«Stephan H. Späthe * First draft: August, 2012 This draft: July, 2013 Abstract This clinical paper gives guidance on how to correctly compute and ...»
How to Evaluate the Share Price Performance during CEO Tenure:
The Case of Josef Ackermann’s Stewardship at Deutsche Bank
Stephan H. Späthe *
First draft: August, 2012
This draft: July, 2013
This clinical paper gives guidance on how to correctly compute and evaluate the share price
performance of a corporation during a specific CEO tenure. Although the emphasis is on
Deutsche Bank’s share price performance between 2002 and 2012, a lot of general questions relevant for this kind of assessment are tackled as well. For instance, it is shown that currency adjustments are essential when comparing global companies if their shares are listed in different currency areas. In addition, the differences of dollar-weighted versus time-weighted returns for peer groups during consecutive observation periods are discussed.
At Deutsche Bank’s AGM at the end of May 2012, the CEO tenure of Josef Ackermann officially ended after ten years in office. During his stewardship, Deutsche Bank established itself in the bulge bracket of investment banking and was one of the few global banks which survived the financial crisis without any direct government support. Whether his regime was successful for shareholders is controversial. It is shown that in spite of the fact that Deutsche Bank’s original share price performance during Ackermann’s stewardship has been negative by −50 %; it significantly outperformed other German financial institutions, international sector indices or hand-selected investment banking peer groups. However, it is not clear if this relative outperformance is solely attributable to Ackermann’s and his board members’ management capabilities or perhaps a result of implicit guarantees by the German government.
Moreover, share price performance figures published by Deutsche Bank are reviewed. It turns out that some figures overstate the real performance. The same is true for e.g. Morgan Stanley’s performance graph in its latest 10-K report. This finally raises the question, how reliable share price performance figures presented by companies are. This is important for the ongoing pay-for-performance discussion because already the underlying data of share price performance could be biased.
JEL classifications: G21, M49, N24 Share Price Performance, Total Shareholder Return, Performance Evaluation,
Performance Graph, Price Index, Return Index, Dollar-weighting, Deutsche Bank, Investor Relations Data Availability: Data used in this study are publicly available from sources identified in the paper.
* I would like to thank Ralf Stromeyer, Sunny Kapoor and Christina Wolff for their helpful comments and suggestions. The responsibility for all remaining errors belongs to nobody but me.
Center for Financial Studies (CFS), Goethe University, Grüneburgplatz 1, 60323 Frankfurt am Main, Germany E-mail: email@example.com, Phone: +49 (69) 798-30080 1 Introduction At Deutsche Bank’s annual general Meeting (AGM) at the end of May 2012, the CEO tenure of Josef Ackermann officially ended after ten years in office. During his stewardship, Deutsche Bank established itself in the bulge bracket of investment banking and was one of the few global banks which survived the financial crisis without any direct government support. Whether his regime was successful from a shareholder perspective is controversial. In the months and weeks before the AMG, almost every German newspaper published some kind of commentary on the Ackermann decade, including an evaluation of Deutsche Bank’s share price performance. Although they all covered almost the same observation period, they came to very different results. The reason for this is a broad mix of measures for share price performance, which mostly have been used or interpreted incorrectly.1 The first objective of this clinical study is therefore to provide a comprehensive evaluation, which covers different observation periods, performance measures and benchmarks. In addition, the impact of currency adjustments and different computation methods are shown. My overall aim is to answer the question, whether Ackermann’s stewardship of Deutsche Bank was valuable for shareholders or not.2 This issue is also important for the current compensation debate of senior executives since the linkage of executive compensation to share price performance is nowadays the standard in most corporations. This debate has attracted a lot of public attention and as a consequence Deutsche Bank has introduced at the end of October 2012 an external and independent compensation panel, which should review the structure and governance of its compensation practice.3 From a theoretical point of
view the following quote summarizes the benefit of a strong linkage:
“The separation of ownership and control in corporations is a central feature of modern economies. The largely unobserved choice of actions by a firm's managers can have a major impact on the wealth of its shareholders. The literature on principal-agent theory suggests that the primary means for shareholders to ensure that a manager takes optimal actions is to tie her pay to the performance of her firm.”4 Although the shareholder value approach is criticized for its focus on short-term earnings at the expense of long-term value creation, share price performance is the ultimate measure to evaluate corporate performance from an investor perspective. In the short-run it could be driven by expectation and rumors but in the long-run they have to materialize. To capture common exogenous shocks, See e.g. DER SPIEGEL (April 7, 2012, p. 77), which highlights the 29 % fall of Deutsche Bank’s market capitalization without adjusting for its latest capital increase of € 10.2 billion; Handelsblatt (February 3, 2012, p. 67) compares Deutsche Bank with international peers without any currency adjustment or Hackhausen and Panster (2001), which compare Deutsche Bank’s price index with the DAX’s return index.
I am aware that Josef Ackermann did not manage Deutsche Bank alone and that other board members and numerous senior executives had a stake in the performance of the bank, too. However, he was the Spokesman and lateron CEO, who had the final responsibility and power of decision. Hence he was the decisive person.
See Deutsche Bank AG (2012d) and for the initial announcement Deutsche Bank AG (2012c).
