | 139 | 2 | 8.1% | 9.8% | 13.5% | 6.5% (2010-04) | 0% | 0% | |
| All Weather (Indian approximation) | 139 | 2 | 7.9% | 9.0% | 10.3% | 6.5% (2010-04) | 0% | 0% | |
| Dual momentum | 139 | 2 | 6.7% | 8.9% | 15.4% | 6.2% (2013-11) | 0% | 1% | |
| Permanent portfolio | 139 | 2 | 7.5% | 8.6% | 11.2% | 6.3% (2010-04) | 0% | 0% | |
| GTAA (10-month trend) | 139 | 2 | 5.6% | 7.8% | 10.7% | 5.4% (2009-08) | 0% | 0% |
A mix shrank the bad tail more than the middle. At five years, 60/20/20's 5th percentile was 7.4% against 4.0% for the Nifty 500, for a median 2.0 points lower.
All 141 portfolios at ten years, sorted by median after-tax XIRR. Click a column to re-sort.
| Portfolio | Windows | Independent periods | 5th pct | Median | 95th pct | Worst (start) | Below FD | Below inflation | More |
|---|---|---|---|---|---|---|---|---|---|
| Momentum (Nifty500 Momentum 50) | 139 | 2 | 15.8% | 19.5% | 24.0% | 11.5% (2010-04) | 0% | 0% | |
| Alpha (Nifty Alpha 50) | 139 | 2 | 14.2% | 18.7% | 25.3% | 9.1% (2010-04) | 0% | 0% | |
| Momentum (Nifty200 Momentum 30) | 139 | 2 | 13.8% | 17.8% | 22.0% | 11.9% (2016-10) | 0% | 0% | |
| Multi-factor (Nifty500 Multifactor MQVLv 50) | 139 | 2 | 12.6% | 16.9% | 21.3% | 8.6% (2010-04) | 0% | 0% | |
| Nifty Midcap 150 | 139 | 2 | 11.5% | 16.9% | 19.9% | 5.2% (2010-04) | 0% | 1% | |
| Value and momentum | 139 | 2 | 8.7% | 15.9% | 20.8% | 3.4% (2010-04) | 1% | 1% | |
| Multi-factor (Nifty Alpha Quality Value Low-Volatility 30) | 139 | 2 | 12.4% | 15.8% | 18.8% | 8.3% (2010-04) | 0% | 0% | |
| Four-factor blend | 139 | 2 | 10.2% | 15.3% | 18.2% | 5.6% (2010-04) | 0% | 0% | |
| Quality (Nifty500 Quality 50) | 139 | 2 | 10.9% | 14.8% | 19.0% | 6.5% (2010-04) | 0% | 0% | |
| Nifty Smallcap 250 | 139 | 2 | 6.7% | 14.5% | 19.0% | −1.1% (2010-04) | 3% | 4% | |
| Nifty 500 + 30% momentum | 139 | 2 | 11.0% | 14.4% | 17.5% | 6.1% (2010-04) | 0% | 0% | |
| Nifty LargeMidcap 250 | 139 | 2 | 10.5% | 14.4% | 17.0% | 4.5% (2010-04) | 1% | 1% | |
| Nifty Next 50 | 139 | 2 | 10.6% | 14.3% | 18.5% | 6.1% (2010-04) | 0% | 0% | |
| Quality and low volatility | 139 | 2 | 10.8% | 14.3% | 17.6% | 7.4% (2010-04) | 0% | 0% | |
| Low volatility (Nifty Low Volatility 50) | 139 | 2 | 11.2% | 14.2% | 17.1% | 7.7% (2010-04) | 0% | 0% | |
| Low volatility (Nifty500 Low Volatility 50) | 139 | 2 | 11.5% | 14.2% | 16.5% | 7.4% (2010-04) | 0% | 0% | |
| Quality (Nifty200 Quality 30) | 139 | 2 | 10.6% | 14.1% | 19.3% | 8.0% (2010-04) | 0% | 0% | |
| Nifty500 Multicap 50:25:25 | 139 | 2 | 9.4% | 14.0% | 16.7% | 3.0% (2010-04) | 1% | 1% | |
| Value (Nifty500 Value 50) | 139 | 2 | 0.9% | 14.0% | 21.1% | −5.0% (2010-04) | 11% | 12% | |
| Low volatility (Nifty100 Low Volatility 30) | 139 | 2 | 11.0% | 13.9% | 16.5% | 6.7% (2010-04) | 0% | 0% | |
| Nifty500 Equal Weight | 139 | 2 | 6.0% | 13.9% | 17.6% | −1.1% (2010-04) | 3% | 4% | |
