Tuesday, December 26, 2017

Teams at Work and Lives at Stake: How to Handle Fast-Paced Complexity

There is a lot of research on how teams make disasters happen, and the answer is clear: teams use cues to make sense of the situation, and disasters happen when sensemaking differs from reality. That’s useful to know, but we would also like to know how it can be prevented. We know that expertise and experience do not help. Experienced commercial pilots, space shuttle subcontractor engineers, chemical plant operators, and fighter pilots have all been studied and found to do faulty sensemaking.  The examples I just gave have led to a total of 4,000 confirmed deaths and more than 10,000 likely deaths.

Finally, an article in Administrative Science Quarterly by Marlys Christianson has some answers. She studied how medical teams went through an emergency room training procedure – treating a young asthma patient with increasing breathing failure – in a simulation designed to invite incorrect sensemaking in the beginning, so they would need to recover later. Fortunately, in simulations the patients are not real, because one quarter of them would have died. Even among the teams that managed to identify and correct the problem (replacing a piece of broken equipment), the speed of doing so varied a lot, so thanks to this research we now know a lot more about how sensemaking can recover.

Teams are in organizations for doing work, not for solving puzzles.  Whenever a situation involves a puzzle that needs to be solved, such as faulty sensemaking that needs to be corrected, the regular work done by the team takes effort and attention away from the correction. This means that cues that may look obvious to someone outside the team are not at all clear to team members who are focused on the regular work and who do this work premised on their sensemaking. In an emergency room, the team will look for cues to how the patient is doing, but they spend much of their time treating the patient. Treating and observing clues are related, but they compete for time.

This means that emergency room teams can solve puzzles only if they manage two trajectories at once – the regular treatment and the interpretation of cues from the patient’s condition. The interpretation trajectory is how sensemaking is updated, and it is complex because it moves from noticing cues that suggest something is wrong, to interpreting them to indicate what the problem is, to acting to check the interpretation. Usually the actions involve changing the treatment, so treatment and interpretation need to be in sync. The trajectory management can fail in multiple places. For example, the treatment takes too much time so cues are not interpreted, or the treatment is based on current sensemaking so changing it to check interpretation does not make sense.

The emergency room teams had a sensemaking problem because the simulation was designed to involve treatment equipment that did not work correctly, so the usual sensemaking (“our equipment works, so all problems can be found in the patient”) was faulty. Similar sensemaking problems are found in many places. In the Black Hawk shooting incident, the fighter pilots saw helicopters without correct friend–foe identification signals and concluded they would be hostile because friendlies signal who they are. Any cues they could see were drowned out by the tasks of flying the aircraft low in mountainous terrain, keeping alert for possible threats, and going through a modified foe identification and engagement procedure while communicating with each other.

Trajectory management can easily fail, with tragic consequences. Now that we know more about the differences between teams that succeed and teams that fail, we may be able to work to make teamwork more reliable, especially when lives are at stake.



On a personal note, I’ve experienced the benefits of the sort of updated sensemaking described in the article.  When I was in the emergency room after an accident, the team scanned me to look for internal bleeding based on their experience of how body folding from being hit by a car while riding a motorcycle can break blood vessels. They found none. The cue of falling blood pressure after closing the external wounds made them re-scan over a broader range, and they found the broken vessel and fixed it. I am alive, thanks to the team’s updated sensemaking.

Wednesday, December 20, 2017

A Paradox of Innovation: Those Who Do It May Be Ignored

We are supposed to like innovations. They drive the world forward, with effects that range from the pleasant (like the camera on your phone) to the vital (like portable ultrasound in developing nations). In fact, many of the heroes in business are known because of their innovations. A classic example is Steve Jobs launching the multi-function iPhone, which relied on knowledge of music storage and playoff, as well as internet connectivity, that previously had not been part of mobile phone technology. This is one of the two classical stories on how to innovate: combine existing knowledge in new ways, or create completely new knowledge.

The only problem with the iPhone story is that it makes us think the world rewards innovation and that firms doing it get Apple-like fame and fortunes. That happens to be the exception. A research paper by Matt Theeke, Francisco Polidoro, and James Fredrickson in Administrative Science Quarterly has shown that firms using new kinds of knowledge for making innovations face a surprising form of risk: they may end up getting ignored.

The details of this research help us see exactly what happens. All kinds of firms want stock brokerage firms to issue analyst reports on them, because that means investors will pay attention to them, which helps them gain financing. This is especially important for firms that rely on innovations, because making innovations means paying money now to get money later, which is exactly what financing is used for. In fact, there are entire industries that are so dependent on innovations that analyst reports are essential. Theeke, Polidoro, and Fredrickson studied medical devices, which is a good example of an innovation-driven industry. Brokerage firms covering that industry need to understand research and knowledge use, because otherwise they cannot estimate future profits well.

So what is the problem?  Well, the brokerage firms have expertise in the conventional use of knowledge, which means that use of new knowledge – innovative use of knowledge – is something they understand less well. As a result, firms incorporating new knowledge are more likely to be ignored, as brokerages drop them from their coverage. The newer the knowledge is, and the more expertise the brokerage firm has in covering other firms in the industry using conventional knowledge, the worse the situation is. Just as expertise makes some firms rigid in their knowledge use, it makes brokerage firms rigid in their knowledge valuation.

So our tales of heroic unconventional innovators are good examples of exceptions, because business rewards convention. Does that mean it is better to follow convention and just make minor improvements? Not really, because easier access to financing is very different from more successful product launches. It just means that firms planning to use new knowledge in making innovations should check their bank accounts first, because they may have to pay the cost themselves.

