Human Resource Management Professor Monika Hamori’s recent report published in the Harvard Business Review in July 2010 (http://bit.ly/byC0ZY) casts doubt on the conventional wisdom that moving jobs can accelerate your promotion through the ranks. In particular she argues this casts doubt on the Boundaryless Career idea (see David Winter’s piece on the Careers in Theory Blog). Indeed Profesor Hamori identifies 4 “myths” associated with advancement. They are a) Job Hoppers prosper; b) A move should always be up c) Big fish swim in big ponds and d) Career and Industry switchers are penalized. On the face of it, such findings appear to cast doubt on the ideas behind the Boundaryless Career. Lets take a look at each of these arguments in their turn, because clearly such provocative conclusions demand closer consideration.
The first thing to say is that Professor Hamori is not some opinionated commentator hollering from the sidelines. She has an impressive evidence-base upon which she draws her conclusions. Specifically, she considered 14,000 career histories of non CEO executives in four sectors of the financial services industry. These records were stored by a large multi-national search firm. She also looked at the career histories of CEOs of Financial Times top 500 European companies and Standard and Poor’s top 500 US firms. In addition she collected interview data from a relatively small number of executive recruiters (45) and Business School alumni (20).
The Job Hopper Fallacy
Hamori’s case for rejecting the notion that job hoppers prosper faster rests on several lines of evidence.
- Firstly she reports that her CEOs worked for three employers on average throughout their careers (with 25% having been with the same company throughout their careers).
- Secondly amongst her 14,000 non CEO executives, she reported that inside moves produced a “considerably” higher percentage and a faster pace of promotions compared to external promotions.
- Thirdly, Hamori provides a couple of selected quotes from her interviews with recruiters to support her speculation that companies prefer to see “stability” in their executives’ career paths.
A closer look at the numbers
On the face of what is presented in the Harvard Business Review Article, it is difficult to make any precise arguments about the interpretation of the data, because too little is presented to do this (this is simply a reflection of the demands of writing for a non scientific audience and not a shortcoming of the work upon which the article is based). However I did locate an earlier paper by Hamori and Kakarika (2009, Human Resources Management, June) that reports the CEO data in great detail. This helps to clarify some of the findings.
Firstly according to Hamori & Kakarika (2009) “We found that CEOs who have spent a higher percentage of their career with the organization they currently lead (% of career spent with organization…or have spent their entire professional career with the organization… take almost one and a half years less time [my italics] to be appointed to the CEO position of a large organization, other factors being equal (Models 2 and 3). In addition…specifically, for each additional employer the predicted time to the top will increase by more than half a year, other factors being equal….On average, lifetime CEOs reached their current position in 23.1 years, while those who had six or more employers took 26.75 years to get to the top.”
For example, the statement that “CEOs who have spent a higher percentage of their career with the organization they currently lead… take almost one and a half years less time to be appointed to the CEO position of a large organization” does not specify “one and a half years less time compared to who?”. In fact, looking at the units in which % of career is assessed, it turns out that one and a half years is the difference between someone who has spent 0% of their career with the organization compared to someone who has spent all their career with the org, which is a fairly extreme comparison. (As an aside, note how this relationship doesn’t quite make sense conceptually. Effectively, it says that someone who has spent all their career with the organization that they lead will have gotten there more quickly than someone who has spent none of their time with the organization that they lead?!).
Also, when you look at the size of the correlations, they are less than .10 for two of the three IVs, which is the benchmark below which Cohen says things start to get trivial.
An alternative interpretation of the data: Job hopping is good for your career
So if we compare the most rusted on CEOs with the most fickle regular movers in the sample, staying put provides a time advantage of at the very most 13.64% over about quarter of a century. Whilst these figures provide no support on the face of it for those who advocate moving to enhance a career, the benefits of staying put are hardly so large that it is self-evident that staying put is the best strategy. Does getting to the top three years earlier real mean very much over such a period of time? When set against some of the plausible benefits of moving around such as greater diversity of experience, and perhaps a richer more storied personal history, three years seems a small price to pay. Indeed it amounts to little more than accrued sick leave and a few other days off every year.
A more serious concern is whether the conclusion that not moving is better for advancement can be substantiated. This is based on correlational data showing a negative relationship between the number of moves and the time taken to make CEO. It makes the assumption that those who leave had an equal shot at the top job compared to their colleagues who remain. This is a very dubious assumption to make.
