Efficient Market Hypothesis (EMH)
1 Introduction
Every valuation in this paper has leaned on a market price at some point. Chapter 14 offered market capitalisation as a valuation method. Chapter 16 computed a cost of equity from a market return and a beta estimated from market prices. Chapter 15 measured a bid premium against the target's share price, and Chapter 6 priced a rights issue at a discount to it.
All of that assumes the price means something. This chapter asks how far it does.
Two words that are not the same
An efficient market is one in which the market price of a security reflects all the information available about it.
A perfect market is one that responds to new information immediately — no delay, no transaction costs, no barriers to trading.
A market that is both is one in which quoted prices are as fair as they can be: they accurately and quickly reflect an entity's position and prospects, so nobody can systematically buy a security for less, or sell one for more, than it is worth.
Note also that “efficiency” here means informational efficiency — whether prices carry the information. It is a different question from whether a market is cheap to trade in, and a different question again from whether it directs capital to the best uses.
This recording covers the whole chapter. It predates the 2027 syllabus and says two different things on the chapter’s most examinable point; carry three things with you.
Weak form: where it first says that studying past price movements beats a weak-form efficient market — and calls that fundamental analysis — disregard it. Studying past prices is technical (chartist) analysis, which weak-form efficiency defeats. The recording corrects itself minutes later and its closing summary is right; the pitfall box under Two kinds of analysis gives the order to remember.
Real markets: it places real markets between semi-strong and strong form. The standard conclusion is that developed markets are at best semi-strong form efficient (How efficient are real markets?).
Dated: its Brexit illustration is dated.
2 The efficient market hypothesis
The efficient market hypothesis asks which information is already in the price. There are three possible answers, conventionally called the three forms of efficiency, and they are nested: each contains the information of the one inside it and adds to it.
The easiest way to hold them in mind is to start at the outside and let information drop out. Strong form contains everything. Take away what is known only inside the company and you have semi-strong form. Take away what has been published and you are left with the record of past prices alone, which is weak form.
Form | Information already reflected in the price | Can an investor beat the market? | What no longer works |
| Weak form | The record of past prices and trading volumes, and nothing else. | Yes — by analysing published information, because it is not yet fully in the price. (And, of course, by using inside information.) | Technical, or chartist, analysis. Past prices are already in today's price, so studying them predicts nothing. |
| Semi-strong form | The above plus all publicly available information: the financial statements, announcements, forecasts, news about the industry and the economy. The price moves when new information is published. | Only by using unpublished information — which is insider dealing, and is a criminal offence in every major market. | Technical analysis and fundamental analysis. Everything public is already priced, so studying it cannot give an edge. |
| Strong form | The above plus all information, published or not, including what is known only to the directors. | No. Not even an insider, because the price moves as soon as the information exists rather than when it is released. | Everything. There is no analysis of any kind that produces a systematic gain. |
A worked illustration of the difference. An exploration company's geologist finds oil. Under strong form the share price rises the moment the discovery is made, before anybody has announced anything. Under semi-strong form nothing happens until the announcement, and then the price adjusts fully and at once — so an insider who bought in the interval would profit, which is precisely why it is unlawful. Under weak form the announcement is not fully absorbed straight away, and an analyst working through what the discovery is actually worth could still find a share that is under-priced.
The random walk
Weak-form efficiency is the claim that successive price changes are independent. A share standing at a thirty-day low is exactly as likely to rise tomorrow as to fall, because everything that produced the low is already in the price; what moves it next is the next piece of news, and news is by definition not predictable from a chart.
This is why weak-form efficiency and technical analysis cannot both be right, and it is the source of the apparent paradox in the chart itself: any pattern reliable enough to trade on would be spotted, traded on by everybody, and thereby destroyed.
3 Two kinds of analysis, and which form of efficiency defeats each
Exam questions on this topic almost always come down to one thing: given a stated level of efficiency, which method of picking shares still works? Answering it requires the two methods to be kept apart.
| Technical (chartist) analysis | Fundamental analysis | |
| What it studies | Past share prices and trading volumes — the chart. Patterns, trends, support and resistance levels. | Published information — the financial statements, the order book, the industry, the economy — in order to compute what the share is intrinsically worth. |
| What it assumes | That patterns in past prices repeat, so the past predicts the future. | That the price will eventually converge on the intrinsic value, so a share trading below its computed value is worth buying. |
| The lowest level of efficiency that destroys it | Weak form. If past prices are already in the price, the chart contains nothing left to exploit. | Semi-strong form. If everything public is already in the price, so is everything the fundamental analyst is reading. |
The two are not the same thing and they are not defeated at the same level. Analysis of past price movements is technical or chartist analysis; analysis of published information is fundamental analysis. Getting them the wrong way round produces an answer that is exactly backwards, and it is the commonest error in this chapter. The order to remember is: weak form kills the chartist, semi-strong form kills the fundamental analyst as well, strong form kills the insider too.
