Best Funds for a Long Term SIP: The 10-Year Evidence

There is a specific investor this article is for: someone who has decided to run a SIP not for two or three years but for ten, fifteen, twenty - through job changes, market crashes, elections and at least one full economic cycle. A long term SIP is a different commitment from an ordinary one, and it deserves a different standard of evidence.

Most "best funds for long term SIP" articles are the same five funds with last year's returns pasted in. This one is built differently: we simulate SIPs through the complete NAV history of every Indian mutual fund - 8,000+ Direct-Growth schemes, refreshed daily - and let the decade-long numbers speak. Descriptive research, not advice; every figure pre-tax.

What ten years of monthly investing actually produced

The cleanest way to judge a long term SIP fund is to run one. Rs 10,000 a month, real NAVs, real dates, full decade - here is what that produced in funds currently in the top tier of our quality ranking:

Two things in that table deserve more attention than the big numbers.

First, every one of these funds fell 34-39% from a peak at some point during the decade. The investors who own those final values are exclusively the ones who kept the instalment running through those falls. The worst-fall column is not a footnote to the returns column; it is the price of it.

Second, the SIP XIRR is higher than the same funds' lumpsum CAGR (Invesco Midcap: 22.1% SIP vs 20.1% lumpsum over the same period). That is rupee-cost averaging doing its job across a decade that included several deep corrections - the SIP kept buying cheap units in months when a lumpsum investor was simply underwater.

Time does the heavy lifting: the arithmetic of long horizons

Why obsess over the "long" in long term SIP? Because SIP compounding is violently back-loaded. At the equity-category median of about 14.7% a year (the median across all equity funds with a 10-year record in our data), a Rs 10,000 monthly SIP builds approximately:

(Illustrative compounding at a constant historical median; real paths are lumpy and nothing about the median is guaranteed.) The lesson is in the last column: at year 5, two-thirds of your value is just your own deposits. By year 20, four-fifths of it is growth. The first five years of a long term SIP feel disappointing by design - that is when investors quit, and it is statistically the worst possible moment to do so.

Add the one habit that outguns fund selection entirely: the annual step-up. Raising the instalment 10% a year - roughly tracking salary growth - takes the 20-year outcome from about Rs 1.3 crore to well over Rs 2.5 crore at the same return. No amount of fund-picking skill is worth as much as that single standing instruction.

What to demand from a fund you will hold for 15 years

Consistency first, and be strict about it

Over 15 years you will live through every kind of market. The statistic that best predicts whether a fund rewards that patience is the share of its rolling 12-month windows that ended positive. Category medians today: mid cap 92%, large and mid cap 89%, ELSS 88%, flexi cap 86%, small cap 85%. The funds in our top table run 95-100%. Under 80%, a fund is repeatedly handing its SIP investors year-long stretches of nothing - fuel for abandonment.

A drawdown profile you can pre-commit to

The table already said it: -34% to -39% is what the best decade-long SIP funds did to their holders on the way to tripling their money. On a mature Rs 40 lakh corpus, a 35% fall is Rs 14 lakh on paper. Decide now, in writing if it helps, what you will do that month: the historically correct answer has been "keep the SIP, add if possible", because those months bought the cheapest units of the entire journey. If honesty says you would stop, choose a gentler category - aggressive hybrids (median worst fall -29%, consistency 92%) exist exactly for that temperament, and their ranking updates daily.

Quality that is measured within category

Raw returns reward whoever took the most risk in a bull market. Our King Score ranks every fund inside its own category on 3-year return percentile, consistency, drawdown and Sharpe - so a leader means "best at its own game", not "most leveraged to the last rally". All six funds in the opening table currently score 83-91. Method, fully public: what King Score measures.

Cost, multiplied by fifteen years

At 15-20 year horizons the expense ratio compounds into lakhs. The Direct-vs-Regular gap alone - roughly 0.5-1% a year - is worth several years of instalments over a long SIP. Direct-Growth only; it is the only share class this site ranks. Full arithmetic: Direct vs Regular.