Consider this: suppose a town has an Easter egg shortage, such that each of the seven shops have only one egg for sale. There are seven shoppers who turn up at the first store. Only one is going to get the egg. The remaining six walk over to the next store, where five miss out. These five move on to the next store. Eventually we have a pattern that shows the person who did the most walking (the hapless customer who had to go to all seven stores before securing their egg) also took the longest time to get their eggs. Moving stores and time to get eggs shows exactly the same relationship as moving organisations and time to get to CEO in Hamori and Kakarika’s work. However nobody would advise the customers who have missed out to hang around in the first store because they might never get an egg.
There are far more executives in any one company than there are CEO positions, and if the company has good succession planning in place, then there will be more executives suitable and capable of being CEO than there are CEO roles (one). So even putting aside the very important fact that not all executives are equally capable or suitable of being CEO, the data does not provide any evidence to support the author’s conclusions that staying within the organisation is a good move for anyone other than the person who ultimately makes it to CEO. Now you could argue in the Easter Egg example, that those shoppers who missed out in the first store and move on actually have to join other shoppers already milling around in the next store waiting for the egg to be tossed into the crowd. However, this is still a better option than staying in the first store that has run out of eggs. Indeed the person who moves will secure an egg faster than had they not moved.
The problem lies in the notion of all other things being equal. If the selection process of CEOs was essentially random and also regular within each company, then all other things are cancelled out (are equal), and so you may as well stay and take your chances. However this is a huge and unjustifiable assumption to make in any top 500 organisation that all have very explicit and structured processes for identifying and developing talent. It illustrates a type of “ecological fallacy”, where a relationship observed at the between-person level (i.e., “people who move jobs more often take longer to become CEO than people who move less often”) is used to make an inference at the within-person level (i.e., “for a given person, moving jobs will increase the time to become CEO”). As the two levels (between-person and within-person) are statistically and conceptually independent of each other, findings from one level do not necessarily generalise to the other level.
Consequently, it is entirely possible for the negative relationship to exist at the between-person level while at the same time for job hopping to be beneficial for some (or even all) people. The important point is that the job hopper fallacy is a within-person (or individual level) phenomenon that isn’t necessarily going to be adequately tested using a between-person analysis. At the individual level, which is where the Career Counsellors work (the ones that are explicitly singled out for perpetuating “myths” about advancement and movement), advising an executive who has been overlooked for a key promotion that precedent indicates is the pathway to the CEO position to remain with the company rather than looking elsewhere makes little sense. The research as presented is not powerful enough to pick up such career reversals or plateaus.
So if the executive has been identified as the “most likely to”, then advising them to stay with the organisation makes sense. Indeed the advice may well be superfluous because that executive is far more likely to enjoy a range of benefits, bonuses, perks, recognition, feedback and training that all serve to enhance engagement. The moving to enhance your career may be a myth for that very select group. But for everyone else, given that everything else is not equal, their chances within their current company are not equal to everyone else’s, moving may be the quickest route to a CEO role, even if it takes longer than those who remain because they’ve already been identified as going places.
Staying with the company does not cause a person to become a CEO quicker, it is merely associated with that promotion for those who made it. Clearly if one included all the executives who didn’t move irrespective of whether they made it to CEO or not, the correlation between not moving and time to make CEO would show a very strong positive relationship between time served and time waiting to become CEO. One final point is that the fact that CEOs worked for three employers on average throughout their careers doesn’t really tell us anything about the usefulness of job hopping unless we also know how many employers the people who didn’t make it to CEO worked for. For all we know, the latter group may have only worked for one or two employers, which would support the job hopping idea.
A limited view of promotion and career advancement
In Hamori’s work, promotion (other than attaining the CEO role) is defined as “a better title with more responsibility or propelled the executive to a larger firm”. This is a very narrow definition. It does not, for instance take into account remuneration or other benefits and conditions. It equates the size of a firm with managerial complexity which may be an over-simplification. Managing a large group of people in a well established and otherwise well run and successful organisation may be a lot simpler than managing a small dysfunctional team in a complex, competitive and rapidly changing environment. It also doesn’t take into account a whole swathe of work rewards such as autonomy, altruism, quality of co-workers, surroundings, skill development, freedom, work-life balance, and lifestyle factors to name just a few.
It doesn’t take into account job satisfaction either. People often move jobs because they are frustrated. The ambitious do so because they often perceive their ambitions are being frustrated. It is questionable whether advising such people to remain with their employer is going to result in positive career outcomes in all or perhaps even most cases.
