Why more funds may not mean better diversification
Many investors discover the weakness in their diversification after the damage has begun.
On ordinary days, a portfolio may look well spread. It may contain several funds, many stocks and enough category names to appear thoughtful. For instance, during the global financial crisis of 2008, many investors who held portfolios diversified across various sectors and funds observed that, despite this perceived diversity, most of their holdings declined simultaneously when the market turned. The portfolio statement had variety but the underlying portfolio had sameness.
That is the problem correlation helps reveal.
Correlation measures how closely two investments move in relation to each other. A reading close to +1 suggests similar behaviour, while a reading near zero points to a weaker relationship. A negative number means the two have often moved in opposite directions. An analogy can clarify this: just as two boats tied together will rise and fall in sync with the same waves, positively correlated investments tend to move together when markets fluctuate. For investors, the concept is less about mathematics and more about dependence. If many parts of a portfolio rely on the same market condition, then the extra holdings may add little resilience.
The illusion of owning more
Portfolio diversification is often confused with portfolio expansion. Adding another fund can feel prudent. Adding another category can create a sense of control.
However, this comfort can be misleading. A portfolio of five equity funds may still be exposed to similar sectors, market-cap segments or investment styles. The names may differ but the underlying behaviour may remain the same. During rising markets, this problem is easy to ignore because everything looks healthy. It is only during market stress that the similarity ceases to be theoretical.
This is why investors need to look beyond the number of holdings. Diversification should be judged by how much the portfolio depends on the same set of outcomes. Correlation is one way to examine that dependence. Portfolio overlap is another.
Correlation looks at behaviour while portfolio overlap looks at ownership. If two funds hold many of the same stocks, their ability to diversify each other is limited. That is, they have a high portfolio overlap. If two strategies hold different stocks but still respond to the same market forces, correlation will show that weakness. Both checks are useful. Together, they offer a better view of whether a portfolio is genuinely spread across different drivers of return.
Why correlation matters in investing
No equity portfolio can escape equity-market risk. When the broader market falls sharply, even well-designed equity strategies can decline together, as seen during the global financial crisis of 2008, when most diversified portfolios suffered significant losses despite containing a variety of holdings. This limitation highlights the fact that correlation analysis, while useful, should be interpreted with caution. A low correlation among portfolio components does not guarantee protection against systemic risk (i.e., broad market risk); during major market downturns, previously uncorrelated or weakly correlated assets may suddenly move in tandem due to broad market forces. Therefore, correlation does not promise safety but rather serves to diagnose potential areas of concentration within the portfolio.
A highly correlated portfolio is one in which many parts behave similarly. That may be acceptable if the investor knowingly wants a concentrated exposure. It becomes dangerous when the investor assumes diversification exists merely because the portfolio has many components.
Lower correlation can help because it reduces dependence on a single pattern of return. When one part of the portfolio struggles, another may hold up better or recover sooner. The benefit appears over full cycles rather than every month or quarter. This is important because investors often abandon diversification precisely when it is doing its job: allowing some parts of the portfolio to lag while others lead.
So, the behavioural trap is obvious: investors prefer what has recently worked. After a strong rally, momentum looks attractive. After a recovery in beaten-down stocks, value appears vindicated. After a volatile phase, quality and low volatility feel more sensible. A portfolio built solely around the latest winner is less a strategy and more a reaction.
Correlation helps resist that impulse. It shows why different return drivers deserve attention before the next phase is known.
Factor investing and different sources of return
In equity investing, factor investing offers a useful way to think about diversification. A factor is a characteristic that helps explain the risk and return of securities. These characteristics may include valuation, price behaviour, volatility, earnings quality or financial strength.
Quality investing focuses on stronger businesses. Value investing looks for stocks that appear attractively priced relative to fundamentals. Momentum investing follows price strength. Low-volatility investing favours stocks that have historically shown more stable price movements.
Each factor reflects a different logic. None leads in every market. And that limitation is the point.
Our internal research on factor performance across market cycles illustrates how factor leadership can shift over time. This research analysed historical half-yearly returns for different factor indices within the Nifty 500 series, using publicly available index performance data spanning multiple market cycles.