| Four assets, 80% equity | 139 | 2 | 9.5% | 13.8% | 16.0% | 4.6% (2010-04) | 1% | 1% | |
| Nifty 500 + 30% quality | 139 | 2 | 10.2% | 13.3% | 15.6% | 4.9% (2010-04) | 1% | 1% | |
| Nifty 500 + 30% low volatility | 139 | 2 | 9.9% | 13.2% | 15.5% | 4.5% (2010-04) | 1% | 1% | |
| Nifty 500 + 30% value | 139 | 2 | 6.8% | 13.0% | 17.2% | 0.8% (2010-04) | 2% | 4% | |
| Defensive asset allocation (DAA) | 139 | 2 | 8.4% | 12.9% | 16.2% | 6.2% (2010-04) | 0% | 0% | |
| Nifty100 Equal Weight | 139 | 2 | 7.5% | 12.8% | 15.4% | 1.8% (2010-04) | 1% | 1% | |
| 100% Nifty 500 / 0% G-sec | 139 | 2 | 9.1% | 12.8% | 15.6% | 3.4% (2010-04) | 1% | 1% | |
| 100/0/0 equity / G-sec / gold | 139 | 2 | 9.1% | 12.8% | 15.6% | 3.4% (2010-04) | 1% | 1% | |
| Nifty 500 | 139 | 2 | 9.1% | 12.8% | 15.6% | 3.4% (2010-04) | 1% | 1% | |
| 90/0/10 equity / G-sec / gold | 139 | 2 | 9.2% | 12.6% | 15.1% | 4.0% (2010-04) | 1% | 1% | |
| Four assets, 60% equity | 139 | 2 | 9.2% | 12.6% | 15.1% | 5.7% (2010-04) | 0% | 0% | |
| Nifty 100 | 139 | 2 | 9.3% | 12.4% | 14.7% | 3.9% (2010-04) | 1% | 1% | |
| CPPI | 139 | 2 | 7.6% | 12.3% | 15.6% | 3.2% (2010-04) | 1% | 4% | |
| 90% Nifty 500 / 10% G-sec | 139 | 2 | 9.3% | 12.3% | 14.8% | 4.2% (2010-04) | 1% | 1% | |
| 90/10/0 equity / G-sec / gold | 139 | 2 | 9.3% | 12.3% | 14.8% | 4.2% (2010-04) | 1% | 1% | |
| 80/10/10 equity / G-sec / gold | 139 | 2 | 9.1% | 12.2% | 14.3% | 4.7% (2010-04) | 1% | 1% | |
| Four assets, equal quarters | 139 | 2 | 8.7% | 12.2% | 14.9% | 5.7% (2010-04) | 0% | 0% | |
| 80/0/20 equity / G-sec / gold | 139 | 2 | 9.0% | 12.1% | 15.6% | 4.5% (2010-04) | 1% | 1% | |
| Nifty 50 | 139 | 2 | 8.5% | 12.0% | 14.5% | 3.4% (2010-04) | 1% | 1% | |
| 80% Nifty 500 / 20% G-sec | 139 | 2 | 9.2% | 11.9% | 14.0% | 4.9% (2010-04) | 1% | 1% | |
| 80/20/0 equity / G-sec / gold | 139 | 2 | 9.2% | 11.9% | 14.0% | 4.9% (2010-04) | 1% | 1% | |
| Value (Nifty200 Value 30) | 139 | 2 | −0.4% | 11.7% | 21.0% | −6.7% (2010-04) | 14% | 15% | |
| 70/20/10 equity / G-sec / gold | 139 | 2 | 9.0% | 11.7% | 13.6% | 5.3% (2010-04) | 0% | 1% | |
| 70/10/20 equity / G-sec / gold | 139 | 2 | 9.0% | 11.6% | 14.7% | 5.1% (2010-04) | 0% | 1% | |
| 70/0/30 equity / G-sec / gold | 139 | 2 | 8.9% | 11.6% | 15.6% | 4.9% (2010-04) | 1% | 1% | |
| Nifty50 Equal Weight | 139 | 2 | 5.7% | 11.5% | 15.1% | −0.6% (2010-04) | 4% | 5% | |
| 70% Nifty 500 / 30% G-sec | 139 | 2 | 9.1% | 11.4% | 13.3% | 5.4% (2010-04) | 0% | 0% | |
| 70/30/0 equity / G-sec / gold | 139 | 2 | 9.1% | 11.4% | 13.3% | 5.4% (2010-04) | 0% | 0% | |
| Endowment-style (Indian approximation) | 139 | 2 | 8.7% | 11.3% | 13.6% | 6.0% (2010-04) | 0% | 0% | |
| Protective asset allocation (PAA2) | 139 | 2 | 8.1% | 11.2% | 13.8% | 7.7% (2010-04) | 0% | 0% | |
| Four assets, 40% equity | 139 | 2 | 8.8% | 11.2% | 14.4% | 6.5% (2010-04) | 0% | 0% | |
