EV/EBITDA Sector Discount on India: High Returns, High Volatility
Indian equities aren't for the cautious. We ran a sector-relative EV/EBITDA screen on BSE and NSE stocks from 2005 through 2025, and the results reflect everything that makes Indian markets both exciting and punishing: 11.18% CAGR with a max drawdown of -54.88%. The strategy outperformed SPY by 3.53% annually, but the ride was brutal in bad years and extraordinary in good ones.
Contents
- Method
- The Screen
- What We Found
- Annual Returns
- 2007: The Bull Run That Built the Lead
- 2008: The Cost of High Beta
- Crisis Comparison
- The Recoveries Are Real
- Limitations
- Conclusion
Here's the full picture.
Method
We screened BSE and NSE listed stocks starting from when sufficient data existed in the FMP warehouse. Each January, we calculated sector median EV/EBITDA across qualifying stocks and selected the 30 names trading at the deepest discount to sector peers. We held for 12 months.
| Parameter | Value |
|---|---|
| Universe | BSE + NSE |
| Market cap minimum | ₹20B (~$240M USD) |
| EV/EBITDA range | 0.5x to 25x |
| ROE filter | > 8% |
| D/E filter | < 2.0 |
| Sector discount required | ≥ 30% below sector median |
| Selection | Top 30 by deepest discount |
| Rebalancing | Annual (January) |
| Cash condition | If fewer than 10 qualify |
| Data source | Ceta Research (FMP financial data warehouse) |
| Filing lag | 45 days (point-in-time, avoids look-ahead) |
| Period | 2000-2025 (effective 2005-2025) |
| Benchmark | SPY (S&P 500 Total Return) |
| Returns currency | INR |
Why EV/EBITDA over P/E in India?
Indian markets have a high concentration of capital-intensive businesses in industrials, materials, and infrastructure sectors where debt loads are significant. P/E is distorted by interest charges in those sectors, making cross-company comparison misleading. EV/EBITDA strips out financing structure and depreciation policy, giving a cleaner picture of operating value relative to the enterprise.
The sector-relative framing matters in India because sector valuations diverge sharply. Consumer-facing companies in India routinely trade at 30-50x earnings while commodity businesses sit at 6-10x. Comparing absolute EV/EBITDA across sectors creates noise. Measuring discount relative to sector peers isolates the genuine mispricing signal.
Note on the benchmark. Returns are reported in INR. The benchmark is SPY in USD. This creates a currency comparison that overstates or understates relative performance depending on GBP/INR moves in a given year. We use SPY as a global reference point, not as a perfect India-specific comparison. Indian market returns should be contextualized against local indices (SENSEX/NIFTY 50) for a cleaner comparison.
The Screen
Here's the current sector-relative EV/EBITDA screen for Indian stocks, running against the live FMP warehouse.
WITH universe AS (
SELECT k.symbol, p.companyName, p.exchange, p.sector,
k.evToEBITDATTM AS ev_ebitda, k.returnOnEquityTTM AS roe,
fr.debtToEquityRatioTTM AS de, k.marketCap
FROM key_metrics_ttm k
JOIN financial_ratios_ttm fr ON k.symbol = fr.symbol
JOIN profile p ON k.symbol = p.symbol
WHERE k.evToEBITDATTM BETWEEN 0.5 AND 25
AND k.returnOnEquityTTM > 0.08
AND (fr.debtToEquityRatioTTM IS NULL OR (fr.debtToEquityRatioTTM >= 0 AND fr.debtToEquityRatioTTM < 2.0))
AND k.marketCap > 20000000000
AND p.sector IS NOT NULL
AND p.exchange IN ('BSE', 'NSE')
),
sector_medians AS (
SELECT exchange, sector,
PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY ev_ebitda) AS median_ev_ebitda,
COUNT(*) AS n_sector_stocks
FROM universe GROUP BY exchange, sector HAVING COUNT(*) >= 5
)
SELECT u.symbol, u.companyName, u.exchange, u.sector,
ROUND(u.ev_ebitda, 2) AS ev_ebitda_ttm,
ROUND(sm.median_ev_ebitda, 2) AS sector_median_ev_ebitda,
ROUND(u.ev_ebitda / sm.median_ev_ebitda, 3) AS ev_ratio_to_sector,
ROUND((1 - u.ev_ebitda / sm.median_ev_ebitda) * 100, 1) AS discount_pct,
ROUND(u.roe * 100, 1) AS roe_pct,
ROUND(u.de, 2) AS debt_to_equity,
ROUND(u.marketCap / 1e9, 2) AS mktcap_b
FROM universe u JOIN sector_medians sm ON u.exchange = sm.exchange AND u.sector = sm.sector
WHERE u.ev_ebitda / sm.median_ev_ebitda < 0.70
ORDER BY u.ev_ebitda / sm.median_ev_ebitda ASC LIMIT 30
Run this query on the live data: cetaresearch.com/data-explorer?q=CIjp4khP-I
What We Found
The strategy was in cash for the first five years (2000-2004). Indian listed companies with sufficient FMP coverage and passing all filters were too few before 2005 to reach the 10-stock minimum. The backtest effectively starts in 2005, giving us 20 invested years.