Aggarwal and Samwick (1999, p. 1999) insulate management from common risk and to obtain an informative measure to assess management performance, a lot of companies use relative performance evaluation (RPE) by benchmarking own results relative to competitors. Recent studies of e.g. De Angelis and Grinstein (2011, p. 17), Gong et al. (2011, p. 1014) and Carter et al. (2009, p. 280) find that between 75 and 91 % of compensation packages are linked to relative share price performance.5 However, other studies found evidence that management often uses its discretion and is “gaming in the peer selection process” to promote its own relative performance.6 In the majority of cases a downward bias of the benchmark could be identified.
But until now nobody has ever checked if corporations correctly compute their share price performance in the first place and if they use appropriate concepts. Hence this examination is the second objective of this clinical paper.
To summarize my analyses, I find that Deutsche Bank’s original share price performance during Josef Ackermann’s stewardship has been negative by −50 % or rather −35 % including reinvested dividends. The decisive observation period had been January 30, 2002 until March 7, 2012 (footprint version) and the performance varies only slightly for different observation periods. Thus, from a European shareholder value perspective his tenure was a disappointment. In a relative comparison, the evaluation depends quite a bit on the benchmark, the respective currency denomination and the observation period. However, all broad equity benchmarks like the DAX, EURO STOXX 50, S&P 500 or MSCI World performed better, by at least 30 up to 65 percentage points over the ten years.7 More appropriate would however be to compare Deutsche Bank with specific industry indices or individual peer groups, since the whole financial sector suffered due to the financial crisis. When doing so Deutsche Bank outperforms all German competitors as well as sector indices and most international peers. Hence, Josef Ackermann’s stewardship rated on a relative performance evaluation would be quite favorable. However, recent studies on implicit guarantees by governments in the aftermath of the financial crisis deliver a possible alternative explanation and hence question the management achievements.
I contribute to the literature in a number of ways. First, I show that a clear compatibility between price indices and return indices for individual shares and especially indices must be guaranteed because otherwise the results overestimate or underestimate the performance. Best practice should be comparisons based on return indices because they include dividends and hence are independent of different payout policies. Second, a correct observation period is essential and multiple subperiods increase the complexity of correct performance computation. Third, relative performance evaluation is a useful way to absorb negative as well as positive external shocks and to measure firm specific share price performance. Individual peer groups can be useful but leave management with great discretion in A study by Perry and Zenner (2001, p. 466) reports contrasting results with share price only in the fifth place with approximately 20 %.
Faulkender and Yang (2010, p. 266) See row ‘Return Index in EUR‘ of Table 3-2 as well as Table 3-3.
the peer group composition. Consequently, only published and well known industry indices should be used as benchmarks. Fourth, I emphasize the importance of currency adjustments when comparing international peers because foreign exchange trends influence performance comparison quite a lot.
Finally, I demonstrate that from a shareholder perspective dollar-weighted returns instead of time-weighted returns are the correct performance measure although their computation is more complicated.
This is the first study which independently recomputed performance data provided by a company and thus reviews it. In the case of Deutsche Bank it turns out that an important graph from the latest annual general meeting as well as some comparisons in the share price section of the annual report are deceptive. All are in favor of the company and hence reinforce the argument of Lewellen et al. (1996,
“… performance matters to senior corporate executives. If it didn't, there would be no reason for firms to report downward-biased performance benchmarks in their proxy statements in order to attempt to create favorable comparisons with their own firms' performances.” However, as this clinical study shows, companies not only wrongly compare return indices of their own share price including dividend payments with return indices of benchmarks without dividend payments, they also influence the computation of share price performance. For unsophisticated readers like the average private shareholders these shortcomings are hard to detect undermine and the basic pay-for-performance principle.
1.1 Research Question At the first glance the following research question of this clinical study may look quite simple, but it turns out that some of obstacles have to be overcome. Especially press article, but investor relation departments of corporations as well tend to disregard some problems or mix approaches, which cannot
be combined. Therefore I want to provide a structured approach which answers the problem:
How to compute and first and foremost evalute share price performance?
In addition the following questions should be solved within this framework too:
What is from a shareholder perspective the best concept to measure share price performance?
How to define the appropriate observation period?
What are good benchmarks to evalute the computed share price performance?
How to deal with share prices in different currencies?
How to compute returns of peer groups given that the overall observation period is divided into several periods of time?
The general research question of this clinical paper applied specifically to Deutsche Bank would be as
Was Deutsche Bank’s share price performance during the stewardship Josef Ackermann superior, average or inferior?
1.2 Related Literature The focus of this clinical paper is the correct computation of share price performance and more importantly its appropriate evaluation relative to different benchmarks. The computational basics are first of all related to classical textbooks in finance such as e.g. Sharpe et al. (1999), Bodie et al. (2011) or Albrecht and Maurer (2008) as a German one. Although the fundamentals and distinctions of original share price performance, price index performance and return index performance are straightforward, they are often mixed or not clearly labeled in newspaper articles as well as annual reports. To my knowledge no research paper has ever analyzed this topic. Therefore this clinical paper highlights that even in publications of a large corporation such ambiguities exist and it seems that these mistakes are not random. Hence, findings of studies that by like Bannister and Newman (2006, p. 37), who used share price performance figures published by corporations and did not recompute them, could be distorted.
A second line of research related to this clinical paper confirms this impression management hypothesis and these are research papers, which analyze financial graphs and performance measures.