Why Diversification Is the Only Free Lunch in Finance
Harry Markowitz in his office (2012)
When Harry Markowitz introduced Modern Portfolio Theory (MPT) in the 1950s, he wasn’t just adding a chapter to the finance textbooks—he was changing the way investors think about risk and return. His famous quip that “diversification is the only free lunch in finance” captured a truth that resonates to this day: by combining assets or strategies that don’t move in lockstep (i.e., they are orthogonal), you can reduce overall risk without necessarily giving up return potential.
Drawdowns, CAGR, and the Value of Offsetting Moves
Markowitz’s insight is best understood through the concept of volatility drag, sometimes called a “tax” on your portfolio. A critical point in understanding diversification is that while average (arithmetic) returns might look fine on paper, what really matters to investors is the compounded return—often referred to as CAGR (Compound Annual Growth Rate). Volatility is the enemy of compounding. When your portfolio suffers a loss, it takes more than the same percentage gain to get back to even. For example, a 10% drawdown reduces a $100 portfolio to $90. To return to $100, you need an 11.1% gain on $90—not 10%. That extra 1.1% is the “tax” you pay for experiencing volatility.
Why is this important? Because what really matters is your portfolio’s compounded (geometric) return, not the simple average of your gains and losses. The gap between arithmetic and geometric returns can become significant when volatility is high. Let’s illustrate with a simple sequence of returns:
Year 1: +20%
Year 2: –20%
Averaging +20% and –20% might look like a 0% average. But in reality, $100 grows to $120 after Year 1, then drops 20% in Year 2 to $96 ($120 × 80%), leaving you with a –4% total return. The volatility (a 20% swing in both directions) “drags” down your actual compound return. This is the volatility drag.
Tale of Two Strategies: Correlation Matters
Now, let’s break this out and imagine two strategies that both average +10% a year. If they’re perfectly correlated—meaning they often move together—your portfolio’s drawdowns can be severe when both dip at the same time, thus increasing the volatility drag. In contrast, if those strategies are less correlated (i.e., more orthogonal), one may hold steady or rise when the other dips, lowering your combined volatility and boosting your long-run compounded returns.
Below are two simplified examples showing how correlation (or lack thereof) affects a portfolio starting with $100 over three years. In these examples, each strategy independently averages +10% a year—but the timing of returns and how they offset each other makes all the difference.
Correlated Strategies Example
| Year | Strategy 1 Return | Strategy 2 Return | Combined Return | End Portfolio Value | CAGR |
|---|---|---|---|---|---|
| 1 | +10% | +10% | +20% | $120.00 | |
| 2 | -10% | -10% | -20% | $96.00 | |
| 3 | +10% | +10% | +20% | $115.20 | 4.80% |
Even though each strategy “averages” +10% a year over three years, the heavy drawdown in Year 2 acts like a “tax” on your portfolio’s compounded return—highlighting how volatility drag reduces long-term growth. The net result is a 4.80% CAGR.
Now, imagine two same strategies that both average +10% a year, but one tends to rise while the other dips. In other words, they are uncorrelated (or othorgonal). In this scenario, the combined portfolio’s swings could be less severe, because losses in one strategy are partly offset by gains in the other. Smaller swings mean less volatility drag and therefore a higher long-term growth path. It’s like trimming the peaks and filling in the troughs so your returns compound from a higher baseline each time. This offsetting behavior reduces the drawdown of the combined portfolio and leads to a more robust CAGR over time.
Uncorrelated Strategies Example
| Year | Strategy 1 Return | Strategy 2 Return | Combined Return | End Portfolio Value | CAGR |
|---|---|---|---|---|---|
| 1 | +10% | 0% | +10% | $110.00 | |
| 2 | -10% | +10% | 0% | $110.00 | |
| 3 | 0% | +10% | +10% | $121.00 | 6.60% |
Markowitz’s Enduring Legacy and Modern Portfolios
Harry Markowitz’s insight—that you could mathematically optimize a mix of assets for both risk and return—laid the groundwork for the entire discipline of quantitative finance. MPT has influenced everything from the Yale Endowment Model, which includes allocations to alternatives (hedge funds, private equity, and real assets), to retail “target-date” funds (yes, the ones that get talked about in retirement portfolios). In each case, the goal is the same: use a blend of assets with lower correlations to smooth out the ride.
Examples of Portfolios Built on This Concept
The “All-Weather” Approach: Popularized by Ray Dalio at Bridgewater, it seeks to balance a basket of assets—such as equities, bonds, commodities, and inflation-linked securities—so that no single economic scenario can sink the portfolio.
The Permanent Portfolio: Introduced by Harry Browne, this classic mix invests in stocks, bonds, gold, and cash in equal parts to harness the strengths of each asset class during different market phases.
Multi-Strategy Hedge Funds: Hedge funds have built portfolios designed to combine equity long/short, macro trading, arbitrage, and other uncorrelated strategies, aiming to capture alpha while managing drawdowns.
Although these portfolios differ in execution, the unifying theme is Markowitz’s free-lunch premise: mixing assets or strategies that don’t move together can yield higher risk-adjusted returns than any single component alone.
The 60/40 Dilemma and Why Orthogonality Matters
For decades, a 60/40 mix of stocks and bonds was considered the poster child of diversification. The idea was simple: when stocks decline, bonds (usually) go up, and vice versa. Yet in 2022, both asset classes declined in tandem—revealing that correlation can change over time, especially in periods of market stress. In other words, “diversification” was little more than a label if both parts of the portfolio moved together.
That’s precisely why orthogonality matters. True diversification isn’t about piling into different asset classes by name; it’s about searching for return sources that genuinely zig when others zag. When you can combine uncorrelated drivers of return, you reduce the overall volatility drag and help your portfolio retain a larger share of its gains. This fundamental principle is what Markowitz identified more than half a century ago, and it still underpins any serious portfolio construction effort today.
At Rembrandt Capital, we recognize that carefully engineered diversification can be the investor’s best friend. We develop and combine multiple quantitative strategies designed to be as orthogonal to each other as possible. By looking beyond traditional asset classes and analyzing deeper statistical signals, we strive to build portfolios that are robust across different market regimes.
Below is a chart highlighting the near-zero correlation (approximately -0.02) between SPY and a quantitative strategy illustrating how truly independent return drivers can help temper overall volatility and preserve capital through varying market conditions.
Similarly, this next chart highlights another near-zero correlation (approximately 0.03), further illustrating how distinct return drivers can help temper volatility when integrated thoughtfully.
Markowitz’s original observation still rings true: reducing volatility is the one reliable way to enhance a portfolio’s compound growth over time. The key is in combining allocations in a manner that genuinely offsets risk, rather than simply relying on conventional labels. Diversification done right is indeed a rare free lunch—and at Rembrandt Capital, our mission is to help investors partake in it by building statistically sound, orthogonal strategies that preserve capital through drawdowns and harness the power of compounding.