30 Years of Wealth Creation

The distilled wisdom of Motilal Oswal’s Annual Wealth Creation Study, 1996–2025

ELI5 / TLDR

For thirty years Raamdeo Agrawal has run the same experiment: take every Indian company that made shareholders rich over the trailing five years, line them up, and ask what they had in common. The answer barely changed. Wealth comes from a high and durable return on capital, compounded by years of profitable growth, in a business protected by a moat, run by honest and competent people — and bought at a price that doesn’t already assume all of this. Everything else — the macro lens, the moat theory, the digital era, economic profit — is the same insight viewed from a different window. The study’s real lesson is in the failures it documents: growth that eats capital (“gruesome” businesses), great companies bought at terrible prices, and the steady drumbeat of frauds. Get the four things right and time does the rest; “time is a friend of good companies and an enemy of bad companies.”

The Arc

The studies read, in sequence, like a single mind teaching itself to invest in public — and occasionally getting humbled by the market in real time.

The foundational nineties: ROE is destiny (1996–2000)

The first study, in 1996, is almost touchingly simple. It is a few pages of Raamdeo Agrawal’s “Inquire” research stapled together, studying the 100 companies that had quadrupled their market cap (32%+ CAGR) over 1991–96. The conclusion that would anchor the next three decades was already there in the first bullet: “Wealth Creating companies have a substantially high ROE and ROCE.” And the second: a high correlation between ROE and the P/E the market awards. The 1996 study found that wealth creators averaged 23% ROE, carried little debt, threw off cash rather than guzzling it, and — counterintuitively — only needed moderate sales growth (median 23%) to deliver 32% to shareholders, because they leveraged fixed costs and widened margins rather than chasing revenue.

The 1996 valuation insight is the seed of everything Motilal Oswal would later formalise: the right P/E to pay is roughly a company’s sustainable ROE, and buying below that is your margin of safety. In 1991, 63 of the Inquire 100 traded at a P/E well below their ROE; by 1996 the gap had closed and the re-rating was the return.

The 1997 study was an accidental masterstroke of timing. The five years it covered (1992–97) had a falling market — the Sensex compounded at −4.7%, and only 45 companies cleared the bar instead of 300. Stress-tested, the philosophy tightened into the line that became the firm’s creed: a good business is one whose economics are “not only distinctly superior but get better with time.” It introduced DuPont decomposition (ROE = margin × turnover × leverage) and the idea that wealth creators are leaders — 30 of the 45 ranked #1 or #2 in their industry. It also, for the first time, named the wealth destroyers — TISCO, ACC, Century — companies that vaporised more capital than the creators built.

By 1998 (Competitive Strengths) the firm had a clean algebraic skeleton: P/B = ROE × P/E, which means the whole game is earning a high return on equity and having the market pay up for it. IT made its first appearance — Satyam and Wipro were the fastest creators — and the study tipped Infosys, HDFC, Hero Honda and Cipla as future winners. That call aged like wine.

Then came the part where the study learned humility. The 1999 study (How to Value Growth) introduced the PEG ratio and observed, correctly, that “no P/E multiple can be regarded as HIGH or LOW unless it is measured relative to a stock’s underlying earnings growth potential.” Infosys, bought at a PEG of 0.52, compounded at 102%; HUL at a PEG of 1.46 managed 24%. But the 2000 study (Characteristics of Multi-Baggers) got caught in the dot-com mania it was describing. Its honest, damning finding: 86% of the top-10 multi-baggers’ returns came from P/E re-rating, only 14% from earnings. New-economy P/E had gone from 20x to 122x. The study half-acknowledged a “bubble factor” and then the bubble burst on schedule.

Picking up the pieces: value, cycles, and the rate regime (2001–2006)

The studies of the early 2000s are the work of an analyst chastened by the crash and rebuilding from first principles. The 2001 study (Components of Value) is the most theoretically dense of the era, breaking intrinsic value into asset value, earning-power value, and growth value — and warning that “solidity of the barrier or competitive advantage is the starting point of Wealth Creation. If it is missing, any amount of growth will be meaningless in the long run.” It also launched the Five Forces of Wealth Creation (high ROE; ROE plus growing capital; free-cash growth; cost of capital; margin of safety via payback). New-economy P/E had collapsed from 122x back to 26x — the study had front-row seats to mean reversion.

