Wednesday, December 18, 2019

Are Football (Soccer) Time-Keeping Rules Fair Play?

Think about this: football games are divided into two halves of 45 minutes each, and time lost during the regulation time can be added by the referee. What can go wrong? Well, to begin with, research by me, Nils Rudi, and Anup Walvekar published in PLOS ONE found that in one of the elite football leagues, most teams played 50-55 minutes on average per game, so 35-40 minutes of time were lost to game stoppages.

The loss of time isn’t fair either. Our research found that when play was stopped, teams that were ahead in the score spent more time restarting the game than teams that tied, and they in turn spent more time than teams that were behind. An advantage meant more time wasting, and this was true for goalies doing kickoffs, goalies kicking or throwing a captured ball, freekicks, throw-ins, corner kicks, and substitutions. See the figure for some examples of the time wasting (the vertical axis shows the time used).


To enforce free play, referees are supposed to prevent time-wasting, but from our research findings it is clear that they can’t do this. In fact, we analyzed the time-wasting and found that the teams were able to follow a nearly optimal (for them) time-wasting strategy, which means that that they are effectively sabotaging fair play, and they are also giving football fans fewer minutes of watching their favorite game than the fans would like to have.

What can be done? Football is special in many ways. It is the oldest organized team play that is currently big business. It is the largest sport worldwide. Larger revenue than any other, more professional players than any other, probably also more amateur players than any other (they are hard to count). And, it is the most important team sport that has a set game duration and no stopping of the game clock when the play stops. Think about it: the American Football, Ice Hockey, and Basketball stop the clock when the play stops; Baseball and Cricket do not have a clock; only Football and Rugby have a game clock and no clock stopping.

If the clock stops whenever play stops, it is impossible to take advantage of game stoppages to waste time and turn a goal advantage early in the game into a win. Most game stoppages would get shorter as a result. Goalie kicks are 80 percent slower under current rules than they would be if the clock is stopped. Goalie restarts are 24 percent slower, and freekicks are 36 percent slower when teams try to waste time. Naturally, the total game time would need to get shorter if the clock is stopped the play stops, but we can calculate how much shorter it should be for the fans to get the same number of actual play minutes.

Football is an important game for all its fans, and in fact the main irritation is when play is stopped, and the opposing team is leading and is slow to start the play. How many football fans have booed players from the opposing team who are suspiciously slow to get back up after a tackle, letting seconds tick off the clock while the whole stadium is waiting? If the clock is stopped, everyone would know that the pain is real, and the game is fair. It would be a big change in a traditional game, but maybe it is worth it?

Thursday, October 31, 2019

Rational Fouls? Yes, Professional Soccer Players Mostly Cheat Rationally

Fans usually watch soccer games with a great deal of excitement and passion. The players look excited too, even passionate. But remember, they are professional soccer players, hired and paid well to perform at top levels. Your local team may not be all that good (it depends on where you live), but the games broadcast on television involve top league sides that can hire elite players from anywhere in the world. Has it occurred to you that they may actually be more calculating than passionate when playing the game?

They are, at least most of the time. In a forthcoming paper in Organization Studies, Nils Rudi, Anup Walvekar, and I studied how soccer players foul each other. In other words, we studied how a soccer player decides to slide down, kick down, elbow, or pull an opposing-team player so that the referee calls a foul.

Fouls are an interesting topic, for a few reasons. First, fouls are against the rules of the game, but can benefit their team. They are a good comparison with other things employees can do that are illegal or immoral but benefits their firm in the short term, such as misleading or defrauding customers. Second, fouls are risky for the player. The referee is watching, and a violent foul not only gives a free kick, but also a yellow card to the fouling player. That can be very expensive for a player who gets a per-game bonus payment, as many of them do. Finally, fouls often look like acts of passion, as when a defending player slides into the attacker to knock him down.

So how do soccer players decide when to foul? Mostly they are rational. We were able to calculate the cost of fouling as change in likelihood that the team of the fouled player will score as a result of the foul. Intuitively, this is easy – fouls lead to free kicks, which can give goals. (Mathematically it is more complicated, but it is completely doable.) We were able to show that players choose are more likely to foul more the better it is for the team to do so. Not exactly surprising, except that the effect was quite strong. Many of you know enough about the off-the-field antics of some soccer players to doubt that they are fully rational, but on the field, they are complete professionals and experts in what they do.