| 60/0/40 equity / G-sec / gold | 139 | 2 | 8.5% | 11.1% | 16.3% | 5.3% (2010-04) | 0% | 0% | |
| 60/30/10 equity / G-sec / gold | 139 | 2 | 8.9% | 11.1% | 12.8% | 5.7% (2010-04) | 0% | 0% | |
| 60/20/20 equity / G-sec / gold | 139 | 2 | 9.0% | 11.1% | 13.7% | 5.6% (2010-04) | 0% | 0% | |
| 60/20/20 | 139 | 2 | 9.0% | 11.1% | 13.7% | 5.6% (2010-04) | 0% | 0% | |
| 60/10/30 equity / G-sec / gold | 139 | 2 | 8.8% | 11.1% | 14.9% | 5.5% (2010-04) | 0% | 0% | |
| Black-Litterman, no views | 139 | 2 | 9.0% | 11.0% | 12.6% | 5.9% (2010-04) | 0% | 0% | |
| 60% Nifty 500 / 40% G-sec | 139 | 2 | 9.0% | 10.9% | 12.5% | 5.9% (2010-04) | 0% | 0% | |
| 60/40/0 equity / G-sec / gold | 139 | 2 | 9.0% | 10.9% | 12.5% | 5.9% (2010-04) | 0% | 0% | |
| 60/40 | 139 | 2 | 9.0% | 10.9% | 12.5% | 5.9% (2010-04) | 0% | 0% | |
| Three-fund (domestic version) | 139 | 2 | 9.0% | 10.9% | 12.5% | 5.9% (2010-04) | 0% | 0% | |
| Vigilant asset allocation (VAA-G4) | 139 | 2 | 8.0% | 10.9% | 12.9% | 7.1% (2010-04) | 0% | 0% | |
| 50/0/50 equity / G-sec / gold | 139 | 2 | 8.1% | 10.9% | 17.0% | 5.6% (2010-04) | 0% | 0% | |
| 50/10/40 equity / G-sec / gold | 139 | 2 | 8.4% | 10.7% | 15.6% | 5.8% (2010-04) | 0% | 0% | |
| 50/20/30 equity / G-sec / gold | 139 | 2 | 8.6% | 10.6% | 14.2% | 5.9% (2010-04) | 0% | 0% | |
| 50/30/20 equity / G-sec / gold | 139 | 2 | 8.8% | 10.6% | 12.9% | 6.1% (2010-04) | 0% | 0% | |
| 50/40/10 equity / G-sec / gold | 139 | 2 | 8.9% | 10.6% | 12.0% | 6.2% (2010-04) | 0% | 0% | |
| Golden butterfly | 139 | 2 | 7.8% | 10.5% | 12.5% | 5.6% (2010-04) | 0% | 0% | |
| 40/0/60 equity / G-sec / gold | 139 | 2 | 7.3% | 10.4% | 17.9% | 5.8% (2010-04) | 0% | 0% | |
| Valuation glide (CAPE) | 139 | 2 | 8.6% | 10.4% | 12.2% | 7.8% (2016-10) | 0% | 0% | |
| 50% Nifty 500 / 50% G-sec | 139 | 2 | 8.9% | 10.4% | 11.8% | 6.3% (2010-04) | 0% | 0% | |
| 50/50/0 equity / G-sec / gold | 139 | 2 | 8.9% | 10.4% | 11.8% | 6.3% (2010-04) | 0% | 0% | |
| 40/10/50 equity / G-sec / gold | 139 | 2 | 7.7% | 10.3% | 16.4% | 6.0% (2010-04) | 0% | 0% | |
| 40/20/40 equity / G-sec / gold | 139 | 2 | 8.1% | 10.2% | 14.9% | 6.2% (2010-04) | 0% | 0% | |
| Volatility-managed equity | 139 | 2 | 6.1% | 10.2% | 13.0% | 3.8% (2010-04) | 4% | 11% | |
| 40/40/20 equity / G-sec / gold | 139 | 2 | 8.6% | 10.1% | 12.1% | 6.5% (2010-04) | 0% | 0% | |
| 40/30/30 equity / G-sec / gold | 139 | 2 | 8.4% | 10.1% | 13.5% | 6.3% (2010-04) | 0% | 0% | |
| 40/50/10 equity / G-sec / gold | 139 | 2 | 8.8% | 10.1% | 11.2% | 6.6% (2010-04) | 0% | 0% | |
| Bold asset allocation (BAA-G4) | 139 | 2 | 8.4% | 10.0% | 11.8% | 8.0% (2013-04) | 0% | 0% | |
| 40% Nifty 500 / 60% G-sec | 139 | 2 | 8.7% | 9.8% | 10.9% | 6.7% (2010-04) | 0% | 0% | |
| 40/60/0 equity / G-sec / gold | 139 | 2 | 8.7% | 9.8% | 10.9% | 6.7% (2010-04) | 0% | 0% | |
| Equal thirds | 139 | 2 | 8.1% | 9.8% | 13.5% | 6.5% (2010-04) | 0% | 0% | |