Friday, December 15, 2017

Open Innovation and Closed Minds: Why NASA Used Open Innovation Sometimes but Not Always

Open innovation is heralded as a way to advance technology and product innovation quickly and cheaply. It is modeled on the open source software movement, which is based on computer programmers donating their time to build software components, check their own work, check others’ work, and correct mistakes. Among the famous software suites made through open source, Linux is a computer operating system that is used in everything from cellular phones to web servers, and is often involved when you are retrieving and reading blog posts like this one. Open innovation extends this model to innovations outside computer programming by organizations posting problems that anyone interested can help solve.


The idea is to use volunteer efforts to get innovations for free (almost a Dire Straits lyric), which sounds like a good deal. Unfortunately, this has proven difficult for many organizations, and research in Administrative Science Quarterly by Hila Lifshitz-Assaf has found out why. Her careful study looks at an open innovation initiative in a very innovative high-tech organization: NASA. In 2009, NASA tried an open innovation experiment that led to some speedy, inexpensive, and impressive solutions. But its relationship with open innovation since then has been inconsistent, with some NASA professionals using it to great success and some not. Why the difference?

In a word, the difference is identity. Innovations are typically done by highly educated people who are trained to follow careful processes specific to their organization and to their scientific and technological specialization. These people have a professional identity built around their unique skills as problem solvers for the organization. For people with such an identity, what does it feel like to have amateurs solve problems instead of them? Open innovation draws much of its strength from individuals who may lack formal education, don’t follow the predefined process, and aren’t even employees of the organization. Naturally there is an inherent conflict between the insiders and the open innovation use of outsiders, and some insiders are tempted to seal the organization off from the outside sources of innovations.

Why did some parts of NASA embrace open innovation? Again the answer is identity. Those who could redefine their professional identity to be a solution seeker, not a problem solver, became adept users of open innovation. For a solution seeker, the existence of a solution is what matters – not who made it, and not how it was made. It is a completely different way of thinking of oneself and of solving problems.

The division between problem solvers and solution seekers resulted in NASA professionals adopting various approaches to the open innovation initiatives advocated by their leadership. Problem solvers maintained boundaries, either explicitly or through the pretense of openness but actual closure. That way they could maintain their focus on their individual efforts and internal innovations. Solution seekers looked for outside solutions, sometimes simply embracing externally developed solutions, and sometimes adapting external solutions so that the final solution became a mixture of outside and inside effort. Problem solvers may hold tight to their identity, but open innovation is sure to continue gaining ground. “Get your innovations for nothing, get your praise for free” is an appealing tune.  

Lifshitz-Assaf, Hila.2017. "Dismantling Knowledge Boundaries at NASA: The Critical Role of Professional Identity in Open Innovation." Administrative Science Quarterly, Forthcoming.

Thursday, December 7, 2017

Minimally Invasive Investment: How Ventures Interact with Investment Partners

Of the many kinds of new businesses that are created every year, researchers and policy makers have been most interested in the ones that pursue innovative technologies and market opportunities. They are the ones with the greatest impact on the world, and much effort has gone into studying what makes them innovative. But let’s take a broader view on this question. What if firms around them also affect their innovation – specifically their investment partners, firms that supply them with money and expect returns from their innovations? The answer would be especially interesting if different investment partners had different effects. And as it turns out, they do.

An article in Administrative Science Quarterly by Emily Cox Pahnke, Riitta Katila, and Kathleen Eisenhardt shows how this happens. The important difference is how each organization doing investment has people trained in specific ways, and adhering to specific norms, as a result of their recruitment and career histories. For innovative ventures, venture capital (VC) firms are special because they invest in potential – in firms that could easily fail, and usually do, but have very significant profits when they succeed. VCs are different from sources of corporate venture capital (CVC), which are investment arms of corporations. They are special because they invest in fit to the corporate strategy – firms that develop products and technologies that match so well that they can become integrated into the corporation or at least use its resources well. Then there is the third kind of special investor—government agencies. They are special because they are interested in science and technology with significant societal impact.

Pahnke, Katila, and Eisenhardt looked at what happened to ventures after receiving funding from each of these sources by examining a specific high-technology industry—medical device firms developing products for minimally invasive surgery. The results were clear. The commercially oriented VC investors were good at exactly that. Their firms launched more products after the investment but did not get more patents approved after the investment. The strategically oriented CVC partners were not good at anything that could be measured independently of their strategy. No more patents were approved, and no more products were launched. That does not mean they weren’t good investors, because they could well have selected and improved the strategic fit of their firms. The government was not good at product development, having no effect there, and appeared to harm patenting, with a reduction in patents after entering the investment. That seems bad—but it is actually unclear, because government may be interested exactly in the type of scientific development that is useful for society but hard to turn into patents that give commercial benefit.

So what is going on here? We can tell that one type of investment partner – VCs – has clear and measurable goals and is good at accomplishing them. For CVC and governments, it is harder to tell. Either they are not doing well, or their goals are not exactly what we can measure. Looks like an interesting topic for further research, because each of these investment partners places big bets on our future.

Pahnke, Emily Cox, Riitta Katila, and Kathleen M Eisenhardt.2015. "Who takes you to the dance? How partners’ institutional logics influence innovation in young firms." Administrative Science Quarterly 60(4):596-633.

Friday, December 1, 2017

When Nanotechnology Shrank: How Communities Police the Boundaries of Their Field

It is ironic that I should write a post on how communities of science police the boundaries of their field shortly after writing a blog post on how nations hurt themselves by policing their boundaries. But a paper in Administrative Science Quarterly by Stine Grodal has exactly that theme and some important conclusions. Plus, it is about nanotechnology, something we have heard about and think will shape the future of the world but don’t understand well.

In fact, what I just wrote echoes the start of the nanotechnology field. It was a term coined and advocated by futurists, it was and still is claimed to be a source of advances in science, technology, and business that will change individual lives and society, and it has been redefined many times. The redefinitions of the field are important because they are partly a consequence of these futurists and others with an interest in the term, especially the government, grappling with the question of how best to define the boundaries of the nanotechnology field so that it attracts the right kind of support from others and makes the kind of advances that are desired.