But there are two exceptions, and both are interesting. First, there is the organizational goal of winning the game. Players and teams hate to lose a lead, and they foul less rationally when they are defending a lead. Second, there is the individual goal of looking good. Players hate to lose the ball, and they foul less rationally when they have just lost the ball to the opposing team. In fact, they completely lose their rationality if they are near the player who stripped them of the ball.

We know this because we can measure not only whether organizational and individual goals affect the likelihood of fouling, but also whether it affects how much the cost of fouling is taken into consideration. Defending a lead and becoming hot-headed after losing the ball both make the player think less of the cost of fouling. So, we know that in one type of organization with a highly expert team, decisions are made rationally except for when it really matters to the decision maker.


Tuesday, October 22, 2019

Punch or Block? How Organizational Members React to Social Movement Anger


The Dick’s Sporting Goods chain store stopped selling military-style semiautomatic rifles after the Marjory Stoneman Douglas High School mass shooting. No wonder: the Stoneman terrorist had bought a gun from one of their stores. Even though he didn’t use it in the attack, the store was one decision away from becoming an accessory to a mass shooting of school children.


Gun control is an issue that provokes significant anger from people on both sides. People who advocate for stricter controls express anger over the lives lost and the failure to quell domestic terrorism. People on the side of no gun control express anger over losing their right to buy any weapon they like.

Not all social movements provoke the degree of anger gun control does, but we have more to learn about those that do. A recent article by Katherine A. DeCelles, Scott Sonenshein, and Brayden G. King in Administrative Science Quarterly shows us something new about social movements that provoke anger and how organizations respond to them. The authors find that anger related to a social movement affects the response of organizational insiders who agree with the social movement, but not in the way we might expect. In the case of Dick’s Sporting Goods, the organizational insiders would be employees who agree that gun control is necessary and that the store should limit its gun sales and/or have stricter background checks.

Employees’ reaction to a social movement they agree with would seem fairly simple to predict: they would express their support of the social movement and try to influence the organization to agree to its demands. This is often the case, and employees are often successful: organizational insiders who agree with a social movement make organizations much more likely to change. In other words, a social movement often creates an opportunity to push an organization for change.

But this article points out that social movements invoking anger are different. Anger leads to a feeling of being under siege, so employees agreeing with the movement face a dilemma. They are angry too and would like to express it. Yet they also depend on the organization for work and pay, and they have reasons to fear that involvement in an angry social movement will lead to negative repercussions, whether from coworkers who disagree or from management. In other words, they have to decide whether an angry social movement is a good opportunity to punch for change.

The research showed that this conflict between anger and fear was resolved in favor of fear. While people outside the organization were more likely to act in response to the social movement when they were angry, organizational insiders were less likely to act. This was because greater anger also led them to fear negative consequences of acting. Employees often did not act on their anger but instead sought to protect themselves. In other words, they decided that the best response to an angry social movement was to block punches that might come from management, not to punch for change.

In Dick’s Sporting Goods, change happened. Not only did they stop selling the semi-automatic rifles that are favored by mass shooters, but they also destroyed them. They are considering stopping sales of guns of all types in their stores. Why did this happen? The decisive factor was that, angry or not, the CEO got on the side of the social movement. Management does matter, especially when there is controversy and anger.


DeCelles, K. A., Sonenshein, S., & King, B. G. 2019. Examining Anger’s Immobilizing Effect on Institutional Insiders’ Action Intentions in Social Movements. Administrative Science Quarterly, Forthcoming.

Monday, October 7, 2019

Angry Friends: How Past Collaboration Can Increase Conflict


Let’s look at something that most of us would agree is true: People and firms that have collaborated in the past will treat each other better in the future. This behavior is widely accepted to be true in business, and plenty of research evidence backs it up. It is part of what we call network theory, where one of the central ideas is that past collaboration or contact between two people creates a tie between them that facilitates future collaborations.

