Understanding the stockmarket rollercoaster

Almost every CEO has complained that the market doesn’t understand their company. But as Kevin Coyne and Jonathan Witter* point out you don’t need to understand the market only the people who make your share price move.

CEOs always want to know how the market will react to new strategies and other major decisions. Will a company's shareholders agree with a particular move, or will they fail to understand the motives behind it and punish the stock accordingly? And what can management do to improve the outcome?

Trying to predict stock price movements is necessary, of course. After all, when stock prices fall, the cost of borrowing and of issuing new equity can rise, and falling stock prices can both undercut the confidence of employees and customers and handicap mergers

But executives can dramatically improve the accuracy of their predictions. By adopting a more systematic, rigorous approach, corporate leaders can learn to understand individual investors as thoroughly as many companies now understand each of their top commercial customers. It is possible to know such customers well because there are only so many of them. Equally, only a finite number of investors really matter when it comes to predicting stock price movements.

Every CEO knows that when buyers are more anxious to buy than sellers are to sell, share prices rise - and that they fall when the reverse happens. But fewer CEOs know that not every buyer or seller matters in this equation.

Our research on the changing stock prices of more than 50 large US and European listed companies over two years1 makes it clear that a maximum of only 100 current and potential investors significantly influence the share prices of most large companies.

By identifying these critical individual investors and understanding what motivates them, executives can predict how they will react to announcements - and more accurately estimate the direction of stock prices.

Armed with these new and solid insights about how critical investors behave in specific situations, executives can make strategic decisions in a different light. Knowing what makes crucial investors buy, sell, or hold the company's stock allows CEOs to calculate what its share price might be after an announcement and to factor this calculation into their strategic and operating decisions.

The few that matter

It should come as no surprise that big trades can significantly move the needle on a company's stock price.

But typically, short-term changes in a company's stock price aren't the result of a single big trade. For the 50 companies whose quarterly stock price variations we studied, we consistently found that the majority of unique changes in each company's stock price resulted from the net purchases and sales of the stock by a limited number of investors who traded in large quantities. (By "unique changes," we mean those occurring relative to the rest of the market. In other words, they do not include price bumps or falls that coincided with the overall movements of the market or the sector.)

Although the number of crucial investors in a company ranged from as few as 30 to (more typically) as many as 100, in each case this set of actors had a dramatic impact on share prices. In the companies we studied, we could attribute from 60 to 80 percent of all unique changes, quarter by quarter, to the net trading imbalances of these investors.

Of course, the correlation between the buying or selling of large investors, on the one hand, and the price of a stock, on the other, can never be perfect. Smaller investors sometimes act in sync and overpower larger holders - as happened twice in two years with the shares of the European industrial company. News, rumours, and world events can spark broad market swings. But among the companies we have studied, the correlation is remarkably persistent (Exhibit 1).

Industrial marketing for investors

Few companies today get to know their top investors well enough to predict with any accuracy what will make those investors buy or sell more of their shares. The CFO of a large financial company, which was about to announce the divestiture of a major division, believed that he was "right on top of [our] investor base."

To develop the ability to make predictions about shareholders, companies should identify their stock price movers and calculate how many additional shares would be offered or sought in reaction to specific announcements.

Through background analysis and interviews, the companies must then analyse in depth the trading behaviour of these movers, developing trading profiles for each of them. Finally, companies should use the information in the profiles to predict which movers would be likely to react to specific corporate announcements by selling or buying in the short term and then calculate what this would mean for share prices.

Getting to know investors isn't a one-shot process. Companies must continually re-examine who is moving their shares -investors come and go. An ongoing dialogue with the movers deepens the knowledge of these companies and, over time, sharpens their ability to predict the actions of their critical investors.

However, most companies will need to beef up their investor relations capabilities to get the job done.

Identify critical investors

A company should begin its assessment by asking who has the potential to move its stock price. Some of the movers could be among the company's largest current shareholders. Some may be smaller holders who want to increase their ownership. And some are potential large players who do not yet own any of the company's stock but could purchase or short it in large quantities.

In other words, investors who count have both weight and a propensity to throw it around. Although the actual calculations needed to put together the list of movers are complicated - requiring more discussion than we can present in this article - a likely mover is someone who does or could reasonably account for at least 1 percent of a stock's trading volume for one quarter.

Movers are not necessarily a company's largest investors. Shareholders (such as family holdings or trusts) that have owned big blocks of the company's stock for a long time don't move it quarter to quarter. Neither do index funds unless the company is added to or dropped from an important index (or unless the fund's assets change dramatically).

Among the largest 20 investors of one big pharmaceuticals company we studied, only 10 were movers, and this proved to be typical of the companies we studied. What is more, nearly half of the large movers of the stock of the pharmaceuticals company over eight quarters from 1999 to 2001 weren't listed among its 20 largest investors during any single quarter.

Moreover, companies should add potential investors to the list of movers. For a large chemical business in our study, we analysed the way the positions of investors in other chemical businesses changed over time. Potential movers include those who have made money investing in other industries in similar circumstances. Investors who bet on the right players in an industry that consolidated, for example, may now be eyeing investments in other sectors on the verge of consolidation.

