Critique: Why Bad Projects Are So Hard To Kill

Critique Harvard Business Review Article

►Why

Bad Projects Are So Hard to Kill Author: Isabelle Royer Originally published 01 February 2003

Teri Abel 2003

Copyright  2003. All rights reserved. No other uses without express author permission.

In her article, “Why Bad Projects Are So Hard to Kill”, Professor Isabelle Royer purports to explain why firms fail to readily abandon projects which she terms are “clearly doomed”. She then offers a prescription for preventing this alleged firm failure. In her analysis, Royer attributes the failure to a “fervent and widespread belief among managers in the inevitability of their projects’ ultimate success”, and to a “desire to believe in something”, the latter of which she calls “collective belief”. In light of this, Royer proposes three firm remedies for preventing this failure. This author regards a critical analysis of Royer’s featured discussion in the Harvard Business Review to prompt the following observations. Neither wreckage from failed projects nor realized returns from successful ones at time (t) constitute forward driving analytics or predictive power at time (0) for those projects’ ultimate value.

Nowhere in the article does Royer actually define what a “clearly doomed” project looks like, vs. a viable one, from any perspective other than hindsight. However descriptive, it is not analytical. Royer touts “strong evidence” of a project’s eventual success as crucial to warranting the continued investment in a project, but does not state what constitutes strong evidence. In effect, Royer mimics the original error of the managers whose leadership is critiqued by advancing no coherent prescription for discerning a “clearly doomed” project from a potentially viable project in advance. Arguably, the reader should expect such a prescription if, as Royer claims, the inevitable doom of “bad” projects is so fundamentally “clear”. Problematically, however, Royer’s discussion begins from an apparent assumption that bad projects are first given. Royer’s description of bad projects that were eventually abandoned is thoroughly similar to a description of what research and development history, for example, tells us have been many successful projects—namely, projects that enjoyed both support and naysayers throughout the organization, and projects that involved the investment of significant firm resources. This can be said, for example, for most MGM film projects—the 1

1 in 20 that are eventually successful, and the 19 in 20 that are not. This observation does not, however, advance MGM toward discerning in advance, which is the 1 and which are the 19. Hence, it leaves unanswered the question of why the 19 were ever funded, which is the article’s indicated aim. Some positive NPV projects that enhance the bottom line can be deemed “bad” or unsuccessful by the market and management.

Alternatively, some completed projects that were clearly profitable and successful by the project’s standards were still deemed bad. This suggests a limitation of Royer’s “bad” descriptor, as the projects were not “doomed”, let alone “clearly doomed”. There is then a category of “bad”

1

According to Written By magazine only when the average slate of American films for a movie production firm was increased to 20 in number was the average slate of films profitable.

2

profitable projects worthy of consideration—bad not because of what they do to the bottom line today, but because of what they have the potential to do later, for example, to diminish a firm’s focus: In a move that was market downgrading, the Coca-Cola company diversified its holdings with its 1982 purchase of Columbia Pictures for $690 million. They profitably unloaded Columbia in 1989 for $1.5 billion. In the short-run, this was, at a return level, a decidedly successful project by most business standards, and one that exceeded Coca-Cola’s own estimates. However, it is unclear whether Royer would, as did the market and eventually Coca-Cola, qualify the Columbia purchase as a bad project, even in hindsight. Private and public firms with dissimilar fiduciary responsibilities should be expected to conduct project management and, hence, abandonment, differently.

Generic project management for a public firm, and, therefore, project abandonment inclinations, can be manifestly different from those of a private firm. Private firms lacking a diversity of shareholders generally exercise latitudes of general and project management unavailable to public firms. This puts the turning of a project’s abandonment partly on the firm’s ownership structure rather than some innate project features. In the case of public firms, considerations are in order about the exchanges on which the shares of those firms are traded and about the applicable universe of real investors in those shares. Differences can exist in, among other things, the listing regulations and transparencies of the exchanges; the cultural aspirations the exchanges can channel; the proportions of institutional and individual investors; and the investing proclivities of disparate investor populations selecting from different arrays of investment vehicles which themselves are not evenly available to the 2

broadest universe of investors. However captured by price are these factors, they are factors which in separate fashion qualitatively influence leadership and project management culture, and theoretically can compete with that of any “collective belief” as Royer described. Different public firms command different cultures of investors; management in turn “plays” differently to different investor groups, effecting different corporate wide project management cultures.