Like all things we believe to be true, we think about this one only when we see that it’s not. A recent article in Administrative Science Quarterly by Jose Uribe, Maxim Sytch, and Yong H. Kim looks at collaborations among lawyers. In most collaborations, people or firms work together for their own benefit. Lawyers also seek to benefit themselves, of course, but they do so by acting as representatives for others. In this paper the lawyers represent firms involved in disputes over intellectual property, which can be a very important and potentially valuable form of legal action.


A past collaboration means that the lawyers have been on the same side in an earlier lawsuit, usually representing different firms that have the same goal. This study considers what these lawyers do when they later represent firms on opposite sides of a lawsuit. One might think that lawyers who have collaborated in the past will treat each other better and come to a better solution in a shorter time. This is possible, but if the firms they now represent are strong rivals, the opposite happens: The lawyers are more aggressive if they have collaborated in the past. That means longer time to settle the case and fewer cases settled out of court.

Does this make sense? It does not if we assume that the firms observe and care about only the final outcome of the case. This article shows that having more-aggressive lawyers on each side typically leads to worse outcomes for both sides, such as dings to their stock prices. But it turns out that firms also look at their lawyers’ history with the other side’s counsel and care about the process, not just the outcome, and that is where things go wrong. Lawyers who have collaborated with opposing counsel in the past know they need to prove their loyalty to their current client, and the easiest way to do so is to be aggressive – the firm is watching the process and will see this as a sign of loyalty.

The results of this research do not mean that what we know about collaborations is wrong – only that it is incomplete. A history of collaborating would likely make most lawyers want to collaborate, but they are trapped by the need to prove loyalty to clients that are strong rivals. There is no way out of this trap, which also hurts the client, but they can make it less serious by agreeing to act collaboratively on smaller procedural steps that simplify the case for the judge and lawyers but are not inspected closely by the client. And indeed, some lawyers did exactly that, presumably because they still wanted some level of collaboration and also knew that this would be better for the client.

So we still know that past collaborations make people want to treat each other better, but the role of representing someone else can reverse that. To represent someone you need them to trust you, and if that takes some aggression against a past collaborator you will consider doing that. At least if you are a lawyer.


Uribe, J., Sytch, M., & Kim, Y. H. 2019. When Friends Become Foes: Collaboration as a Catalyst for Conflict. Administrative Science Quarterly, forthcoming.

Friday, September 27, 2019

Rise of the Moderately Capable: Succession in Family Business


How can the prince of an absolute monarch survive and inherit the throne? How can the child of a family business founder take over the business? Two different questions with the same answer: Don’t be too ambitious, and don’t be too capable. This sounds like a bleak story, so I should explain the research behind it.

The theory and evidence are found in a recent article in Administrative Science Quarterlyby Xu Huang, Louis Chen, Erica Xu, Feifei Lu, and Ka-Chai Tam. They look at the problematic situation that parents and successors of family businesses find themselves in. Family business founders believe that success comes from exactly their actions and values, so successors should imitate them exactly. Yet imitating them too closely creates competition, which encourages the parent to dominate the child instead of giving the successor freedom.

How can this domination be avoided? The answer is simple and discouraging. The successor should not be too ambitious or too capable, because these attributes would increase the competition and lead to dominance by the founder. In fact, highly ambitious and capable children will be dominated by the family business founder because they are too good, just as unambitious and incapable children will be dominated because they are not good enough.

The research team found evidence for this through surveys of family business founders and successors in China. But they went further than that: One part of their study looked at how princes of Chinese monarchies before 1000 AD fared, finding that those who were not too ambitious and capable had the best chance of succeeding the king. Of course, not succeeding an absolute monarch is a little worse than not taking over a family business – those who failed might be killed by their father and replaced by a more amenable prince.

So in some ways the world has improved, because the price of being too capable is no longer death. But this research still does not deliver good news. Family businesses are very important in most parts of the world. If only the moderately capable successors gain freedom, these firms will not become as successful as they might, to the cost of the family, the employees, and society. Strictly speaking this effect has been shown for only the founder generation of family businesses, and we could hope that it disappears in the second or third generation.  