| 30/0/70 equity / G-sec / gold | 139 | 2 | 6.5% | 9.8% | 18.3% | 6.0% (2009-05) | 0% | 0% | |
| 30/20/50 equity / G-sec / gold | 139 | 2 | 7.3% | 9.7% | 15.8% | 6.4% (2010-04) | 0% | 0% | |
| 1/N (monthly) | 139 | 2 | 8.1% | 9.7% | 13.4% | 6.5% (2010-04) | 0% | 0% | |
| 30/10/60 equity / G-sec / gold | 139 | 2 | 6.8% | 9.7% | 17.0% | 6.2% (2010-04) | 0% | 0% | |
| 30/30/40 equity / G-sec / gold | 139 | 2 | 7.7% | 9.6% | 14.3% | 6.5% (2010-04) | 0% | 0% | |
| Ensemble: equal weight of models | 139 | 2 | 8.0% | 9.6% | 10.8% | 6.7% (2010-04) | 0% | 0% | |
| 30/60/10 equity / G-sec / gold | 139 | 2 | 8.5% | 9.6% | 10.4% | 7.0% (2010-04) | 0% | 0% | |
| 30/50/20 equity / G-sec / gold | 139 | 2 | 8.3% | 9.6% | 11.4% | 6.8% (2010-04) | 0% | 0% | |
| 30/40/30 equity / G-sec / gold | 139 | 2 | 8.1% | 9.6% | 12.8% | 6.7% (2010-04) | 0% | 0% | |
| 30% Nifty 500 / 70% G-sec | 139 | 2 | 8.3% | 9.3% | 10.2% | 7.1% (2010-04) | 0% | 0% | |
| 30/70/0 equity / G-sec / gold | 139 | 2 | 8.3% | 9.3% | 10.2% | 7.1% (2010-04) | 0% | 0% | |
| 20/0/80 equity / G-sec / gold | 139 | 2 | 5.8% | 9.2% | 18.6% | 5.1% (2009-05) | 1% | 3% | |
| 20/10/70 equity / G-sec / gold | 139 | 2 | 6.2% | 9.1% | 17.4% | 5.6% (2009-05) | 0% | 0% | |
| Valuation glide (yield gap) | 139 | 2 | 8.2% | 9.1% | 10.3% | 7.2% (2010-04) | 0% | 0% | |
| 60/40 with trend | 139 | 2 | 6.8% | 9.1% | 10.8% | 6.4% (2016-10) | 0% | 0% | |
| 20/30/50 equity / G-sec / gold | 139 | 2 | 6.9% | 9.1% | 14.9% | 6.5% (2009-05) | 0% | 0% | |
| 20/20/60 equity / G-sec / gold | 139 | 2 | 6.6% | 9.1% | 16.2% | 6.0% (2009-05) | 0% | 0% | |
| 20/60/20 equity / G-sec / gold | 139 | 2 | 7.9% | 9.0% | 10.7% | 7.1% (2010-04) | 0% | 0% | |
| 20/40/40 equity / G-sec / gold | 139 | 2 | 7.3% | 9.0% | 13.6% | 6.8% (2010-04) | 0% | 0% | |
| All Weather (Indian approximation) | 139 | 2 | 7.9% | 9.0% | 10.3% | 6.5% (2010-04) | 0% | 0% | |
| 20/50/30 equity / G-sec / gold | 139 | 2 | 7.6% | 9.0% | 12.1% | 7.0% (2010-04) | 0% | 0% | |
| 20/70/10 equity / G-sec / gold | 139 | 2 | 8.1% | 9.0% | 9.5% | 7.3% (2010-04) | 0% | 0% | |
| Ensemble: inverse volatility of models | 139 | 2 | 7.7% | 9.0% | 9.8% | 6.8% (2010-04) | 0% | 0% | |
| Dual momentum | 139 | 2 | 6.7% | 8.9% | 15.4% | 6.2% (2013-11) | 0% | 1% | |
| 20% Nifty 500 / 80% G-sec | 139 | 2 | 7.7% | 8.7% | 9.5% | 7.1% (2016-10) | 0% | 0% | |
| 20/80/0 equity / G-sec / gold | 139 | 2 | 7.7% | 8.7% | 9.5% | 7.1% (2016-10) | 0% | 0% | |
| Permanent portfolio | 139 | 2 | 7.5% | 8.6% | 11.2% | 6.3% (2010-04) | 0% | 0% | |
| 10/0/90 equity / G-sec / gold | 139 | 2 | 4.8% | 8.5% | 19.0% | 4.0% (2009-05) | 9% | 20% | |
| 10/10/80 equity / G-sec / gold | 139 | 2 | 5.3% | 8.5% | 17.7% | 4.6% (2009-05) | 6% | 14% | |
| 10/30/60 equity / G-sec / gold | 139 | 2 | 6.0% | 8.5% | 15.2% | 5.5% (2009-05) | 0% | 1% | |
| 10/40/50 equity / G-sec / gold | 139 | 2 | 6.4% | 8.5% | 14.0% | 6.0% (2009-05) | 0% | 0% | |