Nanotechnology became a very successful field, in part because of government intervention in the traditional way: giving out money to those engaged in research on nanotechnology. The other source of success was interest in nanotechnology companies from venture capitalists, who expressed their interest the same way: they provided money. This initiated an identity crisis because it soon became clear to the futurists that there were many in the world with little interest in their vision but significant interest in money and other resources that were becoming available, and they had the ability to fit their activities into the loosely defined field of nanotechnology. After all, entrepreneurs are well known for creativity in the pursuit of funding, and scientists are (this is less well known) extremely creative in the pursuit of funding.

The result was a backlash. The creators of the field, the futurists, looked at all the newcomers and their flexible definitions of nanotech, and thought they were changing the meaning of the term and were pursuing different futures than the one originally envisioned. Government officials saw a flood of funding applications and realized that the topics were too spread out to provide any kind of consistency unless the funding agency enforced it. Government interest in nanotechnology started with the futurists’ initiatives, so officials could ask the futurists for help in making a stricter definition of the field. The futurists were pleased to help, given that the field was losing clarity and they were losing funding as competition increased. Interestingly, even some interlopers such as scientists and entrepreneurs started rethinking the meaning of nanotech, seeing it as too trendy a term and not well enough connected to their work.

Nanotech started out as a word with a clear symbolic vision and few adherents. Money was added, and it became unclear and populated with many newcomers, members of peripheral communities. This makes sense. The next step is the surprise, because everyone in the field started looking around and seeing a need to sort things out. The founders and funders of the vision stayed, and the newcomers started leaving. That’s how nanotech shrank, and it could well be how many other fields expand and contract over time.

Monday, November 27, 2017

Improving Evidence Presentation: An Example and Some Tips

This is an unusual blog post because it is on how to do research, not on what we have learned from research that has been done. Well, there is also something on what we have learned. I think it is very important that researchers show the data in their manuscripts by making graphs that show the reader the phenomenon and their explanation of it. This is not easy to do currently because management journals like models more than they like graphs, and models don’t show the data as clearly as graphs do.

As the editor of Administrative Science Quarterly, I am encouraging authors, associate editors, and reviewers to use more graphs. I also do it as an author, and this blog post is about a paper in Advances in Strategic Management that I wrote with Seo Yeon Song. We analyzed the ebook business, and our starting point was that there is a big movement toward self-publishing and indie (independent) publishing there, with an increased market share relative to the Big 5 publishers. Here is the graph showing this change:

How did we explain the change? Big 5 publishers can pay for advertisements, unlike indies, so indies must have some other advantage. We thought it was their readers rewarding good indie books with tweets and reviews on the amazon.com website. Here is a comparison of how Amazon reviews affect sales of Big 5 and indie ebooks:



See the difference? Indies don’t have advertisements to support sales, so each new review increases their sales more. This is something that can be seen from the data without any modeling. Of course we also modeled the data. I won’t show the model here, but instead show a graph comparing the effect of Amazon reviews (the count), Amazon review score, tweets (the count), and sentiment (how positive they were). It is easy to see the results, right? Amazon reviews have a much stronger effect than Twitter posts.


Finally, here is a graph that shows the review effect in a model that extracts all other effects we could control for, such as the tweets. This is called a residual graph, and it can be used to check how much of the relation between the reviews and sales is explained by other factors. The answer is… almost nothing. This graph (a residual graph) is visually nearly the same as the earlier one. It also shows how much is left to explain by other factors that are not yet in the model, which is clearly a lot.



Well, this was a short story about ebook sales, but the more important point is that researchers can show their findings well just by graphing the data. If you want to see the program that made these graphs and some sample data to use it on, click here and here.

Thursday, November 9, 2017

Going Back and Doing Good: When Foreign Workers Return Home

Here is an interesting contradiction: Some politicians say that relying less on foreign workers will make their nation more competitive, but in fact it makes the workers’ home country more competitive. Notice that I said contradiction, not paradox, because it is not a paradox at all. It is logical, and it is supported by recent research.

Here is how it works, as explained in an article by Dan Wang in Administrative Science Quarterly. Foreign workers are often used by highly advanced and competitive firms, because those firms are best positioned to take advantage of a worker’s skill wherever it is found, and to transfer it to wherever it is needed. They also have excellent production processes, advanced technologies, and knowledge on how to best operate these. Sometimes their foreign workers go back to their home countries (usually voluntarily). What happens then?

The start is quite simple. These workers may be holding knowledge of great value to firms in their home countries, so the key is whether they can make a knowledge transfer back home. The firms that hire them, or the new firms they form if they become entrepreneurs, will benefit from their knowledge. But the full story is not as simple as the start. These workers differ in how well connected they are to others, in the companies they worked with abroad, and in the companies they work with after returning. Their personal networks differ in how many people they know and how well they know them. It turns out that knowledge transfer depends greatly on these connections, because the greatest transfers happen when a worker is highly connected both abroad and after returning home.

The conclusion is clear. Playing the competitiveness card may be a good way to cater to xenophobia among voters, because those who prefer fewer foreigners around like to hear reasons for their dislike. (Even if the excuse isn’t true, it is nice to have an excuse.) But competitiveness is not a valid reason to send foreign workers home.

Wang’s research had one more important conclusion: it was not just personal networks that made knowledge transfers effective, but also an absence of xenophobia in the home country. Now the contradiction becomes even more interesting. Xenophobic policies of sending people home may be phrased as helping competitiveness, but they usually hurt it — except when the workers come from a country with xenophobic people, because then the knowledge they have won’t transfer back. Xenophobia is a lose–lose proposition.