Family businesses are complicated because they mix family and business, which are usually kept apart in modern society. In the ancient world family and business were the same thing, and I can’t help wondering whether that meant they had better mechanisms for dealing with the mix than we do now.



Monday, September 2, 2019

Buying Elections: How State-Owned Enterprises Can Serve Politicians


Getting re-elected is much easier when voters are content, such as when the economy is doing well and everyone has work. No wonder we see a president tell media that the U.S. economy is fantastic but also tell the Fed that the economy is in trouble and needs a strong boost from a prime rate cut. This is crude, but if everyone complies it will work. However, in nations with a stronger blend of state and private sectors, more sophisticated approaches are also available.

Scholars have long thought that state-owned enterprises (SOEs) have political uses, and in a recent paper in Administrative Science Quarterly by Carlos Inoue we learn more about how politicians can use them. The trick in this case is over-employment, which has to be done very selectively because voters in most nations are wary of the state, or the enterprises it owns, running up excessive bills that have to be paid through taxes. That’s why the use of SOEs has to be sophisticated, not crude.

What to do? Inoue looked at the water-supply sector in Brazil, which has exactly the blend of municipal authorities, SOEs, and privately owned firms needed to check whether SOEs are special. And indeed, they are special. SOEs hired additional staff exactly during election years (for governors, who are the ones controlling them). To maximize the electoral impact, they were careful to hire more staff in communities with greater poverty. Some SOEs also had private investors, who resisted effectively enough that the over-hiring in election years was less. So we see a sophisticated effort to buy elections through influencing SOEs: hire more when needed, where the creation of jobs is most effective, and whenever possible.

What was the cost of all this? In general, SOEs were not costlier than other firms, because they diverted resources from other areas to create jobs. But they did have lower profitability as a result of the greater employment during election years. They spent money – sacrificed profits – exactly when it was useful for politicians seeking re-election, but not otherwise.

Organizations are used in so many ways, some of which are far away from their original purpose. The water-supply industry is important in any nation, because abundant and clean water helps the society and economy, but at least some of these firms have a second purpose. This is something that both scholars and practical readers dislike, but at the same time we have to admire how skillfully it was done in this case.


Friday, July 26, 2019

My power-hungry, algorithmic boss

They say that computers will not replace workers but instead become their bosses. It makes sense. Humans can do many things that computers cannot, but computers are more rational decision makers. That presents all sorts of problems for the humans who have an algorithm for a boss. Praise and ingratiation do not work anymore because algorithms don’t have emotions, only data and processing rules. Algorithms give no explanations of what went wrong and encouragement to do better, or at least not the type of explanations that make sense to a human. What to do?


This question is the topic of a recent article in Administrative Science Quarterly by Corentin Curchod, Gerardo Patriotta, Laurie Cohen, and Nicolas Neysen. They looked at eBay sellers, who are a good example because they had algorithm bosses long before the Uber drivers and Upwork programmers did. As you might imagine, they found that algorithm bosses can be quite difficult.

The main problem with algorithm bosses is their power. They make decisions based on rules and data, and no one can question or appeal their decisions. For example, they can downgrade eBay sellers or even exclude them from the platform if buyers’ feedback is poor. But what if the data are manipulated? For eBay sellers, that is a serious problem because buyers can rate the sellers, which means they can use the threat of low ratings to negotiate better deals. There are ways to use the algorithm to detect manipulative buyers, but these don’t work well because such buyers understand the algorithm well enough to avoid detection, such as by changing identities. That leaves sellers vulnerable because they can’t change identities, so they are stuck with the buyers’ manipulation and the algorithm’s power.

From an eBay seller’s point of view, the algorithm as a boss starts to look like a prison. The boss is always right, even when the data it uses are wrong, and the data are often wrong because buyers use the system to manipulate the seller. There is a way to appeal, but the answers to appeals also sound like they come from an algorithm: eBay can’t monitor buyers’ behaviors because it is a platform that helps buyers find sellers, not more than that. It is not in the business of policing people.

