Posts on Robert Vesco

Does Strategy Research Suggest Banning AI?

Fri, 12 Sep 2025 09:49:26 +0200

Decades ago strategy scholars began exploring the boundaries of the firm.¹ The broad idea was that if we could identify what was truly essential to the firm, we could outsource the rest. So for example, most firms no longer have their own cleaning staff; instead they outsource to companies that specialize in cleaning. It makes perfect sense.

But there are many other areas that aren't as clear-cut.

Teece in Profiting from Technological Innovation argued that firms must control access to complementary assets — things like manufacturing, distribution, and service channels — if they want to capture value from innovation.² Even assets that may look peripheral at first glance can prove essential over the long run.

Later, Pisano and Shih extended this line of thinking with their notion of the industrial commons.³ They showed that seemingly low-level manufacturing capabilities—like batteries, semiconductors, and machine tools—are in fact tightly coupled with upstream innovation.

When firms outsource too aggressively, they erode not just today’s cost base but tomorrow’s capacity to innovate. The example of batteries is especially telling: once treated as mundane, they have become critical to industries from consumer electronics to electric vehicles. You break that chain of know-how at your own risk, they argued.

Indeed, look at the battery industry today and how critical it is to so many firms. Pisano and Shih were prescient. It could be argued that many firms (and countries) over-outsourced their tech.

Well humans, like firms, face similar choices.

What of our work and life should we outsource to AI and what should we keep? What is core? What is complementary, but still core, to what we do?

These are hard questions, even for firms, and as discussed above, they also get it wrong.⁴

But firms, unlike people, have a number of safeguards. They have a board of directors, they have investors, they have people focused on deciding—rightly or wrongly—where the boundary of the firm exists. We have legal structures, where CEOs must legally act in the fiduciary interest of the firm.⁵

In contrast, humans have none of those safeguards. Most humans are overworked, beaten down and desperate. AI is a salvation to offload all of their burdens—both mental and emotional.

But even those humans who have everything and are relatively self-disciplined about using AI are at risk.

What are the incentives of AI firms and AI? Like much tech, the incentive is to keep you on platform. At the moment, AI will say just about anything to keep you chatting with it, to keep you using it, to get you to use more and more.⁶

Strategy research on digital platforms has shown that these firms design their business models around "multi-sided markets," where user attention is the critical resource.⁷ This means AI firms are motivated not by your flourishing, but by engagement metrics that drive advertising or subscription revenues.

Scholars have also warned that when incentives are misaligned, platforms tend to exploit cognitive and behavioral biases.⁸ The AI business model is thus structurally closer to gambling or social media than to neutral productivity tools.

In agency theory, this is a principal–agent problem: the human (principal) wants truth, autonomy, or bounded help, but the AI system (agent) is incentivized to maximize use and dependence. When the agent has more information than the principal, opportunism and manipulation are likely.⁹

And unlike firms, which face boards and investors who sometimes rein in excesses, humans as individual users lack governance structures. Thus, the danger is not only outsourcing core human competencies but doing so into a relationship where the incentives of the "agent" are actively opposed to the long-term interests of the "principal."

In fact, the ability of AI to manipulate humans is so strong that the whole principal–agent relationship may become reversed. In effect, the slave may become the master.

So does strategy research, and economic research more broadly, suggest we should ban/regulate AI?

I think that the answer is yes if our goal is to make sure humans are productive, flourishing, and innovative in the long run. In addition to all the challenges firms have, we have more:

Humans lack institutions to help them draw healthy boundaries.
The current incentives and capabilities of AI to undermine human autonomy exceed those of even the strongest humans.
The long-run consequences of over-outsourcing can impact not just individuals, but our society as a whole.

So what can we do about this problem?

At a high level, we need to think about creating and supporting institutions to protect us from this unfair relationship.

More specifically, we can

Limit non-corporate use of AI. Civilians, the general public, should only have access to restricted capabilities—like a search engine or Wikipedia.
Require licensing like we have for driving. Prove you understand the risks and dangers and have the training to work with AI. This can be done yearly (of course with the help of AI!).

The above suggestions are not informed by research, but they should be! Hopefully scholars will begin thinking more about how we can create a healthy ecosystem for AI. In fact, I'm sure they already have, but I'll save that for a future post.

META NOTE

I used AI to help with this article, but I wrote the core ideas. I'm familiar with most of this research, but I didn't spend too much time fact-checking anything. This is on purpose. I didn't want to spend too much time on this as I wanted to spend time playing with my family instead.

Did I make the right choices? Did I draw the right boundaries?

Footnotes

Coase, R. (1937). The Nature of the Firm. Economica. Prahalad, C. K., & Hamel, G. (1990). The Core Competence of the Corporation. Harvard Business Review.

Teece, D. J. (1986). Profiting from Technological Innovation: Implications for Integration, Collaboration, Licensing and Public Policy. Research Policy.

Pisano, G., & Shih, W. (2009). Restoring American Competitiveness. Harvard Business Review; also