Being public is not even an equalizer in the qualitative assessment of shares and influence upon project management. Different public firms are attached to different histories and agendas and embody different shareholder expectations. Shareholders may uniquely value, for example, a firm’s unconventional beginnings with visionary leadership and unstructured and experimental creative cultures and may expect innovation and more readily reward change—of which all project abandonment can be said to be— whereas the same may not be true for firms conservatively rising from historical journeys through war, for example, that produce

2

For example, a scarcity of issued Chinese treasury bonds relative to the U.S. may marginally affect certain Chinese stock investing as investors seek alternative investment vehicles.

3

reliable industrial grade products. Management is strongly cultural.

Different management and investing populations bound together by different national and economic histories, trajectories, and cultural stories, also bring different risk tolerances and objectives to the management and investing event. For example, a post communist Russian enterprise in the newly charted territory of value maximization, may nurture a very different project management culture from that of any American firm, if it nurtures any at all. Indeed, for a slate of post communist Russian firms, employee and managerial ownership in those firms not only failed to ensure stewardship as generally anticipated under a Western model, but negatively 3

correlated with it. By this example, project management is not even a competency and project abandonment lacks any conscientious drivers. The Enron Corporation’s culture facilitated its famous exercising of certain extremes of project management and abandonment: the culture of creative risk tolerance on the one hand by its leadership and traders, bold adoption of mark-to-market accounting rules, and the periodic axing of some fixed percentage of its talent—a literal project abandonment policy. However, we are led by Royer to believe in a universality of both management and investor behavior, for Lafarge and Essilor in particular, and presumably across publicly traded firms generally, and, hence, in a singular management approach to the project development underpinning those firms’ share values. As all governance is ultimately linked to investor behavior, at least some differences in generic governance ought be considered as possibly emanating from differences in investor behavior and vice versa. Investors are not monolithic. Differences should be expected to affect an individual firm’s project management and hence, its propensity for and culture around project abandonment. As management heterogeneity increases, so too may project analyses and project abandonment potential.

Some proof of management and investor differences is clearly in: it is known, for example, that individual investing behavior in America is statistically different across gender. Governance differentials have been evidenced in American firms with increased gender diversity at a board level and a noteworthy number of high-profile project abandoners and 4

whistleblowers over recent corporate American history were female . Moreover, a phenomenon of collective belief is hardly peculiar to the firm; shareholders themselves are not immune to such a propensity. 3

Contrary to Western expectations, the scheme of official rapid mass privatization of Russia’s largest enterprises in absence of a substantial western-styled regulatory apparatus, led not to stewardship by vested owners and employees but to large scale self dealing. 4 More likely to be "outsiders" to an inner circle of power, the American Bar Association described women as more likely to abide by individual values and potentially break ranks, and less inclined to support more questionable management moves. Three female whistleblowers who dissented on costly high-profile bad projects include Cynthia Cooper of WorldCom, Sherron Watkins of Enron, and Coleen Rowley of the FBI.

4

The fiduciary relationship between management and shareholders then The question is less whether collective belief exists in some group and more whose collective belief is or should be driving in project development.

produces the question of whose collective belief is more driving for some continued investment in a bad project.