The 2002 study (Role of Interest Rates) made the single best macro call in the whole run. Reasoning that “the tiniest change in interest rates changes the value of every financial asset,” and that India in 2002 looked like the US in 1981, it predicted that falling long-bond yields would re-rate Indian equities. The 2003–07 bull market obliged. The 2003 study (Transitory vs Enduring) added the crucial moral distinction — bad businesses can throw up a transitory multi-bagger that round-trips to zero, but enduring wealth needs good business + good management + a cheap price, all three. The earnings-yield-to-bond-yield ratio hit 1.53, the cheapest equities had ever been.

The mid-decade studies tracked the commodity supercycle they were living through. The 2004 study (Business Cycles in Commodity Stocks) — written the year a commodity company topped the list for the first time — mapped the commodity cycle into phases (Gloom → Recovery → Squeeze → Euphoria → Glut) and quoted the old saw that “in the short run the market is a voting machine, but in the long run it is a weighing machine.” The 2005 study (Price & Value) introduced the third ranking category, Most Consistent, and delivered the line everyone knows: “price is what you pay, value is what you get.” It also showed that the TMT bust had concentrated 85% of all wealth destruction in tech-media-telecom. The 2006 study (Terms of Trade) made pricing power its lens and noticed that the fastest wealth came from large, unpopular companies bought cheap — BHEL, SAIL, ONGC at under book value. Its closing warning — that the market was heading “into the overvalued zone” — preceded the 2008 crash by two years.

The mature era: from macro thesis to moat theory (2007–2014)

This is the stretch where the study graduated from describing patterns to building durable theory.

The 2007 study (Next Trillion Dollar Opportunity) was the big macro swing. India had just crossed $1tn of GDP after 30 years; the NTD thesis was that the next trillion would arrive in five, dragging consumption past an inflection point. It tipped Financials, Telecom, Cars, Engineering and Cement. The macro arc was right (though it took until ~2014, not 2012, to hit $2tn), and the “favour the private sector” steer was excellent — but the real-estate bet (“the future will see this sector growing bigger and faster than many others”) was about to become one of the worst calls in the series. Crucially, even amid the euphoria, the study flagged that “margin of safety is low” — a warning issued at the very top.

Then 2008 (Great, Good, Gruesome), written in the teeth of the crash, lifted Buffett’s savings-account taxonomy and gave the series its most useful mental model. A Great business is a savings account whose interest rate (ROE) is high and rising while it consumes almost no capital; a Good business grows but has to “put up more to earn more”; a Gruesome business is the airline — “a bottomless pit, attracted by growth when they should have been repelled by it.” The same year, the leadership-rotation call (commodities about to lose, users of commodities to win) proved exactly right.

From here the framework deepened almost yearly. The 2012 study (Economic Moat) formalised the moat as the product of industry structure (Porter’s five forces) and company strategy, and explained why moated stocks keep beating the market despite premium valuations — the market continuously re-prices the roll-over of their Competitive Advantage Period. (That year both Reliance and ONGC dropped out of the top 100, ending eight straight years of oil-and-gas dominance, and ITC became the biggest creator for the first time.) The 2013 study (Uncommon Profits) sharpened it into the idea of Emergence and Endurance: a company creates “uncommon profit” the moment ROE crosses the cost of equity, and the rare ones sustain it — of 568 companies earning ROE above 15% in 2004, only 86 still did a decade later.

The capstone of the era was 2014 (100x: The Power of Growth), built on Thomas Phelps’ 100 to 1 in the Stock Market. Its arithmetic is the cleanest statement of how a fortune is made: a 100-bagger is roughly 25x earnings growth times 4x P/E re-rating. The 47 enduring Indian 100-baggers had averaged 332x over 15 years — a 47% CAGR — and the hunting framework was SQGLP: small Size, Quality (of business and management), Growth, Longevity, reasonable Price. Phelps’ line is the soul of the whole project: “to make money in stocks you must have the vision to see them, the courage to buy them and the patience to hold them. Patience is the rarest of the three.”