| 10/70/20 equity / G-sec / gold | 139 | 2 | 7.4% | 8.5% | 9.9% | 7.2% (2008-10) | 0% | 0% | |
| 10/20/70 equity / G-sec / gold | 139 | 2 | 5.6% | 8.4% | 16.4% | 5.1% (2009-05) | 1% | 7% | |
| Inverse volatility | 139 | 2 | 7.6% | 8.4% | 9.0% | 7.4% (2013-03) | 0% | 0% | |
| 10/60/30 equity / G-sec / gold | 139 | 2 | 7.1% | 8.4% | 11.4% | 6.9% (2009-05) | 0% | 0% | |
| 10/50/40 equity / G-sec / gold | 139 | 2 | 6.8% | 8.4% | 12.8% | 6.5% (2009-05) | 0% | 0% | |
| Equal risk contribution | 139 | 2 | 7.4% | 8.4% | 8.9% | 7.2% (2012-10) | 0% | 0% | |
| 10/80/10 equity / G-sec / gold | 139 | 2 | 7.5% | 8.3% | 8.8% | 7.3% (2013-03) | 0% | 0% | |
| Maximum diversification | 139 | 2 | 7.3% | 8.2% | 8.9% | 7.0% (2010-04) | 0% | 0% | |
| 10% Nifty 500 / 90% G-sec | 139 | 2 | 7.1% | 8.1% | 8.8% | 6.4% (2016-10) | 0% | 2% | |
| 10/90/0 equity / G-sec / gold | 139 | 2 | 7.1% | 8.1% | 8.8% | 6.4% (2016-10) | 0% | 2% | |
| GTAA (10-month trend) | 139 | 2 | 5.6% | 7.8% | 10.7% | 5.4% (2009-08) | 0% | 0% | |
| 0/20/80 equity / G-sec / gold | 139 | 2 | 4.5% | 7.7% | 16.9% | 4.0% (2009-05) | 14% | 25% | |
| 0/40/60 equity / G-sec / gold | 139 | 2 | 5.4% | 7.7% | 14.3% | 5.0% (2009-05) | 4% | 17% | |
| 0/0/100 equity / G-sec / gold | 139 | 2 | 3.5% | 7.7% | 19.3% | 2.9% (2009-05) | 18% | 29% | |
| 0/10/90 equity / G-sec / gold | 139 | 2 | 4.0% | 7.7% | 18.1% | 3.5% (2009-05) | 17% | 27% | |
| 0/30/70 equity / G-sec / gold | 139 | 2 | 4.9% | 7.6% | 15.6% | 4.5% (2009-05) | 7% | 22% | |
| 0/50/50 equity / G-sec / gold | 139 | 2 | 5.8% | 7.6% | 13.0% | 5.5% (2009-05) | 0% | 10% | |
| Minimum variance | 139 | 2 | 7.0% | 7.6% | 8.4% | 6.7% (2016-10) | 0% | 8% | |
| 0/70/30 equity / G-sec / gold | 139 | 2 | 6.4% | 7.6% | 10.4% | 6.2% (2008-10) | 0% | 6% | |
| 0/60/40 equity / G-sec / gold | 139 | 2 | 6.2% | 7.5% | 11.7% | 5.9% (2008-09) | 0% | 7% | |
| 0/80/20 equity / G-sec / gold | 139 | 2 | 6.7% | 7.5% | 9.0% | 6.5% (2008-10) | 0% | 7% | |
| Hierarchical risk parity | 139 | 2 | 6.5% | 7.5% | 8.4% | 6.1% (2016-10) | 0% | 7% | |
| 0/90/10 equity / G-sec / gold | 139 | 2 | 6.6% | 7.5% | 8.3% | 6.5% (2012-10) | 0% | 8% | |
| 0% Nifty 500 / 100% G-sec | 139 | 2 | 6.1% | 7.4% | 8.3% | 5.7% (2016-10) | 0% | 9% | |
| 0/100/0 equity / G-sec / gold | 139 | 2 | 6.1% | 7.4% | 8.3% | 5.7% (2016-10) | 0% | 9% | |
| Mean-variance (shrunk covariance) | 139 | 2 | 5.8% | 6.9% | 10.2% | 5.5% (2012-07) | 0% | 12% | |
| High beta (Nifty High Beta 50) | 139 | 2 | −6.3% | 4.6% | 14.4% | −16.6% (2010-04) | 50% | 55% |
The longest histories
A few portfolios can be run from before Apr 2005: the Nifty 500 and Nifty 50 and the equity, G-sec and gold mixes. More start months, but the extra windows still overlap, so the independent periods barely rise. The ten-year Nifty 500 SIP had a median of 13.8% across 261 windows, 3 of them independent.