5

Royer’s methodology utilizing non-contemporary research does not appear robust. Arguably, it would have been more credible if the research had proceeded contemporaneously. That is, Royer could have conducted real-time studies within several firms and have effectively shadowed the development of different projects—some of which presumably would have become successful, and others of which would not have. In real time, she could have chronicled any critical project development features that supported any project under study. This could have controlled for differences in perception of project histories due to knowledge of ultimate project outcomes. In real time Royers could have then tracked any differences between those processes, and in hindsight could have examined whether the differences and trends in her study ultimately correlated with a project’s outcome in a way that was predictive. Instead, some of Royer’s most critical analysis rests on managerial memories which should be expected to bear influence from the known 6

outcomes of past projects. Her choice of methodology appears to stem from her conviction that “researching events long after the fact can provide perspective that would be absent from contemporary research”. This is an easy truism, but it does not appear particularly useful for the purpose at hand. She is declaring that recounting in settled hindsight that failed projects did indeed fail is either constructive to a contemporary project’s management or otherwise a useful exercise. Utility notwithstanding, the undertaking is not a novel one for a modern competitive firm or its management team and other advisors. The reader is set up to expect something more, in particular, a revelation for how to see a ‘train-wreck’ coming. Royer’s analysis lacks ‘control groups’.

The argument for contemporary research aside, Royer also did not investigate any ‘control groups’ project development processes at either Lafarge or Essilor. Potentially undermining the thrust of her discussion would have been any discovery of overwhelmingly similar features in the development processes of not just successful projects, but “clearly viable” projects, assuming Royer would equally allow for their recognition.

5

Enron stockholders and an array of Wall Street analysts continued to reward Enron management well after management itself had internally conceded overvaluation of its shares. Theoretically, Enron may have used the “permission” of a greatly devalued stock much earlier in its history, to more expeditiously declare and abandon bad projects and accounting practices or change leadership. 6 Royer’s described correction for this bias is a cross checking of the interview data with the written record followed by a repetitive interview process, if needed. Nonetheless, that this “reconciliation” process does not distort a factual record is unclear.

5

By definition, all failed projects were abandoned, i.e., “killed”, if we define abandonment as the cessation of additional resource investment, maintenance, follow up, development and specific seeking of original project objective. This focuses Royer’s argument as one of degree, or the failure to abandon a project within a certain time frame or beyond a certain level of resource expenditure. As critical as this seems to the heart of the discussion, the reader can fairly wish that Royer had spent more time addressing how firms arrive or should arrive at their expenditure constraints---and under what conditions, if any, they should entertain revisions to those constraints over the course of a project’s life. Companies may suffer more from breaching their internal controls than from lacking them outright.

Her knowledge clearly to the contrary, Royer’s remedies assume that firms didn’t already have certain preventative measures in place. Even if one accepts Royer’s argument for the “distorting effect of collective belief”, her advice to “establish an early warning system” is by her own observation, insufficient: neither Essilor nor Lafarge suffered from a lack of such a system. Royer herself makes the case that they suffered because they ignored the systems they had. Disappointingly, she does not suggest how to remedy any tendency on the part of firms to ignore their established internal controls, and this tendency actually may be more prevalent than the absence of internal controls.

Despite a core assertion, companies engage a continual process of abandonment of declared bad projects.

Companies abandon bad projects all the time—some as nearly 7

conscious core operations ; they spin off businesses, sell stakes to larger entities, retire product lines, outsource operations, implement promotion policies, undertake energy conservation initiatives, and fire their CEOs. This raises the questions of exactly what ilk of bad projects defines Royer’s focus, and whether the proportion of “clearly doomed” projects among all bad projects which are not properly abandoned, is particularly significant. Inherent in the author’s analysis is the assumption that bad projects have an archetype or look the same across industries and, hence, likely share a common remedy. This author can argue otherwise. There is likely a range of profiles of bad projects, and this range likely correlates with a range of uniquely tailored internal controls and remedial approaches. At the National Aeronautics and Space Administration (NASA), many core operational projects have a substantially longer development phase than do 8

their counterparts at Procter & Gamble. The same can be said for Pfizer 7

Big pharmaceutical firms, for example, abandon bad projects in such volume that a significant portion of the small pharmaceutical industry creates their value from big pharma’s refuse. 8 This author recognizes that objectives and stakeholders for governmental research and development and private industry are different, making firms of those categories not fully comparable. However, as governmental agencies face resource constraints they are fairly comparable to private firms as failed governmental projects exact costs to taxpayers as do failed private projects exact to shareholders.