Factor performance changes with market cycles
Different phases tend to favour different factors, which is why relying on a single style can leave a portfolio exposed to the wrong phase
| Recovery | Expansion | Slowdown | Contraction |
| NIFTY500 VALUE 50 TRI (46.81%) | NIFTY500 MOMENTUM 50 TRI (25.95%) | NIFTY500 LOW VOLATILITY 50 TRI (2.35%) | NIFTY500 LOW VOLATILITY 50 TRI (-10.58%) |
| NIFTY500 QUALITY 50 TRI (34.72%) | NIFTY500 VALUE 50 TRI (24.24%) | NIFTY500 MULTIFACTOR MQVLV 50 TRI (0.81%) | NIFTY500 MULTIFACTOR MQVLV 50 TRI (-12.01%) |
| NIFTY500 MULTIFACTOR MQVLV 50 TRI (32.34%) | NIFTY500 MULTIFACTOR MQVLV 50 TRI (19.8%) | NIFTY500 QUALITY 50 TRI (0.39%) | NIFTY500 QUALITY 50 TRI (-13.05%) |
| NIFTY500 LOW VOLATILITY 50 TRI (27.17%) | NIFTY500 QUALITY 50 TRI (17.32%) | NIFTY500 MOMENTUM 50 TRI (0.07%) | NIFTY500 MOMENTUM 50 TRI (-16.73%) |
| NIFTY500 MOMENTUM 50 TRI (23.59%) | NIFTY500 LOW VOLATILITY 50 TRI (16.03%) | NIFTY500 VALUE 50 TRI (-6.84%) | NIFTY500 VALUE 50 TRI (-16.92%) |
Source: NJ AMC Internal Research, CMIE, NSE, NJ AMC SmartBeta Research Platform. Half-yearly periods (Apr 2005–Mar 2026) are classied into four market phases; the shown returns are the average absolute half-yearly returns within each phase. Past performance may or may not be sustained and is not indicative of future returns.
During the recovery phases identified in our analysis, the Nifty 500 Value 50 TRI reported the highest average half-yearly returns among the factors compared. Throughout expansion phases, the Nifty 500 Momentum 50 TRI outperformed the others. Conversely, during periods characterised by slowdown and contraction, the Nifty 500 Low Volatility 50 TRI outperformed the other factor indices included in the study.
The lesson is sober. A single-factor portfolio carries factor-timing risk. The investor may believe they are buying a durable style, but they are also making an implicit bet on the market environment in which that style does well. Since future cycles are unknowable, diversification across factors can reduce reliance on one version of the future.
What factor correlations reveal
Moreover, the correlation data in NJ’s Factor Book adds an important layer to this discussion. It prevents a loose claim that all factors are low-correlated at all times. They are not.
Unlike the US and Europe, in India, equity factors can show positive correlations because they still belong to the same broader equity market. Quality, value, momentum, low volatility and multifactor strategies may all be influenced by broad market direction, liquidity conditions and investor risk appetite. This means that factor diversification should not be sold as a form of insulation from market risk.
Factor correlations in India are positive, but not identical
The differences in Indian equity factors are still meaningful enough to support diversification
| Factor | NJ Quality+ | NJ Enhanced Value | NJ Traditional Value | NJ Momentum+ | NJ Low Volatility+ | NJ Multifactor+ |
| NJ Quality+ | 1.00 | 0.79 | 0.69 | 0.73 | 0.83 | 0.86 |
| NJ Enhanced Value | 0.79 | 1.00 | 0.78 | 0.67 | 0.65 | 0.76 |
| NJ Traditional Value | 0.69 | 0.78 | 1.00 | 0.64 | 0.49 | 0.58 |
| NJ Momentum+ | 0.73 | 0.67 | 0.64 | 1.00 | 0.60 | 0.77 |
| NJ Low Volatility+ | 0.83 | 0.65 | 0.49 | 0.60 | 1.00 | 0.84 |
| NJ Multifactor+ | 0.86 | 0.76 | 0.58 | 0.77 | 0.84 | 1.00 |
Source: Internal research, CMIE, National Stock Exchange of India, NJ’s Smart Beta Platform (in-house proprietary model of NJAMC). Data is for the period starting from 30 September 2006 to 31 December 2025. The correlations are calculated using the daily excess return over the Nifty 500 total return index.
The more accurate point is that factors need not move in lockstep. Their correlations may be positive, but their return patterns still differ enough to matter. Just look at the data below to see what we mean.
Factor leadership shifts from year to year
No factor stays ahead consistently, which strengthens the case for combining distinct return drivers
*Does not represent a complete calendar year. Source: Internal research, CMIE, NJ’s Smart Beta Platform (in-house proprietary model of NJAMC). Calculations are for the specific periods mentioned in the respective column. NJ Quality+, NJ Momentum+, NJ Low Volatility+, NJ Traditional Value, and NJ Enhanced Value are in-house proprietary methodologies developed by NJ Asset Management Private Limited. The methodologies will keep evolving with new insights based on the ongoing research and will be updated accordingly from time to time. Past performance may or may not be sustained in the future and is not an indication of future return. The above is only for illustration purposes and should not be construed as indicative return of offering of NJ Asset Management Private Limited.
That distinction is crucial. Diversification does not require every component to move in the opposite direction. It requires enough difference in behaviour to reduce overdependence on one source of return.
The same idea applies to lower overlap. If a portfolio contains strategies that select stocks using different factor rules, the resulting holdings may differ from broad indices and from one another. Lower overlap can improve the quality of diversification, provided the underlying selection process remains disciplined.
How blending factors can smooth the journey
The case for blending factors becomes clearer when one looks at long periods rather than recent winners.
In our analysis, a blended allocation of 33% quality, 33% momentum and 34% value outperformed the Nifty 500 TRI in 16 out of 22 calendar years. The corresponding numbers for standalone quality and value were 12 out of 22, while those for standalone momentum were 15 out of 22.