4 How efficient are real markets?
The hypothesis is testable, and it has been tested more than almost anything else in finance. The broad findings, which a discussion answer is expected to know:
Weak form: largely supported. Successive price changes in large markets are close to independent, and mechanical trading rules based on past prices rarely beat simply buying and holding once dealing costs are taken into account.
Semi-strong form: broadly supported in large, developed markets. Prices react to announcements — earnings, dividends, bid approaches — within minutes, and the adjustment is on average neither too small nor too large. It holds much less well for small companies whose shares are thinly traded and which few analysts follow.
Strong form: rejected. Directors and other insiders do earn abnormal returns when they deal in their own companies' shares. That is the reason insider dealing is a criminal offence and the reason companies operate closed periods — and the existence and enforcement of those rules is itself evidence that the gains are real.
The standard conclusion, and the reasoning behind it
Developed stock markets are best described as semi-strong form efficient. The argument is short and it is worth being able to write it: if markets were strong form efficient, inside information could not be used to make a gain; laws against insider dealing exist and are enforced, at considerable cost, in every major jurisdiction; therefore gains from inside information must be possible; therefore the market is at best semi-strong.
The qualification matters as much as the conclusion. Efficiency is a matter of degree and it varies with the security: a heavily traded share in a large index is closer to semi-strong efficiency than the shares of a small company that two analysts follow and that trades a few times a week. That is one of the reasons Chapter 14 treats a thinly traded market price with caution.
5 What it means for financial strategy
The syllabus asks for the implications, and each one is a conditional: it says what follows if the market is efficient, and — usually more usefully — what follows if it is not.
5.1 The timing of new issues
In a fully efficient market the timing of an issue does not matter: the price is fair today and will be fair tomorrow, so there is no good moment and no bad one. Unless the market is fully efficient the timing does matter, because an advantage can be obtained by issuing just before or just after information reaches the market.
5.2 The price of a rights issue
If the market is efficient the size of the discount on a rights issue is irrelevant to shareholder wealth: the theoretical ex-rights price adjusts and the shareholder is neither better nor worse off (Chapter 6). The discount exists to make the issue certain of acceptance, not to make it attractive. In a less efficient market, or where shareholders do not understand the arithmetic, the discount may be read as a signal about the company's prospects.
5.3 Project evaluation and the cost of capital
If the market is efficient, the share price is fair, so the return the market requires from the company can be inferred from it — which is what makes a cost of equity calculable, and therefore what makes discounted cash flow appraisal possible at all (Chapter 16). If the price is not fair, the required return cannot be known accurately and the discount rate used to evaluate new projects rests on an unsound figure.
It follows that in an efficient market a company creates value for its shareholders by accepting positive net present value projects, and by nothing else: financial engineering that does not change the cash flows does not change the value.
5.4 Creative accounting and window dressing
A semi-strong efficient market reacts to what is published, so accounts massaged to look better can move the price — until the truth emerges. A strong-form efficient market would price the underlying reality regardless of the presentation, and window dressing would achieve nothing. Since real markets are not strong form, presentation can mislead, at least for a time — which is an argument for the disclosure requirements that Papers F1 and F2 examine, not an argument for using the technique.
There is a related and better-evidenced point: markets are not fooled by accounting changes that do not affect cash. A change of depreciation policy, or a reclassification that raises reported profit without changing what the business earns, has repeatedly been found to have little effect on share prices. What moves prices is news about cash.
5.5 Mergers and takeovers
Where the market is efficient, the price of the target's shares is already fair, so an acquirer paying that price cannot gain unless it can do something with the business that its present owners cannot — synergies, or better management. Unless those are significant, no material premium is justified (Chapter 15).
The premium quoted in the press is measured against the price before the bid, but in a semi-strong market the price will already have moved on the announcement or on rumour of it. The premium that matters is measured against an undisturbed price.
Where the consideration is the acquirer's own shares, an efficient market prices those shares fairly too, so a share offer cannot be made cheap by issuing over-valued paper — and bootstrapping raises reported earnings per share without creating value, because the market re-rates the combined price-earnings ratio (Chapter 14).