Friday, October 13, 2017

Humility Empowers: How CEOs’ Attitudes and Actions Cascade Down the Firm

Managers are often given advice that combines research-based buzzwords with fundamental misunderstandings. Empowerment is a good example. The word contains “power,” and advice is usually given to those who feel a need to improve, so it was perhaps inevitable that empowerment should be associated with ideas of individuals becoming or feeling empowered through some action of their own, such as attending a course or a coaching session on how to become empowered. That’s not what empowerment means or how it works: people are empowered when someone else gives them power and authority to make their own decisions.

The distinction is important because if empowerment improves organizations, then we should start by looking at the person who empowers others. That’s exactly what was done in research published in Administrative Science Quarterly by Amy Ou and collaborators. They looked at how a CEO’s humility could empower others in the firm. Their central insight is simple and powerful: a CEO’s humility makes it easier for the top management team to work together, because each feels empowered and comfortable, and that effect on the top management team cascades down the organization.


They found that the humble CEO is the opposite of the showy CEO in two key ways. First, the humble CEO does not dominate but instead is understated and makes it easier for the closest executives to stand up and perform. For example, the humble CEO does not make public performances and speeches to the whole organization but instead lets the empowerment of the closest executives cascade down. Second, humble CEOs encourage communication to give shared understanding, which in turn lets subordinates feel motivated and confident about their decisions and helps their managers trust their judgment and commitment. This cascading down of shared understanding and trust can bring whole organizations together more effectively than inspirational speeches by showy CEOs.

There is much about that research that appeals to us, because most of us share the suspicion that showy, self-promoting, narcissist CEOs must be flawed in some way. Yet such CEOs are very common, and part of the reason is that it is easier to imagine people who make a big impression also having big effects on the firms they lead. Humility is such a low-key behavior and such (what else can I say?) humble thinking that it is hard to imagine it having a big effect. But it does. How?

The keyword is empowerment.  Humble CEOs empower their top management teams. Empowered top management teams create an organizational climate that empowers workers all the way down. A top management team that has been empowered and in turn empowers others creates norms that are so strong that it is hard to be an authoritarian manager. That’s how humility has big effects on organizations: it creates an expanding circle of empowerment.


Thursday, October 5, 2017

Post the Job or Slot a Person? It Depends on Whether You Want Performance

Five years ago, I posted a blog on award-winning research showing that firms pay more and get less when hiring from the external labor market. The reason is that managers know more about their current employees, and they overestimate the value of external hires because they focus on their formal qualifications. When it comes to filling jobs internally in the organization, a manager still has to choose between placing a known worker into a job or posting the job and assessing applicants, who may or may not be familiar to the manager. With internal hiring, do managers still undervalue the workers they know?

Research in Administrative Science Quarterly by JR Keller answers that question. He looked at the difference in job performance and pay between jobs that were filled through posting and applying, and jobs that were filled through the manager picking someone (slotting).  Managers could choose which way to fill each job, and naturally they used slotting when they knew someone who fit and posting when they were not so sure. This is the same as choosing between internal job mobility and the external job market, because managers pick internal candidates when they know someone who fits the job and hire externally when they are not so sure. They also have more candidates when posting, and they know less about the applicants because they are usually from other parts of the organization. In every way you can imagine, the choice between posting and slotting is similar to the choice between external and internal hiring.

So how wrong is it to fill an internal hire through posting instead of slotting? Here is the surprise: It is not worse to fill by posting, but better. Posting means higher job performance, both absolute and compared with others. It means lower chance of leaving the job, except for leaving for a promotion, which is more likely when filling through posting. Oh, and it also means higher pay for the employee filling the job. So, for the employee this looks like a good thing; more pay and better chance of a promotion. For the firm, it looks like paying more to get more, so the net effect depends on how much more the firm is getting. In this case the answer is easy because the lower turnover from the job alone shows that the firm benefits from using posting. The better job performance is icing on the cake.

But this raises the question of why markets work better inside an organization than outside it. What is it that the manager can see better when posting internally? The answer is, nearly everything. Organizations know a lot about their employees, and this knowledge is readily available when filling jobs through posting. Not only that, the posting process forces the hiring manager to think carefully about what information to use and how to weight it in the decision, giving a more systematic and higher quality choice. All the information is there, and posting gives more choices and a better choice process.

There is an important lesson in this that goes beyond filling positions. We often have beliefs about the benefit of markets relative to social arrangements like networks. We forget that there are many kinds of markets and many kinds of social arrangements, and ultimately decision making comes down to what to choose from and how to make the choice.

Posting and Slotting: How Hiring Processes Shape the Quality of Hire and Compensation in Internal Labor Markets. Administrative Science Quarterly: forthcoming. 

Wednesday, September 27, 2017

There Can Be Only One: Authenticity as Trope or Reality

We live long lives with many new experiences, yet popular culture tells us to be the same. Be true to yourself, they say. Managers have long careers with many roles, yet researchers and self-helpers tell them to be the same. Be the authentic you, they say. It is said so many times that it must be true. Except that things said too many times by too many people need to be researched, because they might be wrong and in consequential ways.


A new article in Administrative Science Quarterly by Brianna Barker Caza, Sherry Moss, and Heather Vough looks at the connection between being authentic and being the same, and it finds that what everyone says is not quite true. They asked whether consistency (being the same) is the same as authenticity (being one’s true self) and found that the answer is no. The problem with saying that authenticity demands consistency is that one’s true self is not a unified whole. We can think multiple thoughts, have multiple beliefs, and take on multiple roles, and each of these can be fully ours even though they are not consistent with each other. People are smart enough that they don’t have to be only one thing, and they are flexible enough that they don’t force themselves to be only one.

To do the research, the authors followed the careers of people who had multiple jobs at once and in some cases also changed these jobs over time. This is a strict test of authenticity because we understand and accept that people can be different at work and privately – like the quiet student who is a very outgoing and improvisational musician. Not surprisingly, the demands of authenticity were a burden for these people with plural careers. They knew that they were asked to be authentic, and that this implied being the same always, but they also felt these demands to be unnatural. Being authentic according to others was the same as being inauthentic according to themselves. So who wins this battle?

