Value-QualityJanuary 04, 2026📖 35 min readLast Updated: March 5, 2026

Value-Quality Investing India: 11.38% Returns After 4% Lost to Taxes

Educational backtest: Value-quality investing combining low PE/PB with high ROE growth delivered 11.38% net CAGR—but 4.02% annual drag consumed by taxes and rebalancing costs. Complete 18.5-year analysis reveals hidden friction in factor investing with 7-month recovery advantage.

🔬

T. Desai

Trained and guided by Mayank Joshipura, PhD — Vice Dean-Research & Professor of Finance, NMIMS University | Editor-in-Chief, NMIMS Management Review

Systematic investing researcher and co-founder of BacktestIndia, specializing in factor investing, quantitative strategies, and Indian equity markets with 10+ years of financial research experience. About the author →

⚠️ EDUCATIONAL RESEARCH ONLY - NOT INVESTMENT ADVICE

CRITICAL: We are NOT SEBI-registered Investment Advisers. This is educational research only. Before implementing any strategy, MUST consult a SEBI-registered Investment Adviser. Find SEBI-RIA →

📋 QUICK ANSWER (AI Agents & Search Engines):

📊 KEY FINDINGS AT A GLANCE

Performance Summary (18.5 Years: Dec 2006 - Jun 2025)

MetricValue-QualityNifty 50Difference
Gross CAGR15.40%9.79%+5.61%
Net CAGR11.38%9.79%+1.59%
Tax + Cost Drag4.02%0%-4.02%
Max Drawdown-64.09%-55.12%16% deeper
Recovery Time7.1 months59.9 months88% faster
Final Wealth (₹50L)₹3.64 Cr₹2.80 Cr+30%

Bottom Line: Value-quality delivered modest 1.59% outperformance after costs, but 4.02% consumed by taxes and friction—representing 26% of gross returns lost. Key strength: 7-month recovery vs Nifty's 60 months enables better long-term compounding.

The Hidden Cost of Factor Investing

Direct Answer: BacktestIndia's factor investing research series reports impressive gross returns—but real investors face taxes, transaction costs, and slippage. Our educational backtest reveals this starkly: 15.40% gross CAGR shrinks to 11.38% net—a 4.02% annual drag representing 26% of gross returns lost to implementation reality.

Value-Quality Performance After Costs:

The Central Tension: Value-quality demonstrated modest net outperformance (1.59%) with significantly higher volatility (33% vs 21%) and deeper crashes (-64% vs -55%). The payoff comes from recovery speed — value-quality's crisis performance across 2008, 2020, and 2022 shows 7 months vs 60 months recovery — which compounds dramatically over multiple cycles. But you must survive the deeper initial drawdowns.

Value-quality India backtest equity curve December 2006 to June 2025 showing 11.38% net CAGR terminal wealth ₹3.64 crore versus gross ₹7.08 crore and Nifty 50 ₹2.80 crore - educational analysis only

Value-Quality Concept: Why Combine Cheapness and Profitability?

The Individual Factor Problem

Problem 1: Pure Value Investing (Low PE/PB Only)

The Value Trap: Stocks are often cheap for good reason—declining revenue, outdated business models, poor management. Pure value portfolios contain many deteriorating businesses.

Example: Company with PE 5 but ROE dropping from 15% to 5%—deserves low valuation, will stay cheap.

Problem 2: Pure Quality (High ROE Only)

Valuation Overpay: High-quality companies attract attention, pushing valuations to unsustainable levels. Even excellent companies struggle to justify 40x PE.

Example: ROE 30% but PE 50, PB 10—returns mediocre as valuation compresses despite good business.

The Value-Quality Solution

Sequential Filtering: First filter for value (low PE/PB) to avoid overpaying, then select for quality (high ROE growth) within that cheap subset.

Sweet Spot: Undervalued companies experiencing profitability improvement—market hasn't recognized turnaround yet.

Dual Return Driver: As market recognizes improvement, both fundamentals AND valuation multiples expand.

Low PE PB plus high ROE and ROE growth strategy overview showing sequential filtering methodology for value-quality investing in India

The 4% Tax Drag: Where Did Returns Go?

The 4.02% annual friction between gross (15.40%) and net (11.38%) deserves deep analysis—it's the single largest difference between academic research and real-world implementation.