Over those 20 years: 11.18% CAGR, with a total return of 1,314.25% turning $10,000 into $141,000. SPY returned 7.64% annualized over the same full 25-year window, giving $10,000 turning into $63,000. The excess of +3.53% per year is real but earned through significant volatility.
| Metric | Strategy | SPY |
|---|---|---|
| CAGR | 11.18% | 7.64% |
| Excess return | +3.53% | — |
| Total return | 1,314.25% | 530.71% |
| Sharpe ratio | 0.146 | — |
| Sortino ratio | 0.274 | — |
| Calmar ratio | 0.204 | — |
| Max drawdown | -54.88% | -34.90% |
| Down capture | 69.13% | 100% |
| Up capture | 142.84% | 100% |
| Win rate | 60% | — |
| Cash periods | 5 / 25 | — |
| Invested periods | 20 / 25 | — |
| Avg stocks held | 26.0 | — |
The Sharpe ratio of 0.146 is low, reflecting the high volatility. The strategy generates alpha but at the cost of a very bumpy ride. Calmar of 0.204 (return per unit of max drawdown) is modest. The up capture of 142.84% is outstanding, but the down capture of 69.13% means you absorb two-thirds of every SPY decline, in a market where declines can be 50%+.
Annual Returns
| Year | Strategy | SPY | Excess |
|---|---|---|---|
| 2000 | 0% (cash) | -10.5% | — |
| 2001 | 0% (cash) | -9.2% | — |
| 2002 | 0% (cash) | -19.9% | — |
| 2003 | 0% (cash) | +24.1% | — |
| 2004 | 0% (cash) | +10.2% | — |
| 2005 | +37.2% | +7.2% | +30.1% |
| 2006 | +17.7% | +13.7% | +4.1% |
| 2007 | +74.5% | +4.4% | +70.1% |
| 2008 | -54.9% | -34.3% | -20.6% |
| 2009 | +62.6% | +24.7% | +37.8% |
| 2010 | +30.3% | +14.3% | +16.0% |
| 2011 | -28.4% | +2.5% | -30.9% |
| 2012 | +44.4% | +17.1% | +27.3% |
| 2013 | -2.5% | +27.8% | -30.2% |
| 2014 | +63.8% | +14.5% | +49.3% |
| 2015 | -7.8% | -0.1% | -7.7% |
| 2016 | +20.9% | +14.4% | +6.4% |
| 2017 | +44.2% | +21.6% | +22.5% |
| 2018 | -31.5% | -5.2% | -26.3% |
| 2019 | -13.3% | +32.3% | -45.6% |
| 2020 | +6.2% | +15.6% | -9.5% |
| 2021 | +49.5% | +31.3% | +18.2% |
| 2022 | +26.3% | -19.0% | +45.3% |
| 2023 | +47.4% | +26.0% | +21.4% |
| 2024 | +12.6% | +25.3% | -12.7% |
2007: The Bull Run That Built the Lead
No single year captures India's potential more than 2007. The portfolio returned +74.5% against SPY's +4.4%, a 70.1 percentage point gap. India's pre-GFC bull market was driven by infrastructure spending, domestic consumption, and capital inflows into a fast-growing economy. Sector discounts collapsed rapidly as institutional money re-rated the entire market upward. Companies that entered January 2007 at 30%+ discounts to sector peers saw that gap close as the broader sector re-rated.