There can be no winner, but the subjects of this study usually found a truce that worked well for them. On the one hand, they had to draw lines between who they were and who they presented themselves as, but these lines did not involve acting – they involved presenting the part of themselves that belonged to the specific job they were doing at the time. Sometimes they could even present the more complete self, but they were careful about when. On the other hand, they incorporated the multiple roles and identities that belonged to them as part of themselves, and they saw this incorporation as authentic and valuable. “There can be only one” was a demand they did not have to follow because they could shape their careers and benefit (and let others benefit) from the learning and flexibility that these multiple identities gave them.

Think about the people around you. Some may seem unusual because they simply do too many things, or too different things, and we sometimes suspect that some of it is inauthentic – they act for some benefit. But you could easily be wrong, and you could underestimate their commitment to each activity and the value they add to it.

Thursday, September 21, 2017

Harm to Similar People: What Awards Can Do

I recently discussed in this blog how status and prestige can do good things for a person, an organization, or a product. This effect has seen much study because it is a social fact that is very influential, and it seems arbitrary and illogical. Identical items are valued differently by supposedly smart people. We have lots of evidence that status gives benefits, including to those who are in some way connected to others with high status. To be seen with the elite is to gain some elite-ness, and casual observation of how people crowd the most prominent people in receptions will tell you they know that.


Can status also cause harm? A new article in Administrative Science Quarterly by Brian Reschke, Pierre Azoulay, and Toby Stuart found an example of this. They looked at the very prestigious appointment to the Howard Hughes Medical Institute (HHMI), which marks appointees and their research as especially accomplished, so much so that it can be used to predict future Nobel Prizes. Receiving this appointment is clearly a good thing. But what happens to those who are doing similar research but were not appointed to the HHMI? Is the similar research a connection that lets them gain some status? Or does it mark them as losers in a race for who is most important and should receive attention? Even worse, does someone else’s HHMI appointment determine what the final answer to a research question is, so that others conclude that this question can now be ignored?

Of course the blog title gave it away: prizes harm the status of similar others. Once a prize is announced, their work is seen as less and less important—their influence erodes. This erosion increases over time, and it is greater for younger (so less established) similar work. This is important to learn about on its own, because it is so different from the positive “status contagion” that we are used to finding. But there is also an exception to this finding, and it is just as interesting. Circling back to the logic that being near the elite can give some elite-ness, a prize might be able to bring recognition to similar others, and through that give them status. Resche, Azoulay, and Stuart found that can be true, but only if the similar others have little recognition to begin with. Once their work is established enough to be seen as important, the prize has negative effects.

So, prizes can give status to the less established, lifting them up. More often, what prizes do is to settle who among the established are most important, bringing one up and pushing the rest down. Anyone working in an area in which prizes are given—and pretty much any occupation or industry that involves design has prizes—will appreciate this research, because it confirms a basic intuition. You should hope to get the prize and fear that the prize is given to someone like yourself. 

Sunday, August 20, 2017

Power from Arbitrary Places: What Awards Can Do

We have a general idea that status and prestige can do things – good things – to a person or organization. We are all familiar with how the prestigious classes of wine demand higher prices, even for a given quality level; there is evidence on this in research on French and California wines. Status effects are also well known from many other contexts, and a more-consequential example is financial markets, in which the most prestigious banks gain price and distribution advantages over all others.


A new paper by Anne Bowers and Matteo Prato in Administrative Science Quarterly gives interesting new details on the effects of status. They look at equity analysts, who seek to help investors in the stock market by issuing reports on firms and estimating their future earnings. This is difficult work, both in getting the estimates right and in gaining the confidence of investors, but some analysts are so highly regarded that their estimates can move the price of stock they report on. They have market power even though they just act as observers and forecasters. But how can an investor determine what analyst to pay attention to?

Conveniently, there is the magazine Institutional Investor, which caters to the large (and very powerful) institutional investors such as mutual funds and pension funds, as well as ordinary investors. The magazine has an annual All-Star award, given mostly for accurate estimates but also for other qualities such as high service to customers (again, institutional investors). This award is prestigious, but it is also a sign of quality. If the market could consider the quality and prestige aspects of the award separately, someone who was nearly good enough for the award should gain nearly as much power as someone actually getting it. You are probably guessing that this is not what happens. And you are right.

Bowers and Prato had a very clever way of finding this out. The award is given across many categories of equities, and these categories often change through addition, deletion, combination, and splitting.  This is neat because it means that an analyst could become an All-Star, or lose an All-Star award designation from a prior year, simply because the categories changed. Focusing only on these changes in awards, they found that the difference between having an award and not having one was a great deal of market power. Gaining an award meant that an analyst moved stock prices much more; losing one meant that an analyst moved prices much less.

The second of these effects should give you pause. Financial markets are supposed to be smart and to be able to predict the average outcome of many future events. But the loss of market power when an analyst loses the All-Star distinction because of a category change suggests that the markets are forgetting what they knew. The quality of an analyst doesn’t suddenly change, so if the market power changes, we know that the market has forgotten the quality. This is not good news for those of us who let institutional investors such as mutual funds or pension funds hold our pension investments.

Monday, August 14, 2017

Underselling Labors of Love: Why Give Discounts to Rich Tourists?

The world is full of people in creative occupations. Taking a broad view of creative occupations as those involving work with personal shaping of the product and service, about 40 percent of the world’s workers are in these occupations. Among them, artists and craftspeople are the ones we most readily associate with creative work because they instill their work not only with personal design and careful craftwork but also with a passion that makes each piece a labor of love. We recognize this most readily with artists making one-of-a-kind works, but many craftspeople turning out decorative items also make each piece an individual expression. They should get paid well for this, right?

