Market Efficiency and Price Movement: Why Prices Move the Way They Do

If you've ever watched a stock price jump 8% in seconds after an earnings call, or wondered why you can't seem to "beat the market" no matter how much research you do, you've bumped into one of the most influential — and most debated — ideas in finance: market efficiency.

This post goes deep on what the theory actually claims, the mechanics of how prices absorb information, where the theory breaks down in the real world, and what all of this means if you're actually investing money.

What "Efficient" Actually Means

In everyday language, "efficient" means something works well or wastes little. In finance, it means something more specific: a market is efficient if prices fully and instantly reflect all available information at any given moment.

This is the core idea behind the Efficient Market Hypothesis (EMH), developed largely by economist Eugene Fama in the 1960s and 70s, work that later earned him a share of the 2013 Nobel Memorial Prize in Economic Sciences. The logic runs like this:

New information about a company or asset becomes available — an earnings beat, a product recall, a Fed rate decision, a geopolitical shock.
A large number of independent, profit-motivated investors process that information and trade on it.
Buying and selling pushes the price to a new "fair" level almost instantly, as people bid the price up (if the news is good) or sell it down (if it's bad) until it reflects the new reality.
By the time you've heard the news, the price has already moved.

If this is true, today's price already contains everything knowable about an asset's value. There's no "free lunch" — no obvious mispricing sitting around waiting to be exploited, because if it existed, someone would have already traded on it and erased it. This is sometimes called the "no free lunch" principle, and it's the single most important intuition to take away from the entire theory.

It's worth being precise about what efficiency does not claim. It doesn't claim prices are always "correct" in some cosmic sense, that markets never crash, or that every investor is rational. It claims something narrower and more mechanical: that the process of price discovery is fast and competitive enough that systematic, repeatable opportunities to profit from public information tend not to survive for long.

The Mechanics: How Does Information Actually Get "Into" a Price?

It's worth pausing on the plumbing here, because the word "reflect" can sound abstract. Information gets into a price through a very concrete mechanism: order flow.

When good news breaks, more people want to buy than sell at the current price. Buyers start submitting orders; market makers and liquidity providers, seeing this imbalance, raise their quoted prices to balance supply and demand. Algorithmic trading firms — many of which exist specifically to detect and react to news milliseconds faster than humans can read it — parse headlines, earnings releases, and even satellite imagery or shipping data, and place trades within fractions of a second. Highly liquid, heavily traded stocks can reprice within milliseconds of a major headline; that's the EMH mechanism working about as fast as physically possible.

This is why, by the time a typical retail investor reads "Company X beats earnings expectations" on a news app, the stock has often already moved most of the way to its new level. The retail investor isn't slow because they're unintelligent — they're slow relative to a structure of latency-sensitive trading infrastructure, professional analysts on conference calls, and institutional desks that are built for exactly this kind of race.

A useful way to think about it: the market price isn't really "the value of the company." It's closer to "the consensus belief, weighted by money at stake, about the value of the company, continuously updated by everyone trading in real time." Prices move not when something is objectively true, but when the market's collective expectation changes.

Expectations, Not Just News

This distinction matters a lot. A company can report record profits and the stock can still fall — because the market had already priced in even higher expectations. Conversely, a company can report a loss and the stock can rise, if the loss was smaller than feared. This is why financial commentators often say "it's not about the news, it's about whether the news beats expectations." Under EMH, this makes complete sense: the old price already encoded the expected outcome. Only the surprise — the gap between what happened and what was expected — should move the price.

The Three Flavors of Efficiency

Fama's original 1970 framework breaks EMH into three nested forms, each making a progressively stronger claim about what kind of information is already "priced in."

Weak-form efficiency

Prices reflect all information contained in past prices and trading volume. If this holds, technical analysis — studying chart patterns, moving averages, support and resistance levels — shouldn't reliably generate excess returns, because any pattern in historical price data that could predict future prices would already have been exploited away by traders looking for exactly that pattern.

This is the form of EMH with the most empirical support. Decades of studies on simple technical trading rules generally find that, after accounting for transaction costs, they don't consistently beat a buy-and-hold approach.

