Come across this article this morning. Quite interesting, at least make me feel I'm not a complete brainless investor :)
From
http://www.sonicboomerang.com/whitepapers/driving_shareprices.pdf.
So What Really Drives Share Prices?
Justin Winfield, Co-CEO, SonicBoomerang Neil Baird, VP Product Development, CCNMatthews
Traditional theories for explaining stock price changes have relied on the assumption that every company has an intrinsic value that is rooted in the forecasts of long-term profits for a company. These theories maintain that the stock market is fundamentally efficient and that rational thinking by the participants in the market will set the correct price for any company, given the set of circumstances in which the company operates.
This thought-piece contends that prices are far more dependent on psychological forces like fear (or greed, such as during 1997-2000) in the market than has been traditionally assigned them – and that these forces are in fact quantifiable. The actions of market participants whose activity sets share prices are dependent on their expectations of what is likely to happen to the price of any given stock or that company’s fortunes, and these expectations are based on the flow of information in financial and media channels.
Expectations are developed by people, and people have emotions – retail investors and professionals alike – so they can be volatile. Moreover, this volatility is increasing: the average mutual fund now turns over its assets entirely within a year, while anyone who has seen a chart of the stock market over the past 10 years can see the huge bubble driven by the media and more investors in the marketplace. Price changes of 3% in one day on the stock market are now reasonably frequent; 20 years ago this was very uncommon.
In essence, as more ordinary people directly invest in the stock market, as the amount of media has exploded (and media typically sensationalizes stories for maximal impact), and as professional investors increasingly resort to non economically derived factors of valuation while boosting their disposition to sell quickly 1, stock prices are increasingly driven by the sentiment, emotion, and psychology, all of which are driven by the various flows of information, such as news, media, and financial information channels.
IR officers of public companies are responsible for doing everything in their power to maximize shareholder value. That being so, the more they can control the flow of information, both from the company and between the market participants, the more likely they are to succeed with this objective. They must also understand the psychological forces at work in the market (indeed, within us all) and then use all of the tools at their disposal to optimise the positive expectations of their existing and potential investors.
1 This is due to several factors, including: all company’s accounting data is now considered suspect by the marketplace at best; more stocks, such as biotechs, do not have current cash flow off which to base future forecasts, forcing the analyst to use non-economic measures; and scandals like Enron, WorldCom etc have heightened the professional investor’s sense of risk while decreasing their confidence in their forecasts. For active investors and money managers, knowing all the drivers of a share price is paramount yet a significant information gap exists: few data sources assess the overall hype or psychology for a particular stock, or data that provides granular understanding of the changing risk profiles and appetites of the market as a whole.
Efficient Market Theory is No Help
Efficient Market Theory - EMT – brought us concepts like beta, the efficient frontier of investment prospects and, most dramatically, ushered in the idea of thousands of completely rational analysts and market participants emotionlessly deriving the real intrinsic values of securities, the result of whose actions drive undervalued stocks up and overvalued stocks down such that the price of a security in the market is nearly the same as the fundamental or intrinsic value of the security.
EMT argues that markets are efficient by comparing the returns from active money managers against the return of the stock market in general and finding that active money managers do not out-perform the stock market on average.
However, this result should be of no surprise. On average money managers cannot out perform the market for the following reasons:
- Active money managers (not passively indexed funds) represent well over half of the $ volume in the stock markets, and an even greater proportion of the $ traded on stock markets. Thus as a group active money managers are the market. It therefore stands to reason that the majority of them will not out-perform it, as the top 50% will always be above the other 50.
- It is probable that active money managers, including self directed investors & day traders, account for over 95% of the daily $ traded. As such they are not only the marginal investor in the market, but nearly all of the investors in the market.
- Given that it is the marginal investor who sets the price at which the market clears, it is therefore impossible for the average active money manager to out- perform the market. Another way of considering this is to recognize that those active money managers who out-performed the market out-performed it by taking money from the others, who therefore, of course, under-performed the market. Since the one equals the other, there cannot be any other conclusion than on average active money managers cannot out-perform the market.
- The existence of “deadweight” costs reduces the returns for all active participants in the market. These include trading costs, the salaries of money managers and analysts, the tools they employ, and the marketing costs of their funds. Such costs are wrapped up in the MER (management expense ratio) of a mutual fund, which is typically in the 2-3% range. An active money manager must therefore out- perform the market by 2-3% so that, net of costs, the fund’s performance is in line with that of the market.