The modern era: assembling and stress-testing QGLP (2015–2025)

The last decade reads as the firm bolting the final planks onto QGLP and then pressure-testing it. 2015 (Mid-to-Mega) added leadership as the engine that carries a mid-cap into the top 100. 2016 (Focused Investing) addressed the question every prior study had ducked — not what to buy but how much — landing on 15–20 high-conviction positions sized by a Kelly-flavoured “confidence-adjusted payoff.” 2017 (CAP & GAP) made longevity measurable: CAP is the number of years ROE stays above 15%, GAP the years earnings outgrow the market, and the two are inversely correlated — fast growth burns out, durable growth is rarely fast. 2018 (Valuation Insights) ran twenty years of data and crowned the winners among pricing rules: PEG below 1x earned +19% alpha, a payback ratio below 1x +17% — while overpaying reliably destroyed returns. 2019 (Management Integrity) turned forensic, cataloguing “sharp practices” under the deadpan rule “Credit P&L, Debit Balance Sheet” and the autopsies of Satyam, Manpasand, Educomp and Gitanjali.

The 2020 Silver Jubilee (The QGLP Checklist) is the master edition — 25 questions, each paired with one of the 25 frameworks the firm had built, mapping a quarter-century of study onto a single tool. Its sobering headline: of the top 500 companies of 1995, only 100 beat the Sensex’s 9.2% CAGR over the next 25 years, and nearly 60 of the top 100 of 1995 failed to. “Time is a friend of good companies and an enemy of bad companies.”

The five studies since have mostly chased themes — Atoms to Bits (2021, digital value migration), Consistents & Volatiles (2022, where only 16% of companies were “consistent” but 87% of those beat the market), Hockey-Stick Returns (2023, recasting everything in terms of economic profit), Bruised Blue Chips (2024), and the grand macro reprise in 2025 (Multi-Trillion Dollar), which extends the 2007 NTD thesis to a $16tn GDP by 2042. The wealth-creation numbers themselves went vertical — a record ₹148 trillion created in 2020–25 at a 38% CAGR — but the philosophy underneath had stopped changing around 2014. The modern studies are variations on a settled theme.

The Enduring Laws

Strip away the annual costumes and the same handful of laws survive all thirty studies.

1. Return on capital is the master variable

It is the first sentence of the first study and the spine of the last. High, sustained ROE/ROCE is what separates creators from destroyers, year after year. The 2013 study put a floor on it (ROE above the ~15% cost of equity = “uncommon profit”); the 2023 study restated it as economic profit — ROE minus a 10% capital charge — and showed why Nestlé, earning 96% on a small equity base, is worth more than Indian Oil earning 7% on a base 56 times larger despite four times the accounting profit. Same truth, sharper knife.

2. Growth only counts when ROE clears the cost of capital — and longevity is everything

Growth is necessary but conditional. The 2001 study’s warning that growth without a moat is “meaningless” matured into the 2017 distinction between CAP and GAP. The recurring, almost depressing finding is how rare durability is: of 116 companies studied for a 20-year competitive-advantage period, only 16 sustained it — and all 16 sat in the same quadrant (favourable industry and favourable strategy). Munger’s line, quoted in 2013, is the whole bet: “if a business earns 18% on capital over 20 or 30 years, even if you pay an expensive looking price, you’ll end up with one hell of a result.”

3. The “gruesome” trap — growth that destroys capital

The most valuable thing the studies teach is what to avoid. The 2008 savings-account taxonomy named it: businesses that grow furiously while earning below their cost of capital are wealth incinerators, and high growth is the bait. The data backs the warning — wealth destruction scaled with euphoria, from 0.2% of wealth created in the 2003–08 bull to 33% in the 2007–12 window, concentrated in a few capital-hungry names (Suzlon, RCom, Tata Steel’s Corus, the telecom price wars).

