Monday, May 4, 2009

The crisis: Timing strategic moves

Timing is key as companies weigh whether to make strategic investments now or wait for clear signs of recovery. Scenario analysis can expose the risks of moving too quickly or slowly.

The McKinsey Quarterly

APRIL 2009 • Richard Dobbs and Timothy M. Koller

Source: Corporate Finance Practice

Corporate Finance, M&A Article, timing strategic moves

In This Article

It may be a nice problem to have, but even companies with healthy finances face a quandary: should they pursue acquisitions and invest in new projects now or wait for clear signs of a lasting recovery? On the one hand, the growing range of attractive—even once-in-a-lifetime—acquisitions and other investment opportunities not only seems hard to pass up but also includes some that weren’t possible just a few years ago. … On the other hand, many indicators suggest that the economy has yet to hit bottom. Companies that move too soon risk catching the proverbial falling knife, in the form of share prices that continue to plummet, or spending the cash they’ll need to weather a long downturn.

… How quickly the world economy returns to normal—and indeed, what “normal” is going to be—will depend on hard-to-predict factors such as the fluctuations of consumer and business confidence, the actions of governments, and the volatility of global capital markets. … In previous recessions, as many as six rallies were followed by market declines before the eventual troughs were reached.1 During the current downturn, market indexes fluctuated by an average of 20 percent each month from November 2008 to March 2009.

… And while better timing of acquisitions, and therefore the prices paid for them, can make a big difference in their ability to create value, the best way to minimize risk is to ensure that investments have a strong strategic rationale.

… To illustrate the risks, we conducted an analysis of a hypothetical acquisition. Real US market and economic data allowed us to build a range of scenarios embodying different assumptions about future US economic performance.2 We found that even scenarios assuming conservative levels of market performance (as indicated by the experience of past recessions) suggest that many industries may be reaching the point when acting sooner would be as appropriate as—if not better than—acting later. …

Analyzing scenarios

… Long-term profits are tightly linked to the economy’s overall performance: over the past 40 years, they have fluctuated around a stable 5 percent of GDP3 (Exhibit 1). … For our scenarios, we assume that US corporate profits will revert to some 5 percent of US GDP, although that estimate could be a conservative one if the trend to higher profits in the years leading up to the crisis resulted from a structural change in the economy. …

Back to top

Growth in the labor force and productivity drive the long-term growth of real GDP. Since it has historically grown in the range of 2.5 to 3.0 percent a year and returned to its former trend line in all US downturns from the Great Depression onward, with no permanent loss in GDP once the economy recovered, our base scenario assumes that both of those trends will continue. (We also examined scenarios reflecting the possibility that long-term GDP growth might be lower as a result of changing demographics, declining productivity growth, and the effects of the current financial crisis or that GDP might fall permanently by as much as 5 or 10 percent.) Finally, in normal conditions the market as a whole has a price-to-earnings ratio ranging from 15 to 17. We used that multiple in our 2.5–3.0 percent growth scenario and a lower one (14 to 16) in our 2.0–2.5 percent growth scenario. Both multiples are consistent with a discounted cash flow valuation of companies.4

Under the scenario that most resembles the course of previous recessions (no permanent loss of GDP and 2.5 to 3.0 percent long-term real growth), the stock market’s normal value in early 2009 (as measured by the S&P 500) would have been about 1,200 to 1,350. This implies that the stock market was trading, as of the end of March, at a discount of about 30 to 40 percent from its normal value (Exhibit 2). … You would have to expect a permanent GDP reduction of around 20 percent to see the March S&P 500 index level (around 800) as normal. A similar analysis for the performance of the stock market in previous recessions finds that at the trough of deeper recessions, it typically trades at a discount greater than 30 percent from its normal value (Exhibit 3).

Back to top

Back to top

Current stock prices can be interpreted in several ways. … [The market] may assign a reasonably high probability to a large, permanent GDP decline, which can’t be ruled out even though it didn’t happen in past downturns since, and including, the Great Depression. Finally, given the different nature of this downturn, the old relationships among GDP, profits, and stock prices may no longer hold, or in the future investors will demand higher returns from the market.

Timing the recovery

… One common analysis calculates how many years must pass before the market will return to normal, assuming growth at the historical long-term average rate (10 percent a year). In past recessions, however, the stock market returned from the trough much more quickly, with cumulative returns, over the two years that followed it, of 50 percent to 130 percent (Exhibit 4).5 If this pattern holds in the current downturn, there’s a real danger that companies waiting too long will miss the upside of the rebound. …

Back to top

Back to top

In this simplified example, only timing an investment perfectly, in six months under scenario two, would produce a net present value (NPV) meaningfully better than the one resulting from investing now (Exhibit 5). …

Much uncertainty surrounds the timing of the downturn’s end, but companies waiting for clear evidence of a turnaround may find that they have been recklessly cautious and missed once-in-a-generation opportunities to acquire or invest. Executives considering when to make their next strategic moves can learn much by examining the course of previous downturns—particularly how valuation levels were related to corporate earnings and how valuations and earnings were related to the economy as a whole.

About the Authors

Richard Dobbs is a partner in McKinsey’s Seoul office, and Tim Koller is a partner in the New York office.

The authors would like to thank Bing Cao, Ezra Greenberg, John Horn, and Bin Jiang for their contributions to this article.

This article is not intended to be used as the basis for trading in the shares of any company or for undertaking any other complex or significant financial transaction without consulting appropriate professional advisers. No part of this article may be copied or redistributed in any form without the prior written consent of McKinsey & Company.

Back to top


1As of the end of March 2009, the present downturn has seen five so-called bear market rallies. This downturn could be different from past ones, so there could be more than the earlier maximum of six such rallies. As of March 2009, the market was about 18 months past its peak. The time to trough was 32 months in the 2000 recession, 21 months in the recession of the 1980s, 21 months in the recession of the 1970s, and 35 months in the Great Depression.

2Managers using this approach for actual strategic decisions would need to refine it by country, economic sector, or both, or to reflect the peculiarities of investments such as capital, R&D, or marketing expenditures, as well as competitors’ moves and regulators’ changes.

3We’ve excluded financial institutions from this analysis because their recent profits have been highly volatile—way above average in 2005–06 and way below average in 2007–08. The inclusion of these companies would not change the results but would make it harder to interpret the long-term trends.

4For more on our model of market valuation, see Marc H. Goedhart, Bin Jiang, and Timothy Koller, “The irrational component of your stock price,”, July 2006.

5This analysis looks at share prices relative to the trough. Much more time may elapse before the market reaches the prior peak, partly because it wasn’t based on fundamentals.