Transcript
There is a claim you will encounter repeatedly in investing: that some product, strategy, or opportunity delivers strong returns with minimal risk. The claim is always worth examining carefully, because the risk-return trade-off — one of the most durable principles in finance — says that genuinely high returns and genuinely low risk cannot coexist for long in a competitive market. Understanding why that is true, where the exceptions are, and where the principle itself gets complicated, is what this episode is about.
This matters to every financial decision you make — keeping money in cash, putting it in a pension fund, paying down a mortgage early, buying into a savings product. Each of those involves a risk-return trade-off whether you consciously choose one or not. Ignoring the trade-off does not make it disappear. It just means someone else's assumptions are governing your money.
Before making that argument, though, it is worth taking the strongest version of the opposing view seriously. Because there is a genuine counterargument here, and it is not a weak one.
The counterargument runs like this: the risk-return relationship, as a reliable empirical fact, is far less stable than textbooks suggest. Researchers have studied U.S. markets across data going back to January 1926. What they found, according to work surveyed in the ScienceDirect literature on the risk-return trade-off, is that the relationship between variance and expected return is often close to zero, and sometimes negative. The dotcom bubble is the most vivid case. During that period, both market prices and market variance increased simultaneously — the opposite of what the theory predicts. If taking more risk reliably produced higher returns, that episode should not have happened. And yet it did.
Researchers studying the period from 1836 to 2003 in U.S. and U.K. markets — work attributed to Lundblad in 2007 and cited in the fresh look at the risk-return trade-off published in ScienceDirect — did find a significantly positive intertemporal relationship between risk and return. But forty years of intensive academic research have not produced a settled consensus. The relationship, as the same source puts it, is "elusive." Some attribute the weak findings to measurement problems — we simply cannot observe expected risk directly, only realised outcomes. Others argue investors sometimes behave irrationally, chasing momentum in ways that temporarily decouple price levels from underlying risk.
So the counterargument is serious: the risk-return trade-off is a theoretical prediction that real markets frequently violate. That is worth acknowledging.
Here is why the principle still holds as a practical guide. The empirical instability does not mean risk and return are unrelated — it means the relationship is noisy, not absent. Over very long horizons, the equity risk premium — the excess return that stocks have historically delivered over bonds — has been positive. A review of the equity risk premium published by Mercer found a long pattern of positive excess returns for equities versus bonds across many decades of recorded history. The question is not whether the premium exists but how stable it is and how much of it any given investor can expect to capture over their personal investment horizon. Rolling ten-year periods, according to the same Mercer review, show the premium swinging from deeply negative to strongly positive — making historical figures a poor predictor of what the next decade will bring. So the trade-off is real, but it plays out over time horizons that are longer and noisier than most people intuitively expect.
The formal framework for thinking about risk and return developed substantially in the mid-twentieth century, growing out of Modern Portfolio Theory as articulated in the 1950s. The core insight was that investors needed to be compensated for bearing uncertainty — that rational agents would only accept the possibility of losing money if the expected payoff was sufficiently higher than a risk-free alternative. That gave rise to the concept of the equity risk premium: the idea that stocks, being riskier than government bonds, should deliver higher average returns over time to attract capital. Without that premium, no rational investor would hold equities when safer assets existed.
That framework, as described across the financial literature including the Business Finance open educational resource from CAUL, treats the risk-return trade-off as a fundamental concept — the organising principle behind why different asset classes behave differently and why diversification matters. Its influence is enormous. It underlies how pension funds allocate assets, how financial advisers explain portfolios to clients, and how most introductory finance courses are taught around the world.
With that foundation in place, it is worth getting precise about what "risk" actually means, because the word covers several distinct things that are often conflated.
The first and most visible form is volatility — the degree to which an asset's price fluctuates up and down over time. Equities are highly volatile. A share market index can drop fifteen percent in a quarter and recover it the following year. Volatility is uncomfortable to experience, but in itself it is not permanent damage. If you do not sell during the drop, the fluctuation does not realise as a loss.
The second form is maximum drawdown — how far an asset can fall from its peak before recovering. Historical share market crashes have been severe. The 2008-2009 global financial crisis saw major equity indices lose roughly half their value. The 2000-2002 dotcom bust took the NASDAQ down nearly eighty percent from its peak. Recovery from those drawdowns took years in some cases and more than a decade in others. Drawdown risk matters enormously for anyone who needs to access money during a down market — and that is where volatility transforms from a temporary discomfort into something more consequential.
The third form is permanent capital loss — what happens when an investment goes to zero or near-zero and never recovers. A single company can go bankrupt. An emerging market bond can default. A highly leveraged position can be wiped out by a margin call. This is categorically different from volatility. Volatility is the price moving around a recoverable path. Permanent capital loss is a different destination entirely.
The fourth is risk of ruin — the scenario where losses are so severe, or so badly timed, that recovery becomes practically impossible even if the underlying asset eventually recovers. This connects directly to what is called sequence of returns risk. Two investors who both earn the same average annual return over twenty years, but in different order, do not end up in the same place. The investor who experiences severe losses in the early years — particularly if they are drawing down from the portfolio during those years, as retirees often do — ends up with far less than the investor who experienced the same losses later. The sequence matters. A portfolio that loses forty percent in year one of retirement may never recover sufficiently to sustain withdrawals, even if the market eventually bounces back strongly.