It is therefore impossible for a group of persons to outperform the market when they are the market, particularly when their actions cost 2-3% of their net assets. In a market that is rising 10% a year, an MER of 2.5% represents approximately 25% of the gains a fund would expect to receive. In short, active money managers are some 25% behind the market at the outset, so it should be of no surprise that most will not out-perform the market.
Do markets efficiently price a stock relative to its intrinsic value?
So are markets efficient? The confusion lies in how one defines efficiency. If a test of efficiency measures whether active money managers as a group can or cannot out- perform the market, then the answer is yes. Markets appear to be efficient as money managers as whole will experience returns on par with or below the market.
But are markets efficient at pricing a stock relative to its intrinsic value? This is a much more serious test of market efficiency, but one that is impossible to measure. The reason is that there is no certain fundamental or intrinsic value to a security. Because of this, the test of market efficiency had to be constructed from an evaluation of active money managers against the market. But as we have just demonstrated, while the answer was satisfying to EMT proponents, this test was predestined to support their conclusions.
Analysts’ estimates of a stock’s fundamental value are also suspect
Perhaps the most common estimate of a stock’s fundamental value is the average of the forecasts issued by financial analysts. However, some notable problems exist with using analysts’ estimates of fundamental value. These include:
- The glaring problems of analyst bias and self-interest in the production of these forecasts. However, this is a problem germane only to sell-side research, which is likely the least used research for the purpose of fundamental analysis
- Forecasting the future outcome for a company’s stock price – that is dependent on up to hundreds of factors, each of which interacts with each other as well as on the stock price - is exceedingly difficult to do accurately. This is particularly true considering that analysts must forecast these variables beyond 5 years to come to a net present value. Few can reliably predict any variable’s level beyond the next 6 months.
- Analysts suffer from irrational thinking during the setting of the assumptions that underlie their models. Stocks they like - for whatever reason - will receive, say, a higher revenue growth forecast than stocks they don’t like. Emotion has now dictated part of the “fundamental value” they ascribe to a stock.
- Analysts must work in large part with company-provided accounting information. This information often has serious flaws, to the point of occasionally being nearly complete misrepresentations of reality
On the Difficulties of Forecasting
We contend that there is no “fundamental” value of a security, and that accurately forecasting the value of a company beyond a short period is extremely difficult if not impossible. The modelled value of a security is simply the net product of all the expectations of all the analysts and money managers about each of the variables that they believe affects the intrinsic value of the security; all of this is subject to great uncertainties.
The following variables are just some on which the share price of a company depends:
- Future revenues
- The future costs to produce these goods or services
- The appropriate discount rate/cost of capital
Each of these variables themselves is created by several other fluctuating variables, in the case of revenue forecasting, future demand for the company’s product or services and the level of competition for these dollars are just two variables that combine to create the overall revenue forecast – yet how does one quantify the competitive dynamics of an industry? This highlights again the difficulty of accurately forecasting any fundamental value, even though most financial modellers create simple linear flow models, which are not as representative of a dynamic entity like a company.
Advances in the field of physics, particularly that from the field now known as complex adaptive systems complexity, show that dynamic systems (ie. those in which the variables within the overall system interact with each other as well as upon the overall system) with as few as three variables can produce essentially random results. In short, chaos and chaotic outcomes exist within some of the simplest systems – yet a company’s economic future has dynamics far more complicated than 3 or 4 variables.
Of equal importance to the number of variables is the fact that all of them are expectations-based. The only basis, outside of the skill of the forecaster and analyst, on which to believe that these future expectations will be realized, is historical precedent, forcing the analyst to drive forward by looking mostly behind. Net net, human forecasting abilities are not very good beyond a few months at most.
So what is the Intrinsic Value of a Security?
Despite the above, it seems reasonable to believe that there should be an intrinsic value of a security based on its delivery of a certain level of profit for a particular period in time. It is reasonable also to believe the stock price would trade in a range around this intrinsic value. This intrinsic value will fluctuate based on the arrival of new information that causes the analysts to revise their expectations for a variable that will lead to a new intrinsic value.
Thus, EMT theorists assume that the intrinsic value of a security should be pretty close to its present stock price. In other words, the price of a stock in the market today impounds all known information, the rational thinking and expertise of the market participants set the fundamental value of a security, and the two are approximately the same.