So what to do when the boss is an algorithm? The eBay sellers know the answer, and their approach will work for others too. With human bosses, you manipulate emotions. With algorithm bosses, you manipulate the data. eBay sellers with some experience are always on the alert for buyers who look suspicious. Maybe their profile matches that of a known manipulative buyer. Maybe the buyer is very recently registered, as would happen if someone changes identities often. Maybe their behavior matches that of a manipulative buyer – the tone or content of email messages before agreeing to buy gives out warning signs. In each case, the response is the same: refuse to sell to that buyer, avoiding the risk of buyer data manipulation. Naturally, these refusals to sell are a form of seller data manipulation because they are done to avoid low ratings.

Of course, eBay sellers protecting themselves by excluding buyers is bad news for buyers who are innocent but match the profile of manipulative buyers. That’s not news in management research, because we know that workers who need to protect themselves against overly powerful bosses end up harming someone or something else. Making the boss non-human does not change the problem for the worker or their need to solve it one way or the other. I find it interesting to see that one can take human bosses out of organizations without also removing the problems that human bosses cause.

Curchod, C., et al. 2019. "Working for an Algorithm: Power Asymmetries and Agency in Online Work Settings." Administrative Science Quarterly, forthcoming.

Wednesday, June 19, 2019

Government by Blackmail: How States Can Influence States

We know that there is a big difference between theory and practice in how states relate to states. The theory is that each state is autonomous and rules what goes on inside it, and for anything that connects states they either follow treaties and conventions or interact as equals. The practice is that larger nations have the most say. The Mekong River goes through China, Laos, Myanmar, Thailand, Cambodia, and Vietnam, but the Mekong River Patrol jointly operated by some of these nations was a Chinese initiative involving significant Chinese police south of China’s border. US drug control south of its border is similar. 

What accounts for the difference between theory and practice? Research by Florian Ãœberbacher and Andreas Georg Scherer looked at this question by examining a specific example: how the famous Swiss banking secrecy was eliminated. Although there can be many ways that large states influence smaller ones, this example is a good illustration of one simple approach: it was done through blackmail by the US. The interesting part is how the blackmail was done.

The problem with states blackmailing states is that statecraft is supposed to be different from running a mob, and any state action resembling what mobsters do can create resistance in the short run and hurt the blackmailing state in the longer run. As of May 2019, we are observing the US placing tariffs on Chinese exports with the explicit purpose of forcing a more favorable trade agreement, an obvious blackmail tactic, and China responding by doing … nothing. The US actions against Swiss banks were more effective, not just because Switzerland is smaller but also because they did not target Switzerland. Instead, they targeted the Swiss banks and used the potential damage to the banks as a way of blackmailing Switzerland, which cares about its banks because they are a large part of the economy.

The script was simple. A whistleblower revealed (to no one’s surprise) that Swiss banks held money that should have been taxed in the US, and US authorities proceeded to demand that the Swiss government turn over information about suspicious accounts. In addition—and this is the important point—US authorities turned increasingly aggressive in pressuring Swiss banks, to the point of making it clear that they were ready to inflict significant economic damage. It was the state version of telling potential snitches that the boss knows where their families live and children go to school and can act on this information. It worked: banks panicked, and the Swiss government agreed to release much more information than before.

All this sounds unusual as a state behavior, and maybe it is something done only when authorities are getting tired of tax evasion and want to act on banks making it easy and on states protecting the banks. But wait, does this resemble something we are also seeing now? The Chinese firm Huawei’s chief financial officer is currently under arrest in Canada at the request of the US, which is seeking extradition. By President Trump’s executive order, Huawei has been banned from using US technology in its products, ranging from mobile telephone infrastructure equipment (such as 5G gear) to components of phones its sells outside the US. The ban is formally for security reasons, but in an interview, the president has explicitly linked its fate to the outcome of the trade negotiations. So statecraft still has a component of blackmail, and we can look forward to seeing how states like Russia and China will learn from this and change how they do statecraft in their vicinity.

Ãœberbacher, F. and A. G. Scherer 2019. "Indirect Compellence and Institutional Change: U.S. Extraterritorial Law Enforcement and the Erosion of Swiss Banking Secrecy." Administrative Science Quarterly, forthcoming.