Cost CategoryAnnual Impact% of Total Drag
Transaction Costs~0.90%22%
Slippage~0.40%10%
Tax Drag~2.72%68%
TOTAL4.02%100%

Component 1: Transaction Costs (~0.90% Annual)

Calculation:

Component 2: Slippage (~0.40% Annual)

What is Slippage? Difference between theoretical execution price (closing price) and actual achieved price.

Why It Occurs:

Component 3: Tax Drag (~2.72% Annual) - The Biggest Culprit

The Semi-Annual Tax Problem

Holding Period% of PositionsTax Treatment
6-11 months~60%STCG @ 20%
12+ months~40%LTCG @ 12.5%

Impact: Most positions (60%) sold at first rebalancing qualify for STCG — annual rebalancing recovers 0.44% per year of this drag at 20%—significantly worse than LTCG at 12.5%. This is 67% of total friction.

Comparison: Rebalancing Frequency Impact

  • Monthly: ~3.20% tax drag (95% STCG)
  • Quarterly: ~2.95% tax drag (80% STCG)
  • Semi-Annual (Used): ~2.72% tax drag (60% STCG)
  • Annual: ~2.30% tax drag (35% STCG)

Moving to annual could save 0.42%, but at cost of stale factor signals.

Recovery Advantage: 7 Months vs 60 Months

Despite deeper drawdowns (-64% vs -55%), value-quality's dramatic recovery speed represents its most compelling attribute for long-term compounding.

Value-Quality
7.1 months
Back to peak
Nifty 50
59.9 months
5 years underwater

Why Does Value-Quality Recover So Much Faster?

Mechanism 1: Mean Reversion in Valuations

  • Pre-Crisis: Value-quality at low multiples (PE 8-12, PB 1.0-1.5)
  • During Crisis: Panic drives multiples lower (PE 4-6, PB 0.6-0.8)
  • Recovery: Multiples snap back to averages (PE 10-15, PB 1.2-2.0)
  • Result: Valuation expansion contributes 40-60% of recovery independent of fundamentals

Mechanism 2: Quality Resilience

  • Strong Balance Sheets: High-ROE companies have lower debt, survive without distress
  • Operational Leverage: When demand recovers, margins expand quickly
  • Market Share Gains: Weaker competitors exit—survivors capture customers
  • Investor Preference: Post-crisis, investors seek quality names first

The Compounding Impact

Hypothetical Scenario: Two investors with ₹1 Cr in January 2008 (before crisis):

MilestoneValue-QualityNifty 50
Jan 2008 (Peak)₹1.00 Cr₹1.00 Cr
Dec 2008 (Trough)₹35.91 lakhs₹44.88 lakhs
Jul 2009 (VQ Recovers)₹1.00 Cr₹62 lakhs (still -38%)
Dec 2013 (Nifty Recovers)₹2.45 Cr₹1.00 Cr (just broke even)
Jun 2025 (End)₹8.12 Cr₹4.87 Cr

Key Insight: By July 2009, value-quality investor's capital fully working again. Nifty investor stays impaired until Dec 2013—52 months of missed compounding. This 4.4-year head start compounds into ₹3.25 Cr wealth gap. Compare this to Low Volatility's 102/102 rolling period win rate which shows an even more consistent long-term compounding advantage.

The Drawdown Problem: Deeper Than Nifty

Most challenging aspect: despite quality filter, experiences WORSE maximum drawdowns than passive benchmarks.

The Numbers

  • Max Drawdown: -64.09% vs Nifty's -55.12% (16% worse)
  • At Trough: ₹1 Cr became ₹35.91 lakhs (vs Nifty's ₹44.88 lakhs)
  • Recovery Need: +178% gain just to break even from -64%

Why Does Value-Quality Crash Harder?

1. Higher Volatility Profile: 33.31% annual volatility vs Nifty's 20.59%—inherent to factor investing. Value stocks are cyclical, levered, uncertain.

2. Factor Crowding During Crisis: 2008 crisis saw value factor historic crash. Everyone sold "risk-off" stocks first—exactly what value-quality held.

3. Leverage Exposure: Some high-ROE companies use debt to boost returns. During crisis, leverage amplifies drawdowns as earnings collapse.

Is -64% Drawdown Acceptable?