4. Management is 90%, and integrity is the gate

Fisher’s “management is 90%, industry 9%, all other factors 1%” recurs across editions and gets its own full study in 2019. Integrity is non-negotiable because a dishonest management is a “race to zero” — and the forensic tells (OCF that won’t convert to PAT, auditor resignations, pledged promoter shares, CEO/CFO churn) recur as the cheapest insurance an investor can buy. “It is not only unnecessary but downright stupid to buy into a company run by men of doubtful integrity.”

5. Price discipline — pay below what the quality is worth

From the 1996 “buy at a P/E below sustainable ROE” to the 1999 PEG to the proprietary Payback Ratio (market cap ÷ next five years’ cumulative profit) that appears from 2007 onward, this is one idea evolving. The 2018 study’s verdict is the most quotable: among all heuristics, PEG below 1x and payback below 1x were the most reliable formulas for outperformance across two decades. “Buy a growth stock, but don’t pay for growth.”

6. Consumer-facing, focused, leading businesses win disproportionately

Across thirty years the same texture keeps appearing: roughly two-thirds of all wealth created comes from consumer-facing businesses (63 of 100 in the 25-year study), market leaders dominate, and focused companies beat diversified ones (87% of the very first cohort were focused). Consumer franchises have proven more durable than production-cost advantages because a brand in a customer’s head is harder to attack than a cost edge in a factory.

QGLP, Assembled

The firm’s house philosophy — Quality, Growth, Longevity, at a reasonable Price — was not designed up front. It was excavated, one plank at a time, and the 2020 checklist edition openly credits the sources.

• Quality of business is the oldest plank, present from 1996 (high ROE) and theorised through the moat work of 2012 and the industry-structure / strategy lens borrowed from Porter.
• Quality of management got its rigour from the 2008 Great-Good-Gruesome typology (which is really a statement about how management deploys capital) and its forensic teeth from the 2019 integrity study.
• Growth was formally added to Buffett’s own four-part framework — the studies are explicit that earnings growth is Motilal Oswal’s distinctive contribution — and was sharpened by the 1999 PEG work and the 2014 100x growth decomposition.
• Longevity is almost entirely the gift of the 2017 CAP & GAP study (folded into the checklist as the framework under the “L”), with roots in the 2005 multi-bagger CAP concept and the 2013 endurance study.
• Price descends from the 1996 ROE-vs-P/E rule, through the 1999 PEG, to the 2018 valuation study that empirically ranked which price heuristics actually work.

QGLP, in other words, is not a slogan — it is a 25-year literature review wearing a four-letter acronym. And the firm’s own data on it is humbling: it is easy to state and brutally hard to satisfy, because longevity (the L) is the rarest ingredient and the one no checklist can guarantee.

Calls That Aged Well (and Badly)

A study that makes predictions every year for thirty years invites a scorecard. Here is an honest one.

Aged well:

• The 2002 interest-rate call. “India 2002 = US 1981.” Falling yields would re-rate equities. The 2003–07 bull market was the proof.
• The 1998 future-creator list. Infosys, HDFC, Hero Honda, Cipla — named in 1998, all went on to be major compounders.
• The 2008 leadership rotation. Commodities to lose, users of commodities to win, value to migrate back to the private sector. Almost perfectly right; commodities were dead money 2008–13.
• The 2006 overvaluation warning that the market was entering the “overvalued zone” — issued two years before the crash.
• Several of the 2014 100x candidates. Of the seven names floated (with the caveat that they were screens, not recommendations), Suven, Granules, Tata Elxsi and Aarti went on to deliver enormous multi-baggers. (DCB Bank and Atul Auto did not — the screen was right about the type, not every name.)