The trouble with this assumption is that the stock market price often does not converge to what the “fundamental” price should be – particularly if the marginal investor pays little attention to the concept of intrinsic value. Examples of this would be the discrepancy of over 40% between the two classes of shares of the Royal Dutch Shell group that lasted for several months in 1998, the price of most stocks from 1999 – 2001, or the spectacular example of the 23% drop in the Dow Jones in one day in October 1987.
Is it reasonable to believe America became 23% less valuable within 6.5 hours on one day? Clearly, market participants – all of them, both retail and institutional, because in 1987 there were few retail investors as there are today – are a lot less rational, a lot less analytical than EMT proposes.
As everyone who has ever seen a trading floor can attest, prices are often set in fast moving situations, and buying and selling decisions are often made in conditions of extreme uncertainty and emotion. The instantly impounding efficiency proposed by EMT theorists is nowhere to be seen and fundamental values suggested by the models of active money managers are reduced to mere rules of thumb.
It boils down to one simple fact: markets are a mechanism that arrives at a net guess, an expectation, of the value of a security every new second of every trading day. They are efficient at ensuring that the average active managers in the market won’t out-perform it, but by no means does this imply that the market is efficient at reaching the fundamental value of a security. Warren Buffet has not become the world’s second richest man from buying accurately priced assets.
Everything connected with stock price evaluation is based on expectations. Prices are based on the expectations of the buyers and sellers of a stock. Expectations are driven by forecasts, which are extremely difficult to do accurately beyond a few days or months, and also by emotions, the hopes and fears of the individual. In each case the process starts and finishes within the human brain and is based on the expectations of the individual of the importance of the various factors, economic and psychological, driving a stock’s price as well as their goals and their emotions.
The propensity of humans to favour avoiding a loss vs. obtaining a gain, our propensity to hold our losers and sell our winners, our inclination to be swept along by group behaviour and the propensity of market participants to buy and sell far more than any rational asset holding theory would predict, all suggest that market values for stocks need not, and do not always approach the fundamental value of a security, if indeed there is an intrinsic share price.
This suggests that adequate forecasting theories should pay attention to not only the economic outlook for a particular company but equally important how market players expectations are set, the kinds of expectations possible and the impact these different expectations have on how certain outcomes are valued and probabilities assigned, how they change under conditions of uncertainty and flux, and the role of information and media flows in yielding useful information and in setting the general psychological framework / emotional makeup of the market participants.
Superior active money managers want to stay ahead of the others with better understanding of information flows, how these impact people’s expectations, and ultimately with quantitative insight into psychological drivers as well as economic drivers; IR professionals can stay ahead by understanding what are the different information flows about their company in the media and how to optimise their dissemination of information – their communications strategy – in order to minimize the impacts of bad news or maximize the impacts of good news.
Conclusions
Given:
- The difficulty of accurately forecasting the future beyond any short period of time, and
- The biases and errors that occur in modelling the fundamental value of a security;
- That the fundamental value of a company is a constantly moving target as information arrives in the market, and
- That active money managers constitute the bulk of the market
It is difficult to conclude:
- That the EMT test of examining active money management performance against overall market returns, particularly net of costs, is of any real use in proving market efficiency
- That fundamental forecasting is the only useful endeavour, or that there is such a thing as an intrinsic price of a security, outside of providing a general rule of thumb to use as one consideration in an investment decision at a particular time. Economic activity occurs in the context of constant flux, and with it the value of any company.
But rather:
- That stock prices depend on the expectations-setting processes of the market participants about many factors, not just the economic outlook for a company, which
- Suggests paying extremely close attention to the flow of information about a stock, such as the amount of information made available, the type of information that is being produced, how that information is being received and interpreted by the market participants, and also
- Suggests paying extremely close attention to how psychology and confidence levels generally, and that of market participants specifically, impacts how information is received, processed, and expectations are formed or altered under differing environmental conditions.
Regardless of whether a money manager uses fundamental analysis or technical analysis or even gut feel –a range broad enough to encompass probably all investment styles – all decisions ultimately proceed from the expectations and beliefs created by and constantly evolving as a result of information processed within the human brain. Clearly then, a properly grounded theory of capital market valuation needs to account for psychology and behavioural roles in the pricing of stocks, in addition to those of an economic or analytic nature.
The insights that rise from a unified view of market value drivers inform active money managers about how to best select stocks and securities to achieve above average returns, while assisting IR professionals in optimising their communications strategy to achieve the maximal share price possible in any set of given circumstances.