5.6 Whether the market price can be used as a valuation
If the market is efficient, the share price is fair: the investor receives a fair combination of risk and return, and the company can raise funds at a fair cost. If it is not, the market price is one piece of evidence rather than the answer — which is exactly how Chapter 14 treats it, alongside asset, earnings and cash-flow bases, and with the reminders that a quoted price is the price of a small parcel of shares and that a thinly traded price may be days out of date.
5.7 Signalling, and what the company should therefore do
Because the market reacts to what is published, what a company chooses to publish is a decision. A dividend cut, a change of accounting policy, a director's share purchase and a delayed announcement are all read as signals about information the market does not have (Chapter 7).
The practical conclusion for a financial manager is short: communicate clearly, promptly and consistently. In a market that prices public information, the way to be valued correctly is to make sure the information is public — and the way to be valued badly is to leave the market guessing.
6 The behavioural challenge
The hypothesis has never been short of critics, and the criticism has two parts: things the model does not explain, and reasons why it might not hold.
6.1 Anomalies — findings the hypothesis struggles with
Momentum and reversal. Shares that have performed well over the last several months tend to keep doing so for a while, and then to reverse. Both are patterns in past prices, which weak-form efficiency says should not exist.
Over- and under-reaction. Prices sometimes move much further on news than the news warrants and then drift back, and sometimes adjust only slowly over weeks.
The size and value effects. Small companies, and companies trading at low multiples of their book value, have historically earned more than their betas alone would predict (Chapter 16's last limitation).
Bubbles and crashes. Sustained, very large movements in whole markets that are hard to attribute to any change in the underlying cash flows.
6.2 Behavioural explanations
Behavioural finance explains those findings by observing that investors are people:
Overconfidence — investors overestimate the precision of their own judgements and trade too much.
Anchoring — a judgement stays close to an initial reference point, such as the price originally paid or the last forecast published.
Herding — following what others are doing, which is individually reasonable and collectively destabilising.
Loss aversion — a loss hurts more than an equivalent gain pleases, so losing positions are held too long and winning ones sold too early.
Availability and representativeness — a vivid or recent event is given more weight than a statistically more relevant one.
A defender of the hypothesis answers that irrational investors need not make prices wrong, because rational ones will trade against them until they are right. The reply is the limits to arbitrage: correcting a mispricing costs money, carries risk, and may take longer than the corrector can afford to wait — so a price can stay wrong for a long time even when somebody can see that it is.
What a financial manager should take from all this
Do not assume the market price is always right. It is the best single estimate available and it is where a valuation starts, but it is an estimate, it moves on sentiment as well as on information, and it prices a small parcel rather than control.
Do not assume you know better either. The evidence that a company's own management can reliably tell when its shares are mispriced is thin, and “our shares are undervalued” is one of the most common and least examined statements in corporate finance.
Act on the things that are not in dispute: take positive net present value projects, tell the market what it needs to know, and do not expect presentation to do the work of performance.
An efficient market is one whose prices reflect the available information; a perfect market is one that responds to new information immediately. The efficient market hypothesis asks WHICH information is in the price, and gives three nested answers: weak form (past prices only), semi-strong form (plus everything published), and strong form (plus everything known privately).
Which analysis still works depends on which form holds. Weak form defeats technical (chartist) analysis because past prices are already in the price; semi-strong form defeats fundamental analysis as well because everything published is in the price, leaving only inside information, which is illegal; strong form defeats even the insider.
The evidence supports weak form, broadly supports semi-strong form in large developed markets, and rejects strong form — which is why insider dealing is a criminal offence and why the standard conclusion is that markets are at best semi-strong efficient, and less efficient the more thinly a share is traded.
For a financial manager: if the market is efficient the timing of an issue and the size of a rights discount do not matter, the share price is a fair valuation, the cost of equity can be inferred from it, and value comes only from positive net present value projects and from real synergies in an acquisition. Because real markets are efficient only in part — and because investors demonstrably over-react, anchor, herd and follow momentum — treat the price as strong evidence rather than as the truth, and treat clear and prompt disclosure as part of the job.
7 Test your knowledge
Two quick checks before you move on: work through the flashcards to fix this chapter’s terms and definitions, then sit the objective questions for exam-style practice. Both mark themselves and explain the answers as you go.
Efficient Market Hypothesis (EMH)
22 questionsAnswer the questions one at a time. Your progress is saved so you can leave and come back.
Open chapter practice