 Earnings Growth: The Two Percent DilutionWilliam J. Bernstein and Robert D. Arnott Financial Analysts Journal, 2003, vol. 59, issue 5, 47-55 Abstract: Two important concepts played a key role in the bull market of the 1990s. Both represent fundamental flaws in logic. Both are demonstrably untrue. First, many investors believed that earnings could grow faster than the macroeconomy. In fact, earnings must grow slower than GDP because the growth of existing enterprises contributes only part of GDP growth; the role of entrepreneurial capitalism, the creation of new enterprises, is a key driver of GDP growth, and it does not contribute to the growth in earnings and dividends of existing enterprises. During the 20th century, growth in stock prices and dividends was 2 percent less than underlying macroeconomic growth. Second, many investors believed that stock buybacks would permit earnings to grow faster than GDP. The important metric is not the volume of buybacks, however, but net buybacks—stock buybacks less new share issuance, whether in existing enterprises or through IPOs. We demonstrate, using two methodologies, that during the 20th century, new share issuance in many nations almost always exceeded stock buybacks by an average of 2 percent or more a year. The bull market of the 1990s was built largely on a foundation of two immense misconceptions:With a technology revolution and a “new paradigm” of low payout ratios and internal reinvestment, earnings will grow faster than ever before. Five percent real growth will be easy to achieve.When earnings are not distributed as dividends and not reinvested into stellar growth opportunities, they are distributed back to shareholders in the form of stock buybacks.In fact, neither of these widespread beliefs stands up to historical scrutiny. Since 1800, the economy, as measured by real GDP, has grown a thousandfold, averaging about 3.7 percent a year. The long-term uniformity of economic growth is remarkable; it is both a blessing and a curse. To know that real U.S. GDP doubles every 20 years is reassuring. But this growth is also a dire warning to those predicting rapid acceleration of economic growth from the computer and Internet revolutions.The relatively uniform increase in GDP implies a similar uniformity in the growth of corporate profits—which does, in fact, occur. Except for the Great Depression, during which overall corporate profits briefly disappeared, nominal aggregate corporate earnings have tracked nominal GDP growth, with corporate earnings staying at 8–10 percent of the GDP growth. The trend growth in corporate profits is identical, to within a remarkable 20 bps, to the trend growth in GDP.For 16 countries, with data spanning the 20th century, we compared dividend growth, price growth, and total return with GDP data from the same period. We found that in stable, non-war-torn nations, per share dividend growth was 2.3 percent less than growth in aggregate GDP and 1.1 percent less than growth in per capita GDP. In the war-torn nations, the situation was far worse—per share dividend growth 4.1 percent less than growth in aggregate GDP and 3.3 percent less than growth in per capita GDP.Data for the comprehensive CRSP 1–10 Index from 1926 to June 2002 show that, after adjustment for additions to the index, total U.S. market capitalization grew 2.3 percent faster than the price index. Thus, over the past 76 1/2 years, a 2.3 percent net new issuance of shares took place, which is the equivalent of negative buybacks. Although net buybacks occurred in the 1980s, by the 1990s, buyback activity had once again returned to historical norms.Earnings growth was indeed high during the 1990s. But the persistence of this growth is dubious for three reasons:The market went from trough earnings in the 1990 recession to peak earnings in the 2000 bubble. Measuring growth from trough to peak is meaningless; extrapolating that growth is even worse.Analysts frequently ignored write-offs while increasing their focus on operating earnings. This behavior is acceptable if write-offs are truly “extraordinary items” but not if write-offs become an annual or biannual event, as was commonplace in the 1990s. Furthermore, what are extraordinary items for a single company are entirely ordinary for the economy as a whole.The peak earnings of 1999–2000 consisted of three dubious components. The first was an underrecognition of the impact of stock options, which various Wall Street strategists estimated at 10 percent or more of earnings. The second was pension expense (or pension “earnings”) based on 9–10 percent return assumptions, which were realistic then but are no longer; this factor pumped up earnings by about 15 percent at the peak and 20–30 percent from recent, depressed levels. The third was Enron-style “earnings management,” which various observers have estimated at 5–10 percent of the peak earnings.In summary, in a dynamic, free-market economy, considerable capital is consumed funding new ventures. For this reason, per share growth of prices, earnings, and dividends will lag aggregate macroeconomic growth by an amount equal to the net issuance of new shares. In peaceful, stable societies, this gap appears to be about 2 percent a year. In war-torn nations, this gap is considerably larger. Although these nations' economies can recover relatively rapidly, the high degree of recapitalization that is required savages shareholders. Date: 2003 Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:taf:ufajxx:v:59:y:2003:i:5:p:47-55 Ordering information: This journal article can be ordered from Access Statistics for this article Financial Analysts Journal is currently edited by Maryann Dupes More articles in Financial Analysts Journal from Taylor & Francis Journals |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.