Blending factors can improve the investment journey
It has historically shown its ability to increase the chances of outperformance
| Allocation | Calendar year outperformance | Maximum drawdown (%) | Volatility (%) | % of instances of 3-year outperformance against Nifty 500 TRI |
| 100% Momentum | 15/22 | -70.3 | 22.3 | 84.9 |
| 100% Quality | 12/22 | -53.6 | 18.1 | 70.4 |
| 100% Value | 12/22 | -66.1 | 25.5 | 56.9 |
| 33% Quality, 33% Momentum and 34% Value | 16/22 | -64.2 | 20.9 | 86.6 |
Note: Returns calculated over the period 1st April, 2005 to 30th April, 2026. Momentum, Quality, and Value are represented by respective Nifty500 Factor 50 TRI counterparts. Quality, Momentum, and Value are assigned a weighted of 33%, 33% and 34% respectively and are rebalanced annually. Numbers are per-tax. CY 2005 and 2026 do not represent the entire calendar year. Back-tested returns shown above are only for illustration purposes and should not be construed as indication for future return of any schemes or offering by NJ Asset Management Private Limited. Past performance may or may not be sustained in future and is not a guarantee of any future returns.
The blended portfolio also recorded the most three-year outperformance instances (86.56% of the time) against the Nifty 500 TRI. That was higher than standalone quality, standalone value and standalone momentum in the same analysis.
This does not mean the journey was painless. The blended allocation still had a maximum drawdown of -64.24%. So, equity investing remained equity investing. But what changed was the reliance on a single factor. With multiple factors, the portfolio had more than one way to participate in equity returns across different phases.
This is the practical benefit of correlation analysis: by combining factors that are less than perfectly correlated and whose performance leadership rotates across market cycles, investors can achieve a more stable overall portfolio experience. While this approach does not guarantee consistently positive results, it effectively reduces the risk of the entire portfolio underperforming when a single investment style falls out of favour. In practice, correlation analysis enables investors to construct portfolios that are more resilient to changing market conditions.
Why a rule-based process matters
Diversification is easy to describe and hard to maintain. The difficulty comes from behaviour. Investors chase what has worked, lose patience with what has lagged and confuse recent performance with permanent superiority. A process that depends on mood will struggle to hold diversified exposures through unfavourable phases.
This is where rule-based investing has a clear role. In a 100% rule-based investment philosophy, the kind that we practice, the key decisions are embedded in the design of the strategy. The rules define the stock-selection universe, the factor parameters, the portfolio-construction logic and the rebalancing discipline. Once the rules are set, the process does not change merely because one factor has had a strong or weak spell.
Keep in mind that this does not remove judgement. Rather, it moves judgement to the right place: the design of the rules, the testing of the parameters, the treatment of overlap and the discipline of implementation. After that, the process has to do what it was built to do.
The real test of diversification
Correlation does not predict which factor will lead next. It does not guarantee lower losses. It does not convert equity risk into certainty.
Its value is more modest and more useful because it helps investors identify hidden sameness.
A portfolio that appears diversified may still be vulnerable to the same market event, valuation regime or style reversal. Correlation helps test whether the portfolio is carrying different sources of behaviour. Overlap helps test whether it is repeatedly holding the same underlying securities. Factor analysis helps show whether return drivers are genuinely distinct.
Good portfolio diversification begins with a simple discipline: reduce avoidable dependence. In equity investing, factor diversification offers one way to do this. The results will still move through cycles. Some periods will favour one factor. Other periods will favour another. That is precisely why a disciplined portfolio should avoid depending entirely on one of them.
Investors cannot know the next market phase in advance. They can, however, build portfolios that are less fragile to being wrong.
FAQs
What is correlation in investing?
Correlation measures how closely two investments move in relation to each other. A high positive correlation indicates they have usually moved in the same direction. A lower correlation means their behaviour has been less closely linked.
How does correlation help in diversification?
Correlation helps investors understand whether different parts of a portfolio are genuinely behaving differently. When investments are less than perfectly correlated, the portfolio may be less dependent on a single source of return.
Is low correlation enough to reduce portfolio risk?
Low correlation helps, but it is not enough on its own. The underlying investments must also be sound, liquid and suitable for the investor’s risk profile. Correlation can also change during market stress.
What is portfolio overlap?
Portfolio overlap refers to the extent to which two funds or portfolios hold the same securities. High overlap can reduce the benefits of diversification, as the investor may be exposed to the same stocks through multiple routes.
How do factors help in portfolio diversification?
Factors such as quality, value, momentum and low volatility represent different stock-selection approaches. Since these factors may perform differently across market cycles, combining them can reduce reliance on a single investment style.
Does diversification remove market risk?
No. Diversification can reduce avoidable concentration, but it cannot remove market risk. Equity investments remain subject to volatility, drawdowns and periods of underperformance.
SEBI Registered Name (Number): NJ Mutual Fund (MF/076/21/02) | Details of Other Regulatory Registrations: https://shorturl.at/SEBua
Investors are requested to take advice from their financial/ tax advisor before making an investment decision.
MUTUAL FUND INVESTMENTS ARE SUBJECT TO MARKET RISKS, READ ALL SCHEME RELATED DOCUMENTS CAREFULLY.