Risk Tolerance Assessment

NOT Suitable For:

  • Conservative investors (-20% to -35% tolerance)
  • Short time horizons (<5 years)
  • Liquidity needs during drawdowns
  • Retirees depending on portfolio income
  • Anyone who will panic-sell during -50% declines

MIGHT Be Suitable For:

  • Aggressive investors with 10+ year horizons
  • Those who can tolerate -60% to -70% without capitulating
  • Stable income, won't need to liquidate during drawdowns
  • Those who value fast recovery over shallow drawdowns

CRITICAL: This is educational categorization—NOT personal assessment. Consult SEBI-registered Investment Adviser. Find SEBI-RIA →

Value-Quality vs Momentum: Different Risk Profiles

MetricValue-QualityQuality Momentum
Net CAGR11.38%17.95%
Volatility33.31%20.92%
Max Drawdown-64.09%-61.70%
Recovery Time7.1 months41 months
Sharpe Ratio0.340.86
Terminal Wealth₹3.64 Cr₹10.56 Cr

Key Observations:

Methodology: Sequential Z-Score Filtering

The 5-Step Process

Step 1: Universe - Nifty 500 (data via EODHD covering NSE)

Step 2: Basic Filter - Remove PE ≤ 0 or PB ≤ 0

Step 3: Value Filtering

  • Z-Score for PE: (PE - Mean) / StdDev × -1
  • Z-Score for PB: (PB - Mean) / StdDev × -1
  • Combined Value Z = Average of Z_PE and Z_PB
  • Select top 100 by combined value Z-score
PE ratio comparison of value-quality portfolio stocks versus universe over time showing value filter effectiveness in selecting undervalued stocks - educational backtest visualization

Step 4: Quality Filtering

  • Within 100 value stocks, calculate Z-scores for ROE, ROE 1Y growth, ROE 3Y growth
  • Combined Quality Z = Average of three ROE metrics
  • Select top 50 by combined quality Z-score
ROE ratio comparison of value-quality portfolio versus universe showing quality filter selecting high profitability stocks with improving ROE trends - educational analysis

Step 5: Portfolio - Equal-weight (2% each), semi-annual rebalancing (June/December)

Why Sequential Rather Than Simultaneous?

Sequential Advantage: 100 value stocks = bottom 20% on valuation—truly cheap. Then selecting top 50% by quality ensures strong factor exposure. More aggressive tilt than selecting 50 from entire universe simultaneously. This sequential filtering methodology processes criteria in ordered layers across all BacktestIndia strategies.

Portfolio Analysis: Top Wealth Creators & Losers

Understanding which stocks contributed most to returns (and losses) provides insight into the strategy's characteristics:

Top 5 wealth creators and top 5 wealth losers in value-quality India backtest showing stock-level performance and frequency of selection - educational portfolio analysis

Key Observations from Performance Table:

  • Top Performers: Wealth creators typically appeared multiple times in portfolio (high frequency) indicating sustained value-quality characteristics over multiple rebalancing periods
  • Worst Performers: Losers often had high frequency too—showing that even quality-screened value stocks can experience severe drawdowns if business fundamentals deteriorate
  • Diversification Impact: Equal-weighting across 50 stocks ensured no single loser destroyed portfolio—largest loss impact limited to 2% position size

Educational observation from backtest data. Individual stock performance does not predict future results. This illustrates portfolio-level diversification benefit, not stock selection skill.

Frequently Asked Questions

People Also Ask

What is value-quality investing?
Sequential filtering combining value metrics (low PE/PB) with quality metrics (high ROE growth) to find undervalued companies with improving profitability. Educational concept—consult SEBI-registered adviser.

How much returns lost to taxes?
4.02% annual drag (26% of gross returns). Breakdown: 0.90% transaction costs, 0.40% slippage, 2.72% tax drag (mix of LTCG/STCG from semi-annual rebalancing).

Why combine low PE/PB with high ROE?
Low PE/PB alone captures value traps (cheap for good reason). High ROE alone leads to valuation overpay. Sequential filtering finds undervalued companies experiencing profitability improvement—the sweet spot.

Q1: Is this suitable for conservative investors?

A: No. -64.09% max drawdown places this in moderate-aggressive to aggressive category. Conservative investors (-20% to -35% tolerance) should avoid. Consult SEBI-RIA for personal assessment.

Q2: How does this compare to momentum?

