Efficient Market Hypothesis: Fact Or Fiction? "Efficient" refers to informational efficiency only.

The efficient markets hypothesis (EMH) in all of its forms, whether strong, semi-strong, or weak, is normative, not positive, i.e., it is an assertion about the way markets should behave in an ideal, utopian world, not a statement about the way markets actually do work in the real, practical world. Simple observation shows that the EMH in all its forms is fallacious. Both Kindleberger and Mackay give historical examples of stock market irrationality and inefficiency.

The efficient markets hypothesis may have advanced many academic careers, but it has not demonstrably increased the wealth of any investor over what would have been created otherwise. The EMH and the related capital asset pricing model, as opposed to the operating enterprise valuation model, may be useful as a standard of market perfection in studies of the market as a whole, but not in the valuation or selection of common stocks for investment.

The term "efficient" in the efficient markets hypothesis refers to informational efficiency only. It does not include mechanical operational efficiency or necessarily societal welfare efficiency.

The EMH explicitly assumes that all market participants have access to the same information in either a strong, semi-strong, or weak sense of the hypothesis.

• This simplifying assumption is chosen because it is necessary for mathematical tractability and thus highly convenient. 
• What makes this assumption unacceptably implausible is the meaning of the term "information" which is often overlooked. 
• Data is not information. Rather, information is data that has been processed and interpreted with judgment based on intelligence, knowledge and experience. 
• Does anyone believe that all market participants are endowed equally, not with access to data, but with the same intelligence, knowledge and experience? 

Competitive, properly-regulated markets may approach the semblance of "data efficiency" in the relative sense of eliminating arbitrage opportunities subject to trading costs and taxes, but no market is efficient in any absolute sense of equating price at all times to intrinsic economic value. This margin between value and price is the major key to successful value investing.

Buffett: What students should be learning is how to value a business.

John Kenneth Galbraith once slyly observed that economists were most economical with ideas: They made the ones learned in graduate school last a lifetime. University finance departments often behave similarly. Witness the tenacity with which almost all clung to the theory of efficient markets throughout the 1970s and 1980s, dismissively calling powerful facts that refuted it “anomalies.” (I always love explanations of that kind: The Flat Earth Society probably views a ship’s circling of the globe as an annoying, but inconsequential, anomaly.)

Academics’ current practice of teaching Black-Scholes as revealed truth needs re-examination. For that matter, so does the academic’s inclination to dwell on the valuation of options. You can be highly successful as an investor without having the slightest ability to value an option. What students should be learning is how to value a business. That’s what investing is all about.

http://www.berkshirehathaway.com/letters/2010ltr.pdf

Risk and Return - Find Investments offering High Returns with Low Risk

A positive correlation between risk and return would hold consistently only in an efficient market.  Any disparities would be immediately corrected, this is what would make the market efficient.

In inefficient markets it is possible to find investments offering high returns with low risk.  These arise 

• when information is not widely available, 
• when an investment is particularly complicated to analyze, or 
• when investors buy and sell for reasons unrelated to value.  

It is also common place to discover high-risk investments offering low returns.  Overpriced and therefore risky investments are often available

• because the financial markets are biased toward overvaluation and 
• because it is difficult for market forces to correct an overvalued condition if enough speculators persist in overpaying.  
• Also, unscrupulous operators will always make overpriced investments available to anyone willing to buy, they are not legally required to sell at a fair price.

Since the financial markets are inefficient a good deal of the time, investors cannot simply select a level of risk and be confident that it will be reflected in the accompanying returns.  Risk and return must instead be assessed independently for every investment.  

In point of fact, greater risk does not guarantee greater return.  To the contrary, risk erodes return by causing losses.

It is only when investors shun high-risk investments, thereby depressing their prices, that an incremental return can be earned which more than fully compensates for the risk incurred.

By itself risk does not create incremental return; only price can accomplish that.