Maybe not. In a new article in Administrative Science Quarterly, Aruna Ranganathan studied the pricing of wood bangles made by craftspeople in southern India, finding that the artistic ambitions of the craftspeople had a surprising effect on the prices they charged, relative to prices charged by traders selling exactly the same goods but not involved in their creation: they gave a discount to buyers who appeared to be especially appreciative of their work. The reason became clear from how they described their work. Unlike traders, who freely admitted selling crafted work just to make money, the craftspeople took personal pride in every piece they made and were especially attached to the best ones. Some items they refused to sell; others they made sure to sell to people who seemed likely to appreciate them and display them prominently.

This makes sense, because every artist wants to be acknowledged and wants the work to be appreciated. Indeed, this was especially important to the craftspeople Ranganathan studied, who saw their work as having such strong elements of the sacred that they viewed their workshops as being like temples. But what’s harder to understand is how craftspeople determine whether someone will appreciate their work. Not every transaction involves words, especially in an area that attracts many tourists who don’t speak the local language, as was the case in Ranganathan’s study. Instead, the craftspeople looked at the customers. And the financial decisions they made based on what they saw might be surprising.

If a customer wore handcrafted jewelry or clothing, or carried a handbag made from natural fibers, the craftspeople considered these clear signals that they would appreciate great craftwork. The craftspeople also believed that foreign tourists, who are fairly easily identified, would see their work as more exotic and be more likely to appreciate it. These two groups have something in common: they are likely to be wealthier than local customers wearing inexpensive items such as plastic jewelry and carrying synthetic handbags. Yet the craftspeople offered discounts to both of the wealthier groups and charged more—market price or even above—to the poorer customers. Market price (or higher) for the poor, discounts for the rich. It seems strange and unfair, but in creative work money is just part of the transaction—appreciation is the other part, and for the artist, this is a tradeoff.


In reporting evidence from social science, we often end up looking at behaviors that make sense on one dimension and not on another. I perfectly understand the artist who is willing to give a discount to have a piece appreciated. I don’t like the idea of the richest customers getting discounts. I suppose the best thing to do is not to bargain too much when buying art as a tourist. Hand-crafted items from local artisans should provide the artisan with both appreciation and a better standard of living.


Wednesday, August 9, 2017

When Occupations Rule the Workplace, Which Matters Most?

 “The modern workplace” is an expression often used as if there is just one kind, while the reality is that workplaces are complex and differentiated. But if one workplace deserves to be ranked as increasingly important, it is the multi-occupational workplace that has not only multiple occupations creating a product or service together, but also no order in the form of top-down hierarchy or start–finish sequence. The occupations in such a workplace work together, and the potential for failure anywhere to ruin the output gives every one of them power. Think of the large and growing medical sector, the many new services using information and communication technology, and the increased customization allowed by computer-aided design and manufacturing.

So how do occupations interact when all are interdependent, all at once? Beth Bechky and Daisy Chung showed in a new article in Administrative Science Quarterly that it depends on how the organization acknowledges occupational power. They studied firms doing equipment manufacturing and film production: both places with multiple occupations interacting at a high level of expertise to achieve customized, high-quality output. It turns out that the procedures in the two types of workplaces were very different, although they shared the feature of each occupation having significant power over its own work and influence on the other occupations. This combination of having power and being influenced was not a battleground, but a complicated and pragmatic interaction.

The differences in procedures were clear. The equipment-making firm maintained the semblance of hierarchy and temporal order, with products starting out as engineering documentation and proceeding to test assembly and manufacturing, but the actual work involved feedback and adjustments that led to cycles as the later stages made clarifications and corrected mistakes. Importantly, everything was in theory documented formally, even the adjustments, and work was done “by the book.” Film production, on the other hand, did not seek to define a hierarchy among occupations and had simultaneous interaction as production proceeded. In film production, the power of each occupation over its own work was fully acknowledged, and interaction among various occupations was direct and egalitarian; in equipment production, the power of each occupation over its own work was hidden, and interaction among various occupations was channeled through the process of documenting the product specifications.

These differences also affected how each occupation functioned internally. Because some occupations in equipment manufacturing were formally seen as subordinate, they conducted close internal quality checks to ensure that their members’ work was perfect along the dimensions they controlled. That way they maintained as much control as possible. Because film production lacked such ranking, the emphasis was not on internal control to keep quality uniformly high, but more on recognizing each member’s specific strengths and mentoring junior members into the occupation.

Why the differences? Keep in mind that these occupations are working together in a pragmatic way to solve organizational problems. The main class of problem has to do with time. Any organization dealing with occupations with well-established hierarchies is dealing with historical time and must not deviate too far from how things were done before, when interdependence was less. So they make the documentation system work in ways that maintain history and handle interdependence. Any organization dealing with unfolding events is dealing with the event clock and must not slow down direct interaction among whatever combination of occupations has the capability to deal with the current emergency; even as the occupations have widely differing formal authority, scarcity, and pay, they interact as equals. Organizations operate to deal with time.

Bechky, Beth A. and Daisy Chung. 2017. Latitude or Latent Control? How Occupational Embeddedness and Control Shape Emergent Coordination. Administrative Science Quarterly, forthcoming

Thursday, July 27, 2017

What Is Healthcare All About? Care, Science, or Just Cost?

I am writing this blog post while Congress is debating whether and how to change the U.S. healthcare system. This is something that, according to the news, will be determined in three days and will be decided by a simple majority without analysis of consequences by the Congressional Budget Office and without hearings involving the affected parties. This seems like a bad time to talk about long-term planning and a broader view of healthcare.

Let me do it anyway, starting with an important research paper by Mary Dunn and Candace Jones in Administrative Science Quarterly.  They looked at a central input to healthcare: medical schools, which educate the doctors who are in charge of providing the type of medical care that requires the MD degree and who direct the medical care done by others, such as nurses and lab technicians.  Although medical care is much bigger than doctors, they are at its core, so how they are educated is consequential. The paper found that medical schools have had an enduring division between two logics on how to think about healthcare, and one of these has become more influential over time. This view of education, and healthcare, is very new and needs some explanation.