Potential movers may also be investors who purchased shares in a company's upstream or downstream suppliers and have a history of investing more broadly in the value chain. Some may have a taste for betting on companies that use certain capital models (high cash flow, say, or high leverage), have new CEOs, or face particular market changes or competitive conditions. To determine how many investors should go on the list - 40? 70? 100? - a company should test the accuracy of its predictions over previous quarters to arrive at the number that works best.

Too few will yield poor correlations between activity and stock prices; too many will add to the cost and complexity of the process. In addition, the list changes frequently. Our experience suggests that a mover typically stays on such lists for six quarters - long enough to give the company time to become familiar with it but short enough so that there will always be new movers to study.

Moving the movers

Once a company has identified its movers, the next step is to develop thorough profiles of all of them. Companies begin by conducting an "outside-in" analysis of each one, including its stated investment criteria and objectives and its trading patterns. Discussions with every investor give a company a chance to fill in the gaps in its understanding of its movers and to confirm its hypotheses about what they trade and why. The resulting profile should first describe how an investor makes decisions. To get this kind of information, companies must phrase the questions carefully in view of a US Securities and Exchange Commission (SEC) regulation that prohibits companies from disclosing material information to some but not all investors.

Typically, indirect questions work best. A company might ask investors why they purchased or sold their holdings in a particular business, for instance. But the company would actually be trying to understand why they sold their holdings after the business announced, for example, that it was investing in China.

Making predictions

With the movers identified and profiled, the investor relations staff and executives can make reasonable judgments about who will sell, buy, and hold. This process isn't merely a mathematical exercise, though it does involve many calculations. Besides assessing whether each investor will approve or disapprove of a given announcement, executives must estimate how many shares the investor is likely to buy or sell. They can be guided in these estimates by such details as the average trade the investor makes and whether the investor historically "bleeds" (buys and sells incrementally over time) or "blasts" (buys and sells quickly and in large blocks). This information gives the company an idea of the extent of the trading imbalance that will likely occur as a result of the announcement. Executives, guided by past imbalances in the company's stock and the way they moved prices, can use this estimate to make a rough assessment of how the stock price will react (Exhibit 2). Secondary, or knock-on, events should also be considered: if the stock price goes up or down, for example, what might momentum players do? This too can be derived from past patterns.

One company, recognising that it would take a hit, decided that it could do little about this except to prepare and manage its board. (In this case, estimates of what would happen to the stock price were extraordinarily accurate.) Another company decided to postpone a restructuring when it realised how far its stock price was likely to fall. In a third case, two companies were about to announce that they were merging. But the estimated dip in the acquirer's stock price after the announcement could have affected the deal (an equity and cash buy), so executives at the two companies used the profiles to identify investors who should be reached immediately and individually. Profiling also helped the companies tailor their communications to those investors. Even if no immediate decisions are pending, a company should try to predict probable moves by investors on a quarterly basis if not more often. Accuracy improves with practice.

Building capabilities

Companies that choose to adopt an industrial-marketing approach to investor relations will need to make at least two key changes. The first is to stop viewing the market as a monolithic entity that is judging a company's performance in an adversarial way. When the company's stock price changes, executives shouldn't ask why the market moved; they should pinpoint who bought, who sold, and why. In fact, managers should view investors much as managers in private companies view their corporate owners - and understand them just as well.

Second, companies will need to overhaul their investor relations units. In the vast majority of companies today, the investor relations function is largely administrative: it oversees the production, but not always the content, of regulatory and annual reports; it administers the registry of shareholders and sets up investor road shows, visits by analysts, and conferences; and it talks to shareholders - when they call. Instead, the investor relations unit will have to take on a more strategic role, almost as an adjunct to strategic planning. It will be responsible for managing the key-account process to identify movers and understand their behaviour. Its staff will have to test all major plans and announcements for their effect on the price of the company's shares and suggest modifications to those plans to bring them into better alignment with the views of key shareholders. Indeed, for the first time, the investor relations unit will become an important adviser to the CEO. But this approach calls for investor relations leaders who can stand up to the CEO and deliver bad news when necessary.

They will also have to be capable of handling tough interviews with investors who are pressing them for information they cannot divulge under SEC regulations or for competitive reasons. Sharp, independent, and analytical investor relations directors may emerge from the ranks of business development, strategic planning, or even, in some instances, internal auditing. Taking a more rigorous, structured approach to investor relations and stock price predictions clearly requires resources, including the time and attention of senior management. But given the importance of share prices, why would a CEO ever want to be left guessing?

* Kevin Coyne is a director and Jonathan Witter is an associate principal in McKinsey's Atlanta office. This is an edited version of an article they wrote for the McKinsey Quarterly

Disclaimer

The purpose of this database is to provide a full-text record of all articles that have appeared in the CDJ since February 1997. It is aimed to assist in the research and reference process. The database has a full-text index and will enable articles to be easily retrieved.It should be noted that information contained in this database is in pre-publication format only - IT IS NOT THE FINAL PRINTED VERSION OF THE CDJ - therefore there might be slight discrepancies between the contents of this database and the printed CDJ.