Semi-strong-form efficiency

Prices reflect all publicly available information — not just past prices, but financial statements, analyst reports, macroeconomic data, news coverage, regulatory filings, and so on. If this holds, fundamental analysis using public information — reading 10-Ks, building discounted cash flow models, comparing valuation multiples — shouldn't give the average investor an edge either, because professional analysts with more resources and faster access have already done that work and the price already reflects it.

This is the most commonly tested and debated form. It's also where most of the real controversy in finance research lives, because there's a long list of documented anomalies (more on those below) that seem to contradict it, at least over certain time horizons or in certain segments of the market.

Strong-form efficiency

Prices reflect all information, public and private — including insider, non-public information. This is the most extreme version of the theory, and it's the easiest to falsify. We know strong-form efficiency doesn't hold in practice, because insider trading is both real and, historically, profitable — which is precisely why insider trading laws exist. If markets were strong-form efficient, insider trading wouldn't be illegal because it wouldn't be possible to profit from it.

Why Prices Move: The Random Walk and Its Cousins

A direct consequence of EMH is the random walk hypothesis: if prices already reflect everything currently known, then future price changes should come only from genuinely new information arriving after today. And new information, almost by definition, arrives unpredictably — if you could predict it, it wouldn't be new, and the market would have already priced it in.

This produces a striking statistical claim: price changes should look like increments of a random walk — each step independent of the last, with no exploitable pattern, not because the underlying economy is random, but because the next piece of news that will move the price hasn't happened yet, and nobody (with rare and illegal exceptions) knows what it will be.

A closely related, slightly more refined version is the martingale model: the best forecast of tomorrow's price is simply today's price (adjusted for an expected return, like the time value of money or a risk premium). You can't use any information available today to predict whether tomorrow's unexpected price change will be up or down.

This theoretical backdrop helps explain a very practical, well-documented empirical fact: most actively managed mutual funds underperform low-cost index funds over long horizons, after fees. Studies tracking active fund performance over 10, 15, and 20-year periods repeatedly find that the majority of active managers fail to beat their benchmark index net of fees — and, strikingly, the managers who do outperform in one period are often not the same ones who outperform in the next, suggesting a lot of what looks like skill is closer to noise.

A Common Misconception: "Random" Doesn't Mean "Directionless"

People sometimes hear "random walk" and assume it means stock markets can't be expected to go up over time. That's not the claim. Markets have historically trended upward over long horizons because of underlying economic growth, corporate earnings growth, and a risk premium investors demand for holding volatile assets. The randomness in "random walk" refers to the unpredictability of short-term fluctuations around that trend — not to the absence of any trend at all.

So Why Do Prices Seem to Overreact, Underreact, and Trend?

Here's where things get genuinely interesting, and where decades of empirical finance research have pushed back hard on the cleanest version of EMH. If markets were perfectly efficient in the semi-strong sense, we wouldn't expect to reliably see:

Bubbles — prices detaching dramatically from any reasonable estimate of fundamental value, like dot-com stocks in 1999–2000, U.S. housing in the mid-2000s, or various speculative crypto assets and "meme stocks" in more recent years. In a bubble, prices aren't just adjusting to new information — they're being driven by expectations that the price will keep rising regardless of fundamentals, a self-reinforcing dynamic that EMH in its purest form struggles to explain.
Momentum effects — stocks (and other assets) that have performed well over the past several months tend to keep performing relatively well for a while longer, and vice versa for losers. This is one of the most robustly documented anomalies in finance, found across many markets, time periods, and asset classes.
Mean reversion over longer horizons — somewhat in tension with momentum, stocks that have done very well over 3–5 years often see below-average subsequent returns, and vice versa, suggesting some long-run overreaction that eventually corrects.
Excess volatility — prices swing more, and more often, than changes in the underlying cash flows or economic fundamentals would seem to justify. Robert Shiller's research on this point — comparing the volatility of stock prices to the volatility of the dividends those stocks are claims on — was influential evidence against a simple, fully rational pricing story.
Post-earnings-announcement drift (PEAD) — when a company's earnings significantly beat or miss expectations, the stock price tends to continue drifting in that same direction for weeks or even months afterward, rather than jumping instantly to a new stable level. This is a direct and uncomfortable challenge to semi-strong efficiency, since it implies a systematic, exploitable pattern following public information.
Calendar effects — historically documented patterns like the "January effect" (small-cap stocks tending to outperform in January) or day-of-week effects. Many of these have weakened or disappeared since being discovered and publicized — itself an interesting piece of evidence, since EMH would predict that publicly known, exploitable patterns should erode as people trade on them.