Category strategy for a 10-20 year SIP

The long horizon changes the category calculus in one specific way: you can afford deeper drawdowns, because your instalments exploit them. Category medians from the live data:

• Mid cap (10Y median 17.3%, consistency 92%) - the standout long-SIP category in our data: near-small-cap returns with the best consistency in equity. Current field.
• Small cap (17.8%, consistency 85%) - the highest ceiling and the roughest ride; median worst fall -32% with long winters. Genuine 15-year money handles it; 7-year money often cannot. Ranking.
• Large & mid cap (15.2%, consistency 89%) - the balanced core: mandated exposure to both halves of the market. Three of the six funds in our 10-year table live here.
• Flexi cap (13.9%, consistency 86%) - one fund, whole-market mandate, manager decides the mix. Ranking.
• ELSS (13.9%, consistency 88%) - equity returns plus the 80C deduction, with a 3-year lock on each instalment that is irrelevant at a 15-year horizon - arguably the most painless place for tax-linked money. Ranking.

A robust long-term structure remains stubbornly boring: one core fund (large & mid, flexi, or index) plus one satellite (mid or small cap), step-up enabled, reviewed once a year. Two SIPs. The data on overlap is unambiguous - beyond a handful of funds, new SIPs mostly rebuy the same stocks.

Maintenance: the once-a-year, twenty-minute review

A long term SIP is not fire-and-forget; it is fire-and-check-annually. The checklist:

1. Consistency trend - has the rolling-positive share decayed for two consecutive years?
2. King Score - still in the top half of its category? A one-year slip is noise; a three-year slide is information.
3. Category drift - did the fund's mandate or manager change in a way that changes what you own?
4. Your own horizon - within 3-4 years of the goal, begin moving accumulated corpus toward hybrids or debt in tranches; the SIP got you here, sequence risk takes over from here. Our SWP guide covers the spending side.

What is deliberately absent from that list: last year's return, this month's star rating, and anything a fund advertisement says.

The step-up, in full detail - because it outguns everything else

The single most valuable paragraph in this article deserves its own section. A step-up SIP raises the instalment by a fixed percentage every year. The arithmetic at the historical equity median (14.7%, illustrative and unguaranteed):

Read the second and third rows against each other. The step-up investor ends ahead of someone who started at double the instalment - while their first-year outlay was half as heavy. The step-up works because it back-loads contributions into the years when your income can afford them, without requiring the willpower to start big. Every major platform supports an automatic annual step-up instruction; setting it once at SIP creation removes the decision forever. If you take one action after reading this article, that is the one.

Two refinements from the data. First, step up on a fixed calendar date (e.g. every April), not "when markets look good" - discretion is where step-ups die. Second, when income jumps unusually (promotion, job change), reset the base rather than waiting for the percentage to catch up: the instalment as a share of income, kept roughly constant, is the real discipline.

When to add a second fund - and when not to

The two-fund structure (core + satellite) covers most long-term SIP investors. The honest triggers for a third:

• The corpus has outgrown single-AMC comfort. Past Rs 50 lakh-1 crore in one fund house, adding a second AMC's fund in the same role diversifies operational and manager risk - a governance hedge more than a return decision.
• A genuinely different exposure is wanted. International equity or a precious-metals sleeve adds something the first two funds do not own. A third domestic mid cap fund does not - our overlap data shows same-category funds sharing 30-50% of holdings routinely.
• A goal has split. Retirement money and a child's education fund at different horizons legitimately justify separate structures - though the separation is for accounting clarity, not because the funds need to differ.

And the non-triggers, which account for most third funds in the portfolios our Doctor analyses: a friend's tip, a new fund offer, last year's category winner, or the platform's "investors also bought" row. The test is one sentence: what does this fund own that my current funds do not? If the answer is "the same stocks with a different label", the instalment belongs in the existing funds instead. Beyond 8 funds, our diagnostic flags the portfolio automatically - because at that point the collection has quietly become an expensive index fund.