Aged badly (or mixed):

• The 2000 multi-bagger study got swept up in the dot-com bubble it was anatomising — and admitted, after the fact, that 86% of those returns were valuation, not earnings.
• The 2007 real-estate bet — “growing bigger and faster than many others” — landed right before real estate and infra became among the worst destroyers of the next five years. (The same study’s broader NTD macro and pro-private-sector tilt aged well, so it nets out to a split decision.)
• The 2018 bearish call that markets would “remain soft” because valuations baked in 16% earnings growth against 3% delivered was directionally sensible but early — Indian large-caps ran hard into 2020–21 before the logic asserted itself.

Too soon to grade — the 2024 Bruised Blue Chips. This is the call worth tracking. In November 2024 the study named a watchlist of high-quality stocks beaten down 30%+ from their highs: Adani Total Gas (−80%), Adani Green (−64%), Adani Enterprises (−42%), Gujarat Gas, SBI Cards, Tata Elxsi, DMart, IRCTC, Berger Paints, Asian Paints, and Indian Oil. The thesis — that bruised blue chips have near-zero mortality (all 99 historical cases still exist) and recover at a 55% CAGR from their lows if bought near a P/B of ~2x and after a healing trigger — is elegant and well-evidenced historically. But the explicit caveat is the catch: “bruising alone cannot be the sole reason for buying… else the Bruised Blue Chip will end up as a value trap.” Whether Asian Paints (structurally bruised by new competition) and the Adani names (sentiment-bruised) heal or stay broken is the open question of this edition.

The Companies That Kept Winning

The most striking thing about reading thirty lists back to back is how few names recur — and how telling those few are.

A small perennial club shows up across decades: HDFC and HDFC Bank, Asian Paints, Hero Honda, Cipla, Infosys, Sun Pharma, Titan, Pidilite, Nestlé, ITC, and Kotak Mahindra Bank. When the 2020 study ranked the best all-round creator of the full 25 years — combining speed, size and consistency — the winner was Kotak Mahindra Bank, which outperformed in 21 of 23 rolling three-year windows. Asian Paints appeared in all ten of the studies that tracked the “Most Consistent” category. HDFC Bank, at one point, grew profit 30%+ for 38 consecutive quarters. These are not the flashiest names in any single year — they are the ones that simply never stopped.

The biggest creators tell a different, more cyclical story. Reliance has been the largest single wealth creator eleven times — but the crown rotates with the macro: Hindustan Lever and the consumer/MNC names in the 90s, the IT majors at the turn of the century, oil-and-gas and metals through the commodity boom, financials from 2011, and digital / Adani / defence-PSU names most recently. The single-company wealth record kept breaking upward — Reliance’s ₹3.1tn in 2008, TCS’s ₹3.6tn in 2014, Reliance’s ₹5.6tn in 2019, Reliance’s ₹11.2tn in 2024 — which is really just a chart of the Indian market getting bigger.

The durability lesson is blunt: the fastest creators almost never repeat (a new name tops the speed list nearly every year, and many later destroy wealth), while the consistent creators are boring, consumer-facing, high-ROE leaders that compound quietly for decades. The study’s own framing — speed thrills but kills — is the data, not a slogan.

What I’d Tell an Investor

Thirty years of the same experiment produce a short, confident answer.

Buy businesses that earn a high return on capital and can keep earning it — the rarest and most valuable trait is longevity of that return, not its peak level. Demand profitable growth, but treat growth that requires endless fresh capital as a red flag, not a virtue. Insist on management that is both honest and competent; walk away at the first whiff of accounting games, because the downside there is total. And refuse to overpay: a payback ratio under 1x (market cap below the next five years of profit) or a PEG under 1x have been, across two decades of the firm’s own data, the most reliable tickets to outperformance.

Then concentrate — 15 to 20 names you actually understand — and hold through the boredom, because the compounding lives in years eight through twenty, not in the first three. Expect most of your wealth to come from a handful of consumer-facing market leaders that never make the year’s exciting headlines. And remember the study’s own scoreboard: of the top 500 companies of 1995, only 100 beat a mediocre index over the next 25 years. The job is not to find a hundred winners. It is to find a few great businesses, buy them right, and let time do the heavy lifting — because “stock returns are slaves of earnings power and growth; in the very long run, valuations matter less.”