Much simplified, the logics are as follows. One is that education is the transfer of scientific knowledge, so that this knowledge can be used in practice. So, medical school is about the science of medicine.  The other is that education instills values, and the guiding value for medicine is to care for people. So, medical school is about care, which can include care for people who are not yet patients and people who cannot be cured.

Dunn and Jones wrote a fascinating paper on how these two logics were in contention over decades. Advocates of each logic recognized the conflict with the other and fought for resources and attention to be devoted to their own logic. They saw an emphasis of their favored logic as essential for the health of medical education and the health of the nation more broadly. I don’t have room to describe their discussion here, but an important conclusion is that the care logic has grown in importance, and a key element of its growth was greater public discussion on managed care, with its emphasis on preventive medicine and focus on reducing costs.

This conclusion has become even more important in light of recent events. Managed care helped make the care logic more prominent, which facilitated useful initiatives such as thinking of public care more broadly to include prevention of disease and thinking of patients as individuals in need of care instead of just as cases with prognoses. But it also had two other effects that play a role in current politics around healthcare. First, the science part of medicine became less important, which now has become part of a greater movement against science having a role in state decisions overall. Second, cost is now a key consideration in health, which is clear given that estimates of tens of millions of people losing health insurance are acceptable as long as costs go down. Cost savings are even more important than the science behind estimating that tens of millions will lose their insurance, with the health consequences this will have.

Thursday, July 20, 2017

The Manager as a Temp: Preparing for the Modern Labor Market

The news keeps telling us that employment is becoming a more-temporary state, with job changes both the result of footloose employees and of firms treating their workers as easily replaced, downsized, and upsized as needed. Not to mention that many now work as contractors, not employees, like Uber drivers. In these stories, the managers are typically cast as the bad guys treating everyone else as expendables. There is some truth to that, but there is a flip side: managers are also temporary. They are quickly moved around or even fired, and they also try to use job changes to move up faster than they could by staying in place.

How can managers be temps? Not only are managers credible to their subordinates only if they are expected to stick around for a while, but their chances of being promoted depend—ironically—on being able to signal that they are in it for the long run, even in a firm that habitually lets managers go. Well, for every problem that can’t be solved, there is a business school claiming to solve it. Research in Administrative Science Quarterly by Gianpiero Petriglieri, Jennifer Petriglieri, and Jack Denfeld Wood looked at how the participants in an MBA program used their education to make themselves more portable across firms and jobs. They were learning to turn themselves into managerial temps and use it to benefit their careers.

Like any education, business school is a journey, and the path and destination are unique for everyone. But there were clear patterns that tell us a lot about careers, and about management, in the current labor market. One path was to use the education to adapt both skills and identity to how firms now treat their workers, including managers. This is an instrumental pathway, where the idea is to understand the rules of the game and play it well. The other path was to use the pathway to explore one’s own preferred role in this world, and shape an identity that matches this discovery. This is a humanistic pathway, where the idea is to understand the parts available in the play and audition for the one that is the best fit.

These paths cannot easily be taken while working, because the everyday demands of actually managing make the learning process difficult, and changing identity isn’t possible either because everyone looks for and values constancy. So education acts as a valuable hiding place—a bubble, or a deep dive—where changes can happen and it is possible to emerge fully formed, or at least nearly so.

The next question is of course what the manager temps will do when they manage worker temps. Will work get easier when both manager and managed understand that they are not in the firm to stay, and the most stable part of their identity is its portability? To know that we have to wait for more research.


Wednesday, July 5, 2017

Committing to Change: How Change Agents Become Effective

Management practice has its fair share of cynics, and one story that many cynics will tell is that there are three sources of inefficiency in organizations. The first is that they can’t change into better practices, the second is that they pay for consultants who can’t help them change into better practices, and the third is that they pay for business schools to teach their managers, who end up not being able to change them into better practices. As a faculty member of a school that benefits a lot from the third source, I can at least say I agree with the cynics on the first two. But then, what can be done?


New research by Melissa Valentine in Administrative Science Quarterly has looked closely at consultants and organizational change, and offers some very helpful insights. She studied efforts to improve a cancer treatment center, so the changes were not just simple matters of reducing cost but had significant health outcomes. What did she find? First, it is completely true that significant consultant effort can be invested with no real change as a result. Money wasted, in other words, but perhaps worth trying because it is likely that nothing would have happened without the consultants either.

But consulting changing nothing was just one result – there were also some consultant efforts that did produce better practices. Importantly, the difference in how the consultants and the cancer center interacted in the unsuccessful  and successful cases was so systematic that this research gives clear guidance on what needs to be done to improve organizations. The difference can be summed up in one word: commitment.  And I will write the rest of the blog without any reference to interpersonal relationships, although I admit to being tempted.

Consultants hear from organizational members what works well and what does not, and they collect ideas on how improvements can be done and who would be in favor of them and under what conditions. This is done every time and has nothing to do with success or failure.  The success came from taking one more step. Whenever possible improvements were suggested and had some level of support across the organization, the managers who would be responsible for making changes were asked to renegotiate their obligations to each other and to implement the necessary changes. The renegotiation is needed because changes in complex organizations typically cross boundaries of managers and are most effectively handled by direct negotiation, not by referring up to the shared manager. Immediate implementation is needed because it is easy to give nice-sounding promises without accepting the cost of actually following through.  In other words, the success came from making managers decide what to commit to and then making them commit.

This was not just done as a final set of activities after delivering a report. It was a continuous effort, step by step, in which managers made adjustments and re-adjustments, set time tables and expected commitments, set new goals and measures, and followed up. The process also went far beyond managers, because hospitals also have another very powerful group: the doctors. Efforts to integrate their concerns were made in both the successful and unsuccessful project, but again, the successful project pushed all the way to commitment. In the successful case, the key decision makers ended up feeling obliged to fulfill promises they had made to others in the organizations – not to the consultants – and as a result, the organization changed.