Friday, June 14, 2019

“Be like DNA”: How Occupations Deal with their Technological Shortcomings

In my work as an editor, I sometimes give talks to authors telling them how social scientists feel inferior to those working in the physical sciences, because the physical scientists are so much more scientific. There is some truth to that inferiority, because most physical sciences definitely have better (at least more expensive) measuring devices than we do. There is also a myth attached to the inferiority, because most social scientists think that they can become better scientists by using various complicated econometric methods to analyze their data. As I pointed out in an earlier blog post, that is exactly wrong. Graphs can be more persuasive than models, and the physical sciences prefer simple methods.

This issue of science and technology envy happens elsewhere too. In a paper published in Administrative Science Quarterly, Beth A. Bechky investigated how forensic scientists reacted when the National Academy of Sciences accused all of them – except those working in DNA profiling – of being insufficiently scientific. This is a great test case because it is consequential – forensic science is important in criminal trials – and because it is a good example of what happens when one occupation gets access to a shiny new and advanced technology, and occupations working nearby watch and react. This is a frequent event in organizations, perhaps especially in healthcare with its influx of new techniques, but also in many organizations that make increased use of data processing and communication technology to improve their work.

The research showed that reactions to the damning report were strongly influenced by how compatible the new DNA technology was with the values and existing technologies of each occupation. “DNA envy” was felt in all the other forensic labs but led to action in only some. The strongest resistance was in firearms analysis, which is an occupation that values individual craft-like judgment and uses a method unrelated to the statistical analysis used in DNA profiling. The firearms examiners didn’t accept the call to become more similar to DNA analysts and saw no way of doing so anyway.

The strongest acceptance was in toxicology. No wonder – toxicologists’ values centered on minimizing errors and quantifying the precision of their measurements, which closely matched the values held by DNA profiling labs (and the new technologies they used). The toxicologists’ main reaction to “DNA envy” was to point out that they had been like DNA profilers all the time but hadn’t been recognized as such.

As usual, the most interesting case is the one in between. Narcotics labs have measurement devices based on technology similar to that used in DNA profiling (GC/MS, in case you wonder) but also rely strongly on a fast, accurate crystal test that has little in common with DNA analysis. While there was some technology similarity, the values of the two types of labs were in conflict because narcotics analysts strongly valued their independence and flexibility when assessing evidence, and they were proud of their speed. To them, becoming “like DNA” would mean becoming less than they already were, at least as individual contributors, and they feared that they would have to adapt in that direction. This fear was especially strong because they knew about other narcotics labs that were moving to the measurement device just to become more scientific.

This research is important news for all who deal with technology. We are used to worrying about whether the right technology will be adopted and whether it will happen soon enough. Now there is another issue to keep in mind: what are the side effects of adopting new technology, and are they good or bad?

Bechky, B. A. 2019. "Evaluative Spillovers from Technological Change: The Effects of “DNA Envy” on Occupational Practices in Forensic Science." Administrative Science Quarterly, Forthcoming.

Wednesday, June 5, 2019

The Sinner as Savior: How Marijuana became Medical

Ideologies and businesses mix uneasily. They can sometimes work together, as when social movements help promote a form of business. For example, socially responsible investing has grown that way. They can also be in conflict, as when an industry runs counter to strong societal beliefs. Other industries can sidestep ideological resistance. For example, coal-fired power plants can ignore global warming because people cannot always choose how their electricity is made. But in other industries, firms are hurt because their potential customers are turned off by the stigma attached to the business, sometimes including its legal status.

What to do when your industry is stigmatized? Kisha Lashley and Tim Pollock examined this question in a recent article in Administrative Science Quarterly, using the cannabis (marijuana) industry as an example. Their study is a good illustration of how an industry that was both illegal and stigmatized, and still is illegal at the federal level in the U.S., can act to overcome its stigma. The change for the cannabis industry hinged on one feature of the product: marijuana is not just a relaxing intoxicant but also a painkiller. This may seem inconsequential considering how many medical painkillers are available, but it matters for two reasons. First, patients differ in how they respond (or not) to painkillers. Second, the most powerful medical painkillers are addictive, such as the opioids that currently kill so many.