After-tax XIRR, per cent a year, over each portfolio's longest usable history.
| Portfolio | Years | Windows | Independent periods | 5th pct | Median | Worst (start) | Below FD | More |
|---|---|---|---|---|---|---|---|---|
| Nifty 500 | 10 | 261 | 3 | 9.3% | 13.8% | 3.4% (2010-04) | 0% | |
| Nifty50 Equal Weight | 10 | 251 | 3 | 6.7% | 13.6% | −0.6% (2010-04) | 2% | |
| Nifty 50 | 10 | 208 | 2 | 8.8% | 12.6% | 3.4% (2010-04) | 0% | |
| 60/20/20 | 10 | 181 | 2 | 9.2% | 11.7% | 5.6% (2010-04) | 0% | |
| 60/40 | 10 | 181 | 2 | 9.1% | 11.1% | 5.9% (2010-04) | 0% | |
| Equal thirds | 10 | 181 | 2 | 8.2% | 10.2% | 6.5% (2010-04) | 0% | |
| Nifty 500 | 20 | 141 | 1 | 11.9% | 14.0% | 10.7% (2000-04) | 0% | |
| Nifty50 Equal Weight | 20 | 131 | 1 | 11.6% | 13.6% | 10.1% (2000-04) | 0% | |
| Nifty 50 | 20 | 88 | 1 | 10.7% | 12.7% | 10.0% (2006-10) | 0% | |
| 60/20/20 | 20 | 61 | 1 | 11.7% | 12.2% | 11.4% (2006-10) | 0% | |
| 60/40 | 20 | 61 | 1 | 10.2% | 11.3% | 9.8% (2006-10) | 0% | |
| Equal thirds | 20 | 61 | 1 | 10.6% | 11.0% | 10.4% (2003-11) | 0% |
The popular refinements
Each refinement is compared with the plain SIP on the same start months. Differences are paired, tested with a Newey-West standard error that allows for the overlap, and checked again on non-overlapping windows. 'Ahead' is the share of windows in which the refinement finished better.
Every refinement trailed the plain Nifty 500 SIP
Median difference in ten-year after-tax XIRR against a plain SIP on the same start months
How to read it. Bars to the left of zero lost to simply investing every month. The table below gives the share of windows each one won and the independent windows behind it.
Every refinement trailed the plain Nifty 500 SIP. Bar chart of 4 items. Median difference, XIRR points: highest Dynamic SIP, CAPE signal at −0.27 pts, lowest Dynamic SIP, yield-gap signal at −1.57 pts.
Valuation-aware SIPs and pausing in a crash
The dynamic SIP invests 1.5 times the instalment when valuations are in their cheapest third, half when they are in their dearest, and parks the rest in a liquid fund. Over ten years it trailed the plain Nifty 500 SIP by 0.27 points at the median. Stopping the SIP when the Nifty 500 was 20% below its peak, and keeping the money in a liquid fund, cost 1.48 points and about ₹1,80,865 of the final value on average.
Ten-year SIPs. Differences are the refinement minus the plain SIP, in XIRR points; value difference in rupees at the end.
| Refinement | Portfolio | Median difference | Ahead | Independent windows | Mean value difference | More |
|---|---|---|---|---|---|---|
| Dynamic SIP, CAPE signal | Nifty 50 | −0.29 pts | 37% | 2 | −₹42,857 | |
| Dynamic SIP, CAPE signal | Nifty 500 | −0.27 pts | 39% | 2 | −₹49,611 | |
| Dynamic SIP, yield-gap signal | Nifty 50 | −1.43 pts | 23% | 2 | −₹1,56,780 | |
| Dynamic SIP, yield-gap signal | Nifty 500 | −1.57 pts | 20% | 2 | −₹1,83,908 | |
| Pause in drawdowns, invest the backlog on restart | Nifty 50 | −0.44 pts | 0% | 2 | −₹96,394 | |
| Pause in drawdowns, invest the backlog on restart | Nifty 500 | −0.47 pts | 0% | 2 | −₹1,08,174 | |
| Pause in drawdowns, keep the cash | Nifty 50 | −1.32 pts | 1% | 2 | −₹1,48,565 | |
| Pause in drawdowns, keep the cash | Nifty 500 | −1.48 pts | 1% | 2 | −₹1,80,865 |
The SIP date barely matters
Nifty 500: median after-tax XIRR difference against an SIP on the 1st
How to read it. Each pair of bars is one date of the month, at five and ten years. Hundredths of a point either way.
The SIP date barely matters. Bar chart of 6 items. 5 years: highest Day 25 at +0.29 pts, lowest Day 5 at −0.03 pts.
The date of the month
Median difference in after-tax XIRR against an SIP on the 1st, in points. Nifty 500, and the median across the 12 headline portfolios.
| SIP date | Nifty 500, 3 years | 5 years | 10 years | All 12, 3 years | 5 years | 10 years | More |
|---|---|---|---|---|---|---|---|
| 5 | −0.02 pts | −0.03 pts | 0.00 pts | −0.02 pts | −0.02 pts | −0.01 pts | |
| 10 | +0.04 pts | +0.06 pts | +0.03 pts | 0.00 pts | −0.04 pts | −0.01 pts | |
| 15 | +0.10 pts | +0.09 pts | +0.05 pts | +0.02 pts | +0.04 pts | +0.01 pts | |
| 20 | +0.27 pts | +0.14 pts | +0.07 pts | +0.17 pts | +0.08 pts | +0.03 pts | |
| 25 | +0.48 pts | +0.29 pts | +0.14 pts | +0.27 pts | +0.19 pts | +0.08 pts | |
| 28 | +0.37 pts | +0.22 pts | +0.11 pts | +0.19 pts | +0.14 pts | +0.07 pts |
Step-ups
Ten-year SIPs with the instalment raised each year. The return barely moves; the corpus grows because more money goes in.