The 15-year seasons: a field guide

Long-term SIP investors in our data live through recognisable seasons, worth naming in advance:

Years 1-3, the invisible years. The corpus is small; even great returns produce boring rupee numbers. A 20% year on Rs 3 lakh is Rs 60,000 - dinner-table irrelevant. The only job is habit formation; performance review is actively counterproductive here.

Years 4-7, the first real test. The corpus reaches meaningful size just in time for its first major drawdown - a six-figure paper loss and a market narrative explaining why equity is finished. This is where the consistency filter pays: high-consistency funds' recoveries kept believers; laggards' slow recoveries manufactured quitters.

Years 8-12, the compounding years. Growth begins visibly outpacing contributions - the corpus moves more in a good month than you deposit in a quarter. Temptations invert: now the danger is redeeming "just some profits" for consumption, which amputates precisely the capital that was about to do the heaviest lifting.

Years 13+, the harvest approach. The corpus dwarfs the instalment; sequence risk gradually replaces return as the dominant variable. The de-risking tranches described above begin. The SIP that survives to this season has already won - the remaining job is not losing.

FAQ

Which fund is best for a 20-year SIP?
Nobody can name the specific winner of the next 20 years - a decade ago, few would have named the funds now topping our table. What the data supports: a top-quartile consistency fund in a high-median category (mid cap, large & mid cap), Direct plan, step-up on, held through drawdowns. The SIP ranking applies these filters daily.

Is Rs 5,000 a month enough for the long term?
At the historical equity median (14.7%), Rs 5,000 monthly is roughly Rs 66 lakh in 20 years from Rs 12 lakh invested - illustrative, pre-tax, median-based. With a 10% annual step-up it crosses Rs 1.2 crore. Enough is a function of the goal, but the instrument is not the constraint.

Should a long term SIP go into small cap funds?
The 17.8% median says yes; the -32% median worst fall and 85% consistency say only if the money and the temperament are genuinely long-term. The compromise the data likes: mid cap as the satellite - most of the return, meaningfully better consistency.

SIP for 10 years vs SIP for 20 - does the fund choice change?
Mostly no; the review discipline changes. What changes at 20 years is the exit plan - the final five years of a 20-year SIP are about protecting a large corpus, which is a different job from growing a small one.

Every fund named here has a complete data page in the terminal: 20 years of NAV history in candles, every rolling window, drawdowns in rupees, and a SIP simulator at any amount and horizon. Free account, and every table in this article regenerates tomorrow morning with fresh data.

Should I stop the SIP once I reach my target corpus early?
Reaching the number early is the good problem, and the answer is structural, not emotional: if the goal date is still years away, the money's horizon has not changed - shift new instalments to a calmer category if you wish, but the accumulated corpus stays invested until the glide path would naturally begin. Money stopped early spends its final years in savings accounts losing quietly to inflation.

Does it make sense to run long-term SIPs in an index fund only?
Entirely defensible: the Nifty 50 index funds' ~12% decade at 0.06-0.26% cost with zero selection risk is a legitimate core, and pairing an index core with one active mid cap satellite captures most of what the active universe offers. What the pure-index route gives up, our data suggests, is the top-quartile active funds' meaningful outperformance - the price of never having to pick one.

What if my fund house merges or my fund changes its name?
Happens regularly across a 15-year horizon (several funds in our own tables carry renamed histories). A name change alone changes nothing. A mandate change - category reassignment, strategy overhaul, new manager with a new style - is a real review trigger: re-run the four-number check as if it were a fresh fund, because effectively it is.

Can I run a long term SIP for my child's education?
It is one of the best-matched uses: an education goal 12-15 years out has exactly the horizon that lets equity SIPs work through full cycles. Two refinements for goal-linked SIPs: run the fund in the parent's name or as guardian per your tax planning, and start the glide path earlier than retirement money would - education dates cannot slip, so the corpus should be substantially de-risked by two to three years before the admission year, not five.