The focus and privileging Fortune Admired and top 100/ 500 companies may well reflect the client-base of the Executive search firm that provided a lot of interviews and career histories for analysis. This may promote the notion that only moves to the higher echelons of these lists can be deemed promotions. If broader conceptions of advancement that go beyond the narrow confines of market indexes are considered, what would be the impact upon the data? A move should be up – is this really a myth? This brings me to the “second myth” – which is that “A move should be up”. This myth came as a surprise to me, because I am not sure how many credible authorities are pushing such a message. Indeed I would argue anecdotally that most credible careers professionals promote privately and publicly the view that “side-ways” moves often provide opportunities, and indeed “moves-down” can provide, as Norm Amundson would say the “backswing” momentum to propel one forward. So this myth seems to be of the straw-man variety.
Performance is overlooked
The study has nothing to say about how the individual CEO’s actually perform. Granted this is a complex and contentious area, however it is important to have a sense of relative performance, because there is a possibility that those who move perform better than those who stay. There is no evidence in their data for this proposition, but supposing it were true, where does that leave the conclusion that moving for advancement is a “myth”? There are several possible confounds that could account for Hamori & Kakarika’s (2009) findings not least of which are ability and personality (e.g., highly able, conscientious and emotionally stable individuals are more likely to stay at a firm and also more likely to become CEO). Another big confound relates to people in the sample who started their own corporations early on (see H&K, 2009, p.361 under Dependent Variable).
For example, a person who starts his own corporation out of grad school would score 0 for time taken to become CEO (i.e., the minimum score) and would score 100% for % of career with organisation (i.e., the maximum score). Obviously this greatly inflates the observed negative relationship!
Big fish and big ponds
Here Hamori uses lists like the Fortune’s most Admired lists to judge whether executives are moving from bigger name to smaller name companies. It would be intriguing to get a close look at this data, because such lists have very significant volatility and turnover. The so called “stumble rate” of companies knocked off their most admired perches from one year the next is quite high (49%), and even in the highest echelons of the top 50 all stars, only 17 companies remain on the list that appeared on the initial one in 2001. Consequently with all this volatility, trying to say anything definitive about moves from big names to smaller names seems fraught with difficulty. Afterall Apple’s market value was $7.09 billion in 2001 (when the Fortune list began), and Microsoft’s was $332.73 billion. So leaving Microsoft for Apple presumably was seen as a bad move. In 2010 the Apple’s capitalization was $225.98 billion to Microsoft’s $225.32. Things change and rapidly.
The arguments are supported with appeals to interviews with executive search recruiters, who reinforce the view that too many moves are a bad thing. They claim they and their clients prefer to see only a few moves perhaps interspersed with periods of stability (e.g. An “eight-year run”).
There is nothing here to support the claim that moving advances your career is a myth. Hamori’s work is interesting, well presented and thoroughly analysed. It makes a provocative contribution to our understanding of career advancement. However her criticism of the idea that movement can lead to success as being a “myth” is premature, and is not supported by the data she presents. There are too many other variables that plausibly might be at play here that are simply not considered within the narrow definitions of career success or within the dataset.
The size of the effects she reports in practical terms seem underwhelming over the period of time it takes to become a CEO. Another way of looking at this data would be to say, if your path to the CEO route looks to be blocked in one company then moving a couple of times may only delay you by about 12 – 18 months in the worst case scenario that you would have made it to CEO had you stayed put. If, however, you wouldn’t have made it had you stayed put, then moving has probably got you to the CEO role faster than staying put. Given there can only be one CEO (in nearly all companies), then my alternative scenario is more likely to apply. In other words this data can just as easily be interpreted to draw precisely the opposite conclusion.
There are valid criticisms that can be levelled at the Boundaryless career idea, for instance Rodrigues and Guest (2010) review evidence suggesting that moving jobs is not on the increase to the degree that some commentators claim but has always been part of the scene. They conclude that “what we seem to be witnessing is not the demise but rather a redefinition, a growing complexity, and a more subjective perspective on career boundaries (Heracleous, 2004).” pp 1170. I’d agree.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New Jersey: Lawrence Erlbaum.
Hamori, M & Kakarika, M. (2009) EXTERNAL LABOR MARKET STRATEGY AND CAREER SUCCESS: CEO CAREERS IN EUROPE AND THE UNITED STATES Human Resource Management, May–June 2009, Vol. 48, No. 3, Pp. 355– 378
Heracleous, L. (2004). Boundaries in the study of organaization. Human Relations 57(1): 95-103.
Rodgrigues, R.A. & Gust, D. (2010). Have Careers become boundaryless. Human Relations, 63, 1157. DOI: 10.1177/018726709354344
I’d like to thank my colleague Dr Amirali Minbashian for his feedback and comments that have helped to shape my thoughts for this piece. I’d also like to thank David Winter from the Careers in Theory Blog http://bit.ly/aPBGq9 for agreeing to us posting our thoughts simultaneously on The Factory Blog and the Careers in Theory Blog.