This wasn't alpha through careful stock picking in a normal year. It was the right type of strategy at the right market phase. Sector-relative value in a re-rating bull market earns extraordinary returns.
2008: The Cost of High Beta
The same characteristics that produced +74.5% in 2007 drove -54.9% in 2008. Indian markets are high-beta relative to global equity. When GFC deleveraging hit, foreign institutional investors sold emerging market positions. The correlation of all assets went toward one. No fundamental filter could protect against forced selling at any price.
This isn't a flaw in the EV/EBITDA methodology. It's the structural reality of investing in a market that still depends on foreign capital flows and moves with global risk appetite. The strategy fell harder than SPY (-54.9% vs -34.3%) because the underlying market fell harder.
Crisis Comparison
| Event | Strategy | SPY | Excess |
|---|---|---|---|
| GFC 2008 | -54.9% | -34.3% | -20.6% |
| Eurozone stress 2011 | -28.4% | +2.5% | -30.9% |
| Taper tantrum 2013 | -2.5% | +27.8% | -30.2% |
| India slowdown 2018-2019 | -37.6% cumulative | +27.4% cumulative | -65.0% |
| COVID 2020 | +6.2% | +15.6% | -9.5% |
| Rate shock 2022 | +26.3% | -19.0% | +45.3% |
The 2018-2019 period was particularly painful. The strategy lost -31.5% in 2018 (India NBFC crisis, IL&FS default, rupee depreciation) and -13.3% in 2019 (slow recovery, credit squeeze). SPY gained in both years. Two years of heavy underperformance require conviction.
The Recoveries Are Real
For every painful drawdown, the recoveries were sharp. After -54.9% in 2008, the portfolio gained +62.6% in 2009. After -28.4% in 2011, it gained +44.4% in 2012. After the 2018-2019 trough, the portfolio gained +49.5% in 2021, +26.3% in 2022, and +47.4% in 2023.
The mean-reversion mechanism works. Companies that fall to deep sector discounts tend to recover when the macro environment normalizes. India's growth story reasserts itself.
Limitations
Currency mismatch. Returns are in INR, benchmark in USD. A strengthening dollar or weakening rupee will mechanically increase the apparent performance gap. A weakening dollar will compress it. This is a real consideration for non-Indian investors.
High drawdowns. A max drawdown of -54.88% will exceed many investors' risk tolerance. The strategy requires the ability and willingness to hold through 50%+ portfolio declines.
Early cash periods. Five cash years from 2000-2004 reflect data thinness, not strategy choice. Investors with a true India strategy would have been invested in something else during that period.
Systemic exposure. India remains a high-beta market tied to global risk appetite. The EV/EBITDA screen provides no protection against EM-wide selloffs. The diversification benefit investors expect from emerging markets can disappear precisely when they need it.
Execution friction. Indian mid-cap stocks can have meaningful bid-ask spreads. Annual rebalancing in a 30-stock portfolio is manageable, but transaction costs and market impact aren't modeled.
Sector thinness. Some Indian sectors have fewer than 10 large-cap companies meeting quality filters. In thin sectors, sector median EV/EBITDA is computed from a small sample, making the relative discount signal noisier.
Conclusion
The sector-relative EV/EBITDA screen on Indian stocks delivers real outperformance: +3.53% per year over 20 invested years, with extraordinary up capture. The mechanism works in India. Sector discounts do mean-revert.
The honest trade-off is volatility that would test most investors. A -54.88% max drawdown, two consecutive bad years multiple times, and an early period in cash make this a strategy for patient capital with genuine conviction in Indian growth.
For investors already holding Indian equities, the sector-relative EV/EBITDA filter is a structured way to tilt toward the undervalued names within the markets you're already exposed to. It doesn't reduce India's inherent volatility, but it does pick the better-valued names within that volatility.
Have you backtested other value signals in India? Curious whether P/B or EV/Revenue behaves differently given the capital structure mix.
Part of a Series: Global | US | UK | Switzerland | Japan | Hong Kong | Germany | China
Data: Ceta Research (FMP financial data warehouse). Backtest covers 2000-2025 (effectively 2005-2025) using point-in-time financial data with 45-day filing lag. Returns in INR. Benchmark (SPY) in USD. Results are hypothetical and don't reflect actual investment results. Past performance doesn't guarantee future results.