These anomalies fueled the rise of an entire competing — or perhaps complementary — field: behavioral finance. The central argument of behavioral finance is that real human investors aren't the perfectly rational, instantly informed, emotionless calculating machines that the cleanest version of EMH implicitly assumes. Instead, real investors are subject to well-documented psychological biases, including:

Overconfidence — traders, especially after a string of wins, tend to overestimate the precision of their own information and trade more often and more aggressively than is rational, which tends to hurt net returns once costs are included.
Herding — following the crowd's behavior rather than independent analysis, which can amplify price moves well beyond what fundamentals justify, in both directions.
Loss aversion — the well-documented tendency to feel the pain of a loss roughly twice as intensely as the pleasure of an equivalent gain, which distorts decisions like holding onto losing positions too long ("hoping they come back") and selling winners too early.
Anchoring — fixating on a reference point, like the price you originally paid for a stock, rather than weighting current information appropriately.
Recency bias / availability bias — overweighting recent, vivid, or easily recalled events (a recent crash, a recent hot stock) relative to their actual statistical importance.
Disposition effect — a specific, empirically documented pattern combining loss aversion and overconfidence, where investors sell winning positions too quickly to "lock in gains" while holding losing positions too long to "avoid realizing a loss."

There's also a structural, non-psychological category of explanation worth mentioning: limits to arbitrage. Even if some sophisticated investors spot a mispricing, they may not be able to correct it. Short-selling can be expensive, risky, or restricted; capital available to "smart money" arbitrageurs is finite and sometimes pulled away exactly when opportunities are biggest (margin calls during a crisis, for instance); and mispricings can persist or even worsen before they correct — the old Keynes-attributed line that "markets can remain irrational longer than you can remain solvent" captures this well.

Reconciling the Two Views

Most economists and practitioners today land somewhere in between full EMH and full behavioral skepticism. The most defensible synthesis looks something like this:

Markets are largely efficient in the sense that matters most for everyday investors — it really is extraordinarily hard to consistently and reliably beat a well-diversified market index after accounting for fees, taxes, and transaction costs, especially in large, heavily traded, well-covered markets like U.S. large-cap stocks. But markets are not perfectly efficient, and mispricings can and do exist, particularly:

In less-followed, smaller, or less-liquid assets. A mega-cap stock like a major tech company has hundreds of analysts, instantaneous news coverage, and immense trading volume working to keep its price near consensus fair value. A thinly traded small-cap stock or an obscure corporate bond might have almost no analyst coverage at all, leaving more room for mispricing to persist.
During periods of panic, euphoria, or forced selling. Efficient pricing relies on enough rational, well-capitalized participants being willing and able to trade against mispricing. During a true panic — a margin-call-driven crash, a liquidity crunch — that capacity can temporarily vanish, and prices can move on flows and forced liquidation rather than considered judgment about value.
Over very short time horizons, where market microstructure (bid-ask spreads, order book depth, the mechanics of how trades clear) and short-term order flow matter more than fundamental analysis — this is the domain of high-frequency trading, which is really a different game from long-term investing.
Around structural quirks, like index fund rebalancing days, where stocks being added to or removed from a major index see predictable price pressure simply from funds mechanically buying or selling to match the new index composition — a pattern that has nothing to do with the company's actual prospects.