| Portfolio | Step-up | Median XIRR | Median final value | More |
|---|---|---|---|---|
| 60/40 | None (plain SIP) | 10.9% | ₹21,16,725 | |
| 60/40 | 5% a year | 10.9% | ₹25,53,421 | |
| 60/40 | 10% a year | 10.9% | ₹31,09,846 | |
| Nifty 500 | None (plain SIP) | 12.8% | ₹23,33,217 | |
| Nifty 500 | 5% a year | 12.8% | ₹27,99,188 | |
| Nifty 500 | 10% a year | 12.8% | ₹33,95,562 |
A lump sum moved in over months (STP)
Rs 12 lakh moved from a liquid fund in equal monthly transfers, against investing it all on day one. Difference in final value, per cent.
| Portfolio | Transfer over | Years held | STP minus lump sum, median | Lump sum ahead | More |
|---|---|---|---|---|---|
| Nifty Midcap 150 | 12 months | 1 | −4.5% | 61% | |
| Nifty Midcap 150 | 12 months | 3 | −4.9% | 65% | |
| Nifty Midcap 150 | 12 months | 5 | −5.7% | 63% | |
| Nifty Midcap 150 | 12 months | 10 | −6.8% | 68% | |
| Nifty 50 | 12 months | 1 | −3.6% | 68% | |
| Nifty 50 | 12 months | 3 | −3.8% | 69% | |
| Nifty 50 | 12 months | 5 | −4.4% | 69% | |
| Nifty 50 | 12 months | 10 | −4.5% | 71% | |
| Nifty 500 | 12 months | 1 | −2.8% | 63% | |
| Nifty 500 | 12 months | 3 | −3.6% | 66% | |
| Nifty 500 | 12 months | 5 | −4.0% | 67% | |
| Nifty 500 | 12 months | 10 | −4.8% | 70% | |
| Nifty Smallcap 250 | 12 months | 1 | −3.2% | 59% | |
| Nifty Smallcap 250 | 12 months | 3 | −3.7% | 59% | |
| Nifty Smallcap 250 | 12 months | 5 | −4.1% | 59% | |
| Nifty Smallcap 250 | 12 months | 10 | −5.6% | 65% | |
| Nifty Midcap 150 | 6 months | 1 | −2.5% | 61% | |
| Nifty Midcap 150 | 6 months | 3 | −3.1% | 61% | |
| Nifty Midcap 150 | 6 months | 5 | −3.3% | 62% | |
| Nifty Midcap 150 | 6 months | 10 | −3.4% | 62% | |
| Nifty 50 | 6 months | 1 | −2.0% | 63% | |
| Nifty 50 | 6 months | 3 | −2.3% | 62% | |
| Nifty 50 | 6 months | 5 | −2.5% | 62% | |
| Nifty 50 | 6 months | 10 | −2.5% | 58% | |
| Nifty 500 | 6 months | 1 | −2.0% | 62% | |
| Nifty 500 | 6 months | 3 | −2.2% | 63% | |
| Nifty 500 | 6 months | 5 | −2.5% | 64% | |
| Nifty 500 | 6 months | 10 | −2.4% | 60% | |
| Nifty Smallcap 250 | 6 months | 1 | −2.0% | 58% | |
| Nifty Smallcap 250 | 6 months | 3 | −2.0% | 60% | |
| Nifty Smallcap 250 | 6 months | 5 | −2.1% | 60% | |
| Nifty Smallcap 250 | 6 months | 10 | −2.1% | 60% |
How often a ten-year SIP beat a lump sum
Share of start months in which the SIP's XIRR beat the lump sum's CAGR
How to read it. Below half, putting the whole amount in at the start usually did better. These are two different investors, so the STP table above is the fair comparison.
How often a ten-year SIP beat a lump sum. Bar chart of 12 items. Share of start months the SIP finished ahead: highest Nifty Midcap 150 at 67%, lowest GTAA (10-month trend) at 45%.