This painkilling feature was used to create a new ideology built around marijuana as a product that could help patients who were suffering and did not have other good options for relief. Activists, politicians, and others worked to legalize marijuana sales so these patients would not need to buy marijuana on the black market, with all the risks and dangers that could involve. Medical marijuana sold in stores – which industry leaders called “dispensaries” – was presented as a solution to a problem of human suffering and could thus occupy a moral high ground.  This was a story that marijuana proponents could tell politicians and canvass to voters, pushing for legalization in many states.
 
But what about traditional marijuana users who love the intoxication but don’t necessarily need the pain relief? They were both good and bad for the new stores. Good, because patients specifically seeking pain relief from cannabis will always be a minority in a community, so having an additional customer base is helpful for this new industry. Bad, because buyers wanting the intoxicant are exactly the stigmatized kind of “stoner” customer the dispensaries didn’t want politicians, the medical community, the media, and the general public associating them with. What to do?

The dispensaries needed to draw a fine line by taking strategic action. Through clean, stylish store designs they could look enough like pharmacies to avoid the stigma, even if many customers came for the high. Through logos and packaging (such as childproof containers) that mirrored those found in the medical field, they could establish legitimacy. And through changes in terminology, including the term “recreational use” to refer to the other part of their customer base, they could retire many of the stigmatized words describing marijuana and its users.

So is marijuana not stigmatized anymore? Well, it is a growing industry in the places that have made it legal, and a recent Financial Times article asked whether it can become a $100 billion industry. That does not mean it is completely destigmatized, but it has come far enough that there are opportunities for many entrepreneurs. Legal ones.

Lashley, K., and T. G. Pollock
2019. "Waiting to Inhale: Reducing Stigma in the Medical Cannabis Industry." Administrative Science Quarterly, forthcoming.


Sunday, June 2, 2019

Liberal and Conservative Companies: How Organizational Location and Politics Intermingle

You may know Nike as a liberal company in US politics – it is known for marketing statements with a liberal bent, and it also has liberal corporate policies such as transparency about manufacturing locations and labor practices. It is based in Beaverton, Oregon, a state that generally sends  Democrat representatives to the US Congress. Dell is a conservative company, though it is more obviously conservative in its political donations than in public statements. It is based in Round Rock, Texas, a state that most often elects Republicans senators and representatives. Are these facts connected?

A recent paper by Abhinav Gupta and Forrest Briscoe in Administrative Science Quarterly looked at the connections between firms‘ politics and their locations, and it found some relations that should be surprising. Let’s start with the more obvious ones. First, firms differ in how conservative or liberal they are. This can be measured by the political donations of their employees, and liberal firms are more sensitive to their surroundings and likely to concede to social movements’ demands. Second, the relation between employees’ political leanings and firms’ behaviors is stronger when employees are closer to the headquarters. The employees in Nike-owned stores far away from headquarters will not matter as much for the headquarters’ thinking and decisions as the employees in the headquarters.

Now for the really new stuff. Both Nike and Dell have political leanings that match their locations pretty well. That seems quite normal, and it matches some other firms we know about such as Seattle-headquartered Starbucks. But there are also firms that are more liberal or conservative than the average voter in their state, such as the software company SAS, which is in Cary, North Carolina and much more liberal than its state. If we compare equally liberal firms in a conservative and a liberal state, which one will be more likely to concede to the pressures of a social movement? You might think that it would be the liberal firm in the liberal state because its management feels more secure taking such actions. In fact, it is the opposite. The liberal firm in a conservative state will take a stand, so it will be more liberal than a liberal firm in a liberal state. Similarly, a conservative firm will take a more conservative stance if it is in a liberal state.

So, location and politics intermingle in organizations in ways that go beyond what you might expect if you thought of organizations as sponges that absorb whatever is around them. Employees can shape their organizations’ messages and actions in the political sphere. They are clearly more influential than people who live or work near the organization but who aren’t employees, because employees’ views not only shift organizational politics but also prompt the organization to take a stand against community members’ opposing views. People outside the organization can try to influence its beliefs and actions, but whether they succeed is largely a function of the views held by those on the inside.

Gupta, A., and F. Briscoe
2019. "Organizational Political Ideology and Corporate Openness to Social Activism." Administrative Science Quarterly, forthcoming.