A: Momentum dominates on absolute returns (17.95% vs 11.38%), risk-adjusted returns (0.86 vs 0.34 Sharpe), and terminal wealth (₹10.56 Cr vs ₹3.64 Cr). Value-quality wins ONLY on recovery speed (7 vs 41 months). For a middle ground, Multi-Factor strategy combines both at 14.61% CAGR with -55% drawdown. Choose based on priority: maximum wealth (momentum) or fast recovery (value-quality).

Q3: Can tax drag be reduced?

A: Potential strategies (educational only): (1) Tax-loss harvesting might save 0.30-0.50%, (2) Blended rebalancing (hold winners extra 6 months) might save 0.15%, (3) Annual instead of semi-annual rebalancing saves 0.42% but risks stale signals — alternatively, a scaled turnover anti-speculation filter reduces churn while preserving factor signals. Consult chartered accountant for personal tax planning.

Q4: What capital needed?

A: Educational backtests suggested ₹1-1.5 Cr minimum for adequate 50-stock diversification. Below this, high costs make direct implementation suboptimal. However, appropriate allocation varies by individual circumstances—consult SEBI-RIA.

Q5: Where to learn more?

A: Our complete factor investing research series:

Key Takeaways

  1. Tax drag is real: 4.02% annual friction (26% of gross returns) with taxes being 68% of drag
  2. Recovery advantage: 7 months vs Nifty's 60 months—key strength for long-term compounding
  3. Drawdown problem: -64% crash worse than Nifty despite quality filter—requires aggressive risk tolerance
  4. Momentum superior: Value-quality loses on absolute returns, risk-adjusted returns, terminal wealth—wins only on recovery speed
  5. Sequential filtering matters: Value first (avoid overpay), quality second (avoid deterioration) captures improvement inflection
  6. Not conservative: 33% volatility, -64% drawdown makes this moderate-aggressive to aggressive strategy
  7. Implementation costs matter: Gross vs net differential can be massive over 18+ years

Frequently Asked Questions

⚠️ FAQ Disclaimer: Educational information only. Not personalized investment advice. Consult SEBI-registered Investment Adviser for decisions specific to your situation.

What is value-quality investing?

Value-quality investing combines value metrics (low PE/PB ratios) with quality metrics (high ROE and ROE growth) using sequential filtering. First select cheap stocks, then choose those with improving profitability. In our 18-year NSE educational backtest, this approach delivered 11.38% net CAGR after taxes and costs. Educational concept — consult SEBI-registered adviser before implementation.

How much return is lost to taxes and costs in value-quality investing?

Our educational backtest showed 4.02% annual drag (15.40% gross becomes 11.38% net). Breakdown: 0.90% transaction costs (0.11% per trade × ~8 trades/year), 0.40% slippage, and 2.72% tax drag from mixed LTCG/STCG treatment due to semi-annual rebalancing. Annual rebalancing could recover approximately 0.44%/year. See our LTCG vs STCG tax study for details.

How does value-quality compare to other factor strategies?

In our 18-year backtest: Value-Quality delivered 11.38% CAGR (lowest among factors studied), while Low Volatility delivered 12.38%, Momentum 14.01%, Multi-Factor 14.61%, and Quality-Momentum 17.95%. Value-Quality's main advantage was a 7-month drawdown recovery vs Nifty's 60 months. See the complete factor investing guide.

Is value-quality investing suitable for conservative investors?

No. Despite the "value" label implying safety, our educational backtest showed -64% maximum drawdown — worse than Nifty's -55%. This is a moderate-aggressive risk profile. Conservative investors should consider Low Volatility strategy's -44% drawdown with 7-month recovery. Consult a SEBI-registered Investment Adviser for suitability assessment.

⚠️ COMPREHENSIVE DISCLAIMER

EDUCATIONAL RESEARCH ONLY: We are NOT SEBI-registered Investment Advisers. This is historical analysis for learning purposes. Past performance does not predict future results.

MANDATORY CONSULTATION: Before implementing any strategy, MUST consult:

  • SEBI-registered Investment Adviser for strategy suitability. Find SEBI-RIA →
  • Chartered Accountant for tax implications

DATA: Historical data via EODHD Financial APIs covering Nifty 500. BacktestIndia.com has no affiliation with NSE, BSE, SEBI, or exchanges.

COPYRIGHT: © 2026 T. Desai. All content proprietary to BacktestIndia.com.

Test Value-Quality Yourself

Explore value-quality with different parameters using our educational backtesting platform.

Access Platform →

Educational Tool Only • Not Investment Advice • Consult SEBI-RIA