A useful, more modern framing comes from economist Andrew Lo's Adaptive Markets Hypothesis: rather than treating efficiency as a fixed, binary property a market either has or lacks, think of markets as an evolving ecosystem. Efficiency rises and falls with the population of competing strategies, the amount of capital chasing inefficiencies, regulatory changes, and the broader economic environment — closer to evolutionary biology than to a fixed law of physics. A strategy that worked brilliantly in one decade can stop working once enough capital piles into it, and entirely new inefficiencies can open up as market structure changes (the rise of retail trading apps and social-media-driven trading in the early 2020s being one recent example).

Famous Real-World Illustrations

A few well-known episodes help make these abstract ideas concrete:

The 1987 "Black Monday" crash, where major indices fell over 20% in a single day with no comparably sized piece of fundamental news to explain it, is often cited as evidence that prices can move on positive-feedback trading dynamics (in that case, portfolio insurance strategies that mechanically sold as prices fell, triggering more selling) rather than purely on updated information about company values.
The dot-com bubble saw companies with little or no revenue valued in the billions based on speculative future growth narratives, followed by a collapse of more than 75% in the Nasdaq Composite from its 2000 peak to its 2002 trough — a textbook example cited by behavioral economists as detachment from fundamentals.
"Meme stock" episodes in 2021, where coordinated retail buying (notably in GameStop) drove prices to levels far beyond what any reasonable fundamental analysis supported, illustrate how social coordination and short-squeeze mechanics can move prices independent of new information about the underlying business.
Conversely, the speed with which large, liquid markets digest major scheduled news — a Federal Reserve interest rate decision, for instance — is often cited as strong evidence for efficiency: within seconds of a rate announcement, futures, bond yields, and currency pairs across global markets reprice in a coordinated, economically sensible way.

Both kinds of episodes are real and well documented. That's exactly why the debate between EMH and behavioral finance hasn't been "won" by either side — the evidence is genuinely mixed, and which model fits best seems to depend a lot on which market, which time horizon, and which historical period you're looking at.

What This Means in Practice

For everyday investors, the practical takeaways are fairly grounded, even if the academic debate above never gets fully resolved:

Beating the market consistently, net of costs, is hard. Not impossible — but the odds are stacked against active strategies once you account for trading costs, taxes, fees, and the fact that you're competing against extremely well-resourced professional and algorithmic participants.
Low-cost, broad index investing is a rational response to this reality, not a concession of defeat. If a market is efficient enough that few people can reliably beat it, paying low fees to simply own "the market" tends to outperform paying high fees to a manager trying and mostly failing to beat it.
News moves prices fast — often before you can act on it. By the time a headline reaches a typical retail investor, professional and algorithmic traders with faster information access and execution speed have frequently already repriced the asset substantially. This doesn't mean retail investors can't invest successfully — it means reacting to news as a trading strategy is a much harder game than it looks.
Volatility and even apparent overreaction don't automatically mean irrationality — short-term price swings can reflect genuinely fast-changing expectations about an uncertain future. But persistent, extreme, multi-year deviations from any reasonable fundamental story are where behavioral and structural explanations tend to get the most empirical traction.
Diversification still does real work. Whether you lean toward the efficient-markets camp or the behavioral camp, neither view tells you which specific stock will outperform tomorrow — which is exactly why spreading risk across many assets remains one of the few strategies almost everyone in finance, regardless of theoretical camp, agrees is sound.

The Takeaway

Market efficiency isn't a binary switch you can flip to "on" or "off." Markets aren't perfectly rational pricing robots, and they aren't chaotic casinos either. They sit somewhere in between: mostly efficient most of the time, with information getting absorbed into prices remarkably — almost unsettlingly — quickly, especially in large, liquid markets with intense competition among professional participants. But that efficiency is punctuated by human psychology, structural frictions like the limits of arbitrage, and quirks of market mechanics that create real, if often short-lived and hard-to-capture, pockets of inefficiency.

Understanding both sides of this debate — the elegant, self-correcting logic of the Efficient Market Hypothesis, and the messy, very human reality documented by behavioral finance — gives you a far more realistic picture of why prices move the way they do, and a healthy dose of humility about how hard it actually is to outguess millions of other participants doing the same thing you are.

This post is for informational purposes only and isn't financial advice.

Market Efficiency and Price Movement