Thursday, May 30, 2019

Will an Unprofitable Airline have Safe Planes?


Do you feel safe when you board an airplane for a flight? Typically we do, as we should because air travel is much safer than driving. Well, safer per mile (kilometer), but airplanes are much faster than cars, so per minute the difference is less impressive. And occasionally we get disturbing news like the two crashes of 737 MAX 8 aircraft, one of Boeing’s latest models, with a technical problem being the reason for the crash. We need the speed and convenience of air travel, for a reasonably priced ticket, but we also want aircraft builders and airlines to keep air travel as safe as possible. Are they doing that?

In the research paper“Safe or Profitable? The Pursuit of Conflicting Goals” that will be published in Organization Science, Vibha Gaba and I looked at that question. We knew that aircraft models build up different safety records through their years of operation because some models never (or hardly ever) crash, while others have more crashes. Airlines have this information, but do they act on it? It is expensive to sell an aircraft when it turns out that its model is doing less well in the air than other models, and tempting to hold on to it because the differences are small. Even models that have twice the usual crash rate are safe to fly, nearly all the time.

We found that airlines treat safety as a goal, and act on this goal. They buy and sell more aircraft when their fleet safety is lower than the average airline, and these transactions improve the fleet safety. The simple rule is “out with the bad, in with the good,” and if your fleet is good to begin with, hold on to your airplanes.  This is good news, but there are two complications in this picture.

The first complication is profit. Airlines also would like to be profitable, and these aircraft transactions cost money. Will the less profitable airlines hold on to their less safe models in order to save money? Actually the opposite is true – unprofitable airlines are particularly concerned with fleet safety, while profitable ones pay less attention to it. This sounds like a paradox but is easy to explain. An aircraft crash and loss of lives is very costly for an airline. A profitable airline can handle this cost without going out of business, but it could be fatal for an airline that is already. Aircraft safety is about survival for airlines too, at least some of them.

The second complication is the buyer. It is easy to buy safe aircraft, because both the aircraft maker and some airlines will be happy to sell. But how to sell aircraft models that everyone knows are less safe? Here we need to confess that our data are from airlines in the more developed parts of the world, not from the whole world. It is well known that less safe aircraft models and older airplanes are found in many developing nations. Some are also rebuilt and used as freighter aircraft.

So what can we learn from this? The most important lesson is that safety is maintained because firms view it as a goal and act to maintain it, at least up to the same level as other firms. They treat safety as a goal because society keeps on eye on safety and reports on unsafe events like accidents – so the media and our own choices in response keep the airlines, and other companies, focused on safety. The second lesson is that profits are important, but not in the way you might think. Profit gives safety for the firm, but that safety does not translate into safety for its customer. On the contrary, it is the unsafe firm that is compelled to provide safety to its customers. Those are important lessons in management, and they will also make me think more carefully about the choices I make as a consumer.


For those interested in the technicalities of the 737 MAX 8 crash, here are some details (which may change as the investigation continues). The 737 is a small airplane with wings mounted low, so as larger engines became popular for their greater fuel efficiency, engine placement has been a problem. There simply isn’t much room between the engine and the ground, so the engine is mounted more forward than is conventional. With the MAX model, an additional problem is that an increase in thrust from the engine can push the nose too high because the engine is low relative to the rest of the plane. A too high nose can cause a stall, which means that the wings no longer lift and the aircraft goes into freefall. To prevent this, the MAX model has an automatic stall prevention software, which makes adjustments to push the nose down when it rises too high. This system works even when the pilot tries to lift the nose. The system relies on an accurate reading of how high the nose is (angle of attack), and if the sensor measuring the angle of attack is incorrect, the system can incorrectly push the nose down. If that happens, the pilot needs to turn the stall prevention system off and fly without it. The system is designed to override the pilot, so it is not possible to pilot against it. Currently the Lion Air flight crew are suspected of failing to turn the system off, and Boeing is suspected of failing to teach crews how to turn the system off. The Ethiopia Flight crew did turn the system off and tried to fly manually as they were supposed to, but the manual controls are extremely difficult to use during takeoff because the airflow pushes against the horizontal tail.