These distinctions — volatility, drawdown, permanent loss, and ruin — are not interchangeable, but they are often lumped together under a single word: "risk." Separating them helps explain why the same asset class can look very different depending on an investor's time horizon and cash flow needs.
Consider what happens at the other end of the spectrum — the supposedly safe assets.
Cash held in a savings account involves almost no volatility and carries deposit protection in many jurisdictions. But it is not risk-free. In the low-rate years around 2020, many savings accounts paid well under one percent while inflation ran higher; and even when interest rates rise, the return left after inflation and tax is often close to zero, or below it. The purchasing power of cash held in a low-interest account erodes quietly and continuously. Sarasin and Partners make this point explicitly: "taking too little risk may limit the ability of investments to grow over time, particularly after inflation." The risk is hidden — it shows up not as a price drop on a screen but as a gradual reduction in what your savings can actually buy.
Government bonds occupy a middle position. They are widely described as lower-risk than equities, but as Sarasin and Partners note, they are explicitly "not risk-free." Bond prices fall when interest rates rise, when inflation expectations increase, or when credit quality deteriorates. Any investor who held long-duration government bonds through a period of rising rates has experienced meaningful capital losses — without ever touching anything that looks like a high-risk asset.
This is the underlying logic of the principle: something promising high returns with low risk is, somewhere, concealing a risk. The risk might be inflation erosion. It might be liquidity risk — the inability to sell when you need to. It might be counterparty risk buried in the product structure. Or it might be the tail risk of catastrophic loss in precisely the scenario the product claims to protect against. The trade-off does not disappear; it relocates.
Which brings us to diversification — the one mechanism that actually does reduce risk without necessarily reducing expected return, and which comes closest to a genuine exception to the general rule.
The logic of diversification operates at the level of portfolio construction rather than individual assets. When you hold a single stock, your return is exposed to everything that affects that specific company — its management decisions, its competitive position, its regulatory environment, its luck. That is called unsystematic or idiosyncratic risk. It is company-specific and, crucially, it is not compensated with higher expected return, because rational investors can eliminate it simply by holding many different assets.
When you combine assets whose prices do not move in perfect lockstep — ideally assets with low or negative correlations — the volatility of the combined portfolio is lower than the average volatility of its individual components. The Business Finance open educational resource from CAUL frames this directly: the total risk of a portfolio can be lowered without necessarily sacrificing expected returns. That is the "free lunch" label that sometimes attaches to diversification — you are not giving up return to get lower risk, you are removing the unnecessary, uncompensated portion of risk.
But the free lunch has limits. What diversification cannot eliminate is systematic risk — the risk that affects all assets simultaneously because it comes from economy-wide forces: recessions, financial crises, pandemics. When the COVID-19 shock hit markets in early 2020, almost every asset class fell together. Correlations that looked low in normal conditions spiked upward precisely when diversification was most needed. M1 states this plainly: diversification "does not guarantee a profit or protect against loss in declining markets." Diversification reduces risk relative to concentrated positions; it does not provide immunity to broad market downturns.
There is also a behavioural dimension that theory tends to underplay. Loss aversion — the well-documented tendency to feel losses more acutely than equivalent gains — causes investors to make decisions that damage their long-term position. Selling during a drawdown converts a temporary price decline into a permanent loss. Holding onto a losing position because of sunk-cost reasoning — because selling would make the loss "real" — can produce the same outcome through the opposite mechanism. The Business Finance CAUL resource identifies these patterns explicitly. They are not peripheral to the risk-return trade-off; they are central to why the theoretical relationship so often fails to translate into individual investors' actual experience.
One additional layer of complexity: recent academic work cited in the ScienceDirect overview on risk-return trade-offs found that forecasting the equity premium improved by as much as twenty percent when measures of cross-market dependence were included in the model. And a separate piece of research found that macroeconomic state variables explained about thirty-five percent of the relationship between policy uncertainty and expected returns, leaving roughly two-thirds unexplained by known factors. The implication is that even sophisticated models capture only a portion of what drives the risk-return relationship at any given moment. The rest reflects real uncertainty about future conditions that no model can fully resolve.
Pulling this together: the risk-return trade-off is real, durable, and worth taking seriously as an organising principle. Higher expected returns are associated with accepting greater risk. That is affirmed by decades of historical data on equity premiums and by the basic logic of competitive markets — if a high-return, low-risk opportunity were genuinely available, capital would flow to it until the excess return was bid away.
But the relationship is noisy. It does not hold cleanly over short periods. It can be temporarily reversed by speculative bubbles. And the word "risk" covers several distinct phenomena — volatility, drawdown, permanent loss, ruin, and hidden risks like inflation erosion — that require different frameworks to understand.
Diversification genuinely reduces unsystematic risk without reducing expected return, making it the most defensible near-exception to the general principle. But it does not protect against systematic, market-wide shocks.
A claim that something offers high returns with low risk is not automatically false. But it requires identifying precisely what kind of risk is being reduced and what kind remains — or has been relocated somewhere less visible.
What remains genuinely open is the question of how much of the equity risk premium reflects compensation for volatility, and how much reflects compensation for tail risk — the small probability of catastrophic permanent loss. The academic literature has not cleanly separated those two. And whether the empirical instability of the risk-return relationship reflects a fundamental truth about markets or a measurement problem that better models will eventually solve is, as of the current state of research, unresolved. Those open questions are not a reason to dismiss the framework. They are a reason to hold it carefully — to understand both what it tells you and where it reaches the limits of what any principle can reliably predict.