The implication is clear. Consulting is often seen as an effective way of making changes because changes require a time investment, and organizations typically don’t have the resources to do their regular work and make time investments all at once. But increasing the capacity to propose change does not relieve the organization of the responsibility to negotiate, decide, and commit. Without those added activities, it is paying for nothing.


Thursday, June 22, 2017

In the Fight: What U.S. Army Mental Health Tells Us about Goal Conflict

Goal conflict is a necessary and central part of how organizations work. This is true even if we ignore personal goals that may differ from the organization’s and focus only on the goals that all have to meet, at the same time, for the organization to function well. Cars are made in production lines that require quality, speed, and low cost. Airlines require safety, service, and low cost. Health care requires personal attention, standardized procedures, treatment of all possible conditions, and again low cost. And finally, important for this blog, an army requires its soldiers to inflict injury on others, risk or experience injury themselves, and maintain mental health good enough to go out and do it all over again.

A paper in Administrative Science Quarterly by Julia DiBenigno looked at the goal conflict between the U.S. Army’s commitments to providing mental health care and keeping its force mission-ready, and her findings are important for any organization. She addressed a fundamental problem of goals that are in conflict: usually each goal is assigned to specialists with expertise in that specific goal, so resolution does not happen inside someone’s head but rather as an interaction between the people in charge of each specific goal. Usually that is done by prioritizing one goal and assigning the other goal to a service-providing or supervisory function in the organization.

The U.S. Army exists for fighting, and naturally commanders are in charge. But mental health care is also a high-priority goal because the recent wars have put a heavy load on each soldier, and post-traumatic stress disorder and affiliated conditions take highly trained soldiers out of action. Many even commit suicide, spreading the pain more broadly to also affect families of military personnel. This is recognized as a key problem by everyone involved, but solving it involves negotiation between specialists. This leads to push-and-pull with two frequent results: the health care provider is coopted by the commander and serves the commander’s purpose, or the health care provider stays anchored in the care identity and interferes with the commander’s purpose. As a result, most conflicts are poorly solved: analysis found that 5 percent ended with a good mutual solution, in 85 percent either the commander or the health care provider won the battle, and in 10 percent both lost out.

But here is the key message of the article. The statistics I cited were for only two of the four brigades DiBenigno studied. In the other two, 89 percent of conflicts led to a good mutual solution, in 7 percent one party won but not the other, and in 4 percent both lost. This is a really large difference, and the reason for it boiled down to one minor change in organizational structure with major consequences for the process. In the successful brigades, each health care provider was embedded in the clinic but also assigned as a point of contact with specific commanders, which led to longer and more personal interactions than in the other two brigades. The result was an anchored personalization: the provider was anchored in a group of other mental health professionals who shared knowledge and norms, and the provider had a personal network of commanders that allowed learning each commander’s needs and earning trust as well.

The personal interaction proved to be central to understanding each other’s thinking and finding adaptive actions in each situation. It had a massive effect on the ability to find good solutions, especially because the goal conflict was unique in each case. Mental health issues are complex, but so are the needs of military units and their commanders. Perhaps most remarkable is the origin of the difference in problem-solving capacity: just a simple change in organizational structure that regulated which care providers interacted with which commanders created a total change in how these interactions were done. It’s an important lesson for organizational design – how it is done determines what happens later.

DiBenigno, Julia. 2017. "Anchored Personalization in Managing Goal Conflict between Professional Groups: The Case of U.S. Army Mental Health Care." Administrative Science Quarterly, forthcoming: 0001839217714024.

Friday, June 16, 2017

Protest Outside, Protest Inside: How Social Movements Create Labor Unions

Occupy Wall Street protest with union members
Executives of large firms have been known to worry about social movement activity of three kinds. There are movements that encourage various kinds of costly state actions, such as cleaning up pollution or reducing carbon emission, which at some level will lead to taxation to cover the cost. There are movements that engage in boycotts and other actions to discourage firms from various cost-saving misbehaviors such as farming out production to nations with very loose labor and environmental protections. And inside the firm, labor movement advocates take action through established unions or through trying to form new unions where none yet exist. Responding to all this activity can exhaust executives, and they might not like to hear that these movements are related to each other.

How they are related is the topic of an article in Administrative Science Quarterly by John-Paul Ferguson, Thomas Dudley, and Sarah SouleThey look at how social movements outside the firm but in the same city influence unionization drives inside the firm. This is interesting because social movements and unions operate very differently, with unions under much stricter rules and restrictions, so the influence is not a result of workers learning anything useful about unions by taking part in social movements. In fact, it is not even clear that they do take part in social movements, because the mechanism behind this effect requires only that workers can see social movements, not that they participate.

Unions are built on procedures and ideas, with workers’ rights and equal opportunity among the most important ideas. It would make sense that the presence of similar progressive ideas in social movements in the same community could inspire union activity in firms, whereas social movements with more conservative ideology might have less effect on unionization because they have much less overlap with the ideology driving unionization.

This is exactly what the authors found to be true in U.S. cities. Protests in a city led to unionization drives in the same city, and this effect was stronger when the protests were related to progressive causes, including civil rights and gender equality. So protests outside a firm filter into unionization inside, specifically when the outside protests concern issues that workers inside also care about.  But there are additional details that make things even more interesting. Unions are not the only way for workers to solve problems. The Civil Rights movement and the women’s movement also had successes with changing the law, which meant that workers could contest gender or racial discrimination through the legal system rather than through unionizing. As a result, these movements’ effects on unionization were significantly reduced after the legal changes. So ideology matters, but competition from the law does as well.

Protests outside create unions inside, except when there are laws outside that make unions less necessary. What does that mean for our situation now? The laws outside are being weakened, and protests are getting stronger. Could it be a time for more unions?