6

vs. Universal Pictures. These firms’ projects vary in their consumption of R&D, their marketing, their required overhead, their stakes in creating public goodwill, their frequency and their payback periods. If in fact projects have different archetypes across industries, one can ask not only whether the standards for their development, funding and management are in fact singular, but whether indeed they should be. Royer’s analysis procedes from a small homogeneous sample.

Royer’s conclusions flow from her analysis of only two public French firms. No scientific argument is made for why either this sample size or mix should be statistically significant or otherwise compelling. Therefore, her extrapolations to and recommendations for firms generally do not appear supported. Royer’s description of the phenomenon she calls “collective belief” seems plausible enough. Confidence in industry can instigate confidence. Enron, marketing confidence in its business model and its portfolio of projects, had no problem instigating confidence across the market, enjoying “buy” and “strong buy” recommendations from major Wall Street analysts well near its collapse. What Royer does not detail is exactly how the collective belief surrounding a bad project is inherently different from that which surrounds an eventually successful project. This comparison is where the article would have held more utility. If we only consider the Quaker Oats acquisition of the Snapple brand, we observe that a failed project can have less to do with any germane feature of a particular product or project and more with how it is perceived, packaged, distributed or managed. Under the subsequent management of Triarc which acquired Snapple from Quaker Oats, the brand represented a positive NPV project with a competitive shareholder return and payback period. This suggests that project success or failure can rest on what appear to a firm to be noncritical details. This should render some bad projects especially difficult to discern before making a large investment. Royer’s discussion of group dynamics and the roles of “exit champions” and “project champions” reduces to an argument for robust diversity and objectivity on a managerial team. This is a fair and interesting point of examination, but it is also not a novel consideration for a firm to make, if not consistently implement, by current standards.

7

Conclusion This author believes that those bad projects which are hard to kill are so because the managers who nurture them along with many in management don’t exactly know what they look like, and often enough fail to follow internal controls about how to abandon them expeditiously. With an aim toward utility, an objective to describe why bad projects were deemed bad of their own design might have better advanced the cause for acting in accord with the best evidence for a project’s likely outcome, rather than immediately examining actors in project development. For example, there are basic considerations like whether some new product will be regarded as a truly differentiated product in some market; whether a firm’s research and development supports an incremental change or some game changing breakthrough change and why such thresholds are met by the particular research; whether a brand is strengthened or possibly contradicted by the project even if profitable in the immediate term; and whether an even profitable project is a strategically timely utilization of institutional resources—for example if a firm enjoys the benefit of a longstanding entry barrier for some project, it may rightly command a lower priority than other more opportunistic aims at market growth. Any “good” project is not a good one in isolation or even good indefinitely, but as a function of its temporal place in the scheme of all firm and broader market variables. This moving target of “goodness” or “badness” makes any static and monolithic discussion of failed projects as this author perceives in this article, limited in applicability of the insights the article’s concern aimed to render. More instructive inquiry may lie beyond Royer’s propositions in their entirety and consider the valuation by an organ like the Harvard Business Review put upon Why Bad Projects Are So Hard to Kill. The publishing decision generally marks an occasion for readers’ exercise of not just the benefit of branded identification of the value additive and of new ideas, but of the readiness to confirm a brand as needed, to ensure its meeting of an assumed value proposition. █

8

Bibliography

Black, Bernard, Reinier Kraakman, and Anna Tarassova. “Russian Privatization and Corporate Governance: What Went Wrong?”, Stanford Law Review, Working Paper Number 269a, May 2000: 17-23. Chiang, Harriet. “Women speak up -- big names go down, Female whistle-blowers play by 'outsider' rules”, San Francisco Chronicle, June 17, 2002: A - 1. Written By, the magazine of the Writers Guild of America, West, 2003.

9