SIP against a lump sum
Ten years. The SIP's XIRR against the CAGR of investing the same total at the start. These are two different investors, so this is the comparison people make rather than a fair one; the STP table above is the fair one.
| Portfolio | SIP, median | Lump sum, median | SIP ahead | More |
|---|---|---|---|---|
| Nifty Midcap 150 | 16.9% | 15.8% | 67% | |
| Nifty Smallcap 250 | 14.5% | 13.3% | 66% | |
| Nifty 500 | 12.8% | 12.5% | 62% | |
| Nifty 50 | 12.0% | 11.4% | 61% | |
| Equal thirds | 9.8% | 10.1% | 54% | |
| Permanent portfolio | 8.6% | 8.8% | 52% | |
| 60/20/20 | 11.1% | 11.5% | 52% | |
| All Weather (Indian approximation) | 9.0% | 9.3% | 49% | |
| Four-factor blend | 15.3% | 14.9% | 47% | |
| 60/40 | 10.9% | 11.0% | 47% | |
| Dual momentum | 8.9% | 9.0% | 46% | |
| GTAA (10-month trend) | 7.8% | 7.9% | 45% |
What tax took
Ten-year SIPs. Median XIRR points lost to tax, and what switching the equity exemption off changes.
| Portfolio | Lost to tax | Exemption off vs on |
|---|---|---|
| Nifty Smallcap 250 | 0.07 | −0.03 pts |
| Nifty Midcap 150 | 0.21 | −0.06 pts |
| Nifty 500 | 0.31 | −0.07 pts |
| Four-factor blend | 0.32 | −0.08 pts |
| Nifty 50 | 0.40 | −0.08 pts |
| Permanent portfolio | 0.45 | −0.10 pts |
| All Weather (Indian approximation) | 0.47 | −0.10 pts |
| 60/40 | 0.49 | −0.09 pts |
| 60/20/20 | 0.54 | −0.09 pts |
| Equal thirds | 0.61 | −0.10 pts |
| GTAA (10-month trend) | 1.15 | 0.00 pts |
| Dual momentum | 1.38 | 0.00 pts |
Checking published studies
Before trusting my engine, I ran it in replication mode: the Nifty 50 total-return index, no costs, no tax, instalments on the first session of each month. Published Indian SIP studies can then be set against it.
Nifty 50 TRI, no costs or tax. XIRR across start months, per cent a year.
| Years | Windows | Lowest | Median | Highest | Above 12% | Negative | More |
|---|---|---|---|---|---|---|---|
| 5 | 268 | −4.4% | 14.2% | 47.9% | 65% | 0.7% | |
| 8 | 232 | 2.2% | 13.2% | 34.6% | 67% | 0.0% | |
| 10 | 208 | 3.8% | 13.7% | 23.5% | 73% | 0.0% | |
| 12 | 184 | 5.5% | 13.3% | 18.2% | 70% | 0.0% | |
| 15 | 148 | 6.9% | 13.4% | 17.6% | 85% | 0.0% | |
| 20 | 88 | 10.9% | 13.6% | 15.7% | 90% | 0.0% |
Freefincal's study used the same index, and I reproduce the fall it describes in ten-year SIP returns across the end dates it covers. WhiteOak Capital's and Geojit's studies used the Sensex, which I do not hold, over a sample that starts earlier than mine. My figures have the same shape and come out a little lower, as a later start and a later end would predict. Because the index differs, neither counts as a match under the specification's rule, and nothing was tuned to close the gap. Their published figures are quoted in the results log.
What I predicted, and what happened
Five predictions were written down before any SIP result was computed. 3 held; the others are reported with the same weight.
S1 Failed
The SIP date of the month makes no practical difference: the median after-tax XIRR differs by less than 0.2 points between any two dates, and no difference is significant once overlap is accounted for.
Across the 12 headline portfolios, later dates differed from the 1st by up to 0.75 points at the median at three and five years, with 60 paired tests beyond two Newey-West standard errors; at ten years and beyond the largest gap was 0.21 points. The results log traces this to a turn-of-month effect in Indian indices. Six dates on twelve portfolios and six horizons is a lot of tests, and a well-known effect may weaken.
S2 Held
An all-equity (Nifty 500) SIP had a lower after-tax XIRR than the fixed-deposit SIP in some 5-year windows but in no 15-year window.
It fell below the FD SIP in 10.6% of 5-year windows (4 independent periods) and in 0.0% of 15-year windows (1 independent period, so the second half rests on very little).
S3 Failed, in the other direction
The valuation-aware dynamic SIP beats the plain SIP by less than 0.5 points of XIRR at the median 10-year window, with the interval for the difference including zero.
It did not beat the plain SIP at all: the median 10-year difference was −0.29 pts for Nifty 50 and −0.27 pts for Nifty 500, with the CAPE signal. Keeping money in the liquid sleeve while valuations looked dear cost more than buying more when they looked cheap gained.
S4 Held
Pausing during drawdowns lowers the terminal value relative to an uninterrupted SIP in most windows that contain a drawdown of more than 20%.
At five years and longer, the paused SIP finished ahead in at most 4% of windows, for either way of handling the skipped money. That share is over all windows, which is stricter than the prediction's subset.
S5 Held
Lump sum beats a 12-month STP in about two thirds of start months for an all-equity portfolio, in line with Vanguard's US finding (2012).
Investing on day one beat a 12-month STP in 59% to 71% of start months, across the equity portfolios and horizons tested.
The workings
Data
Every chart has a table view, a CSV download and a link to its published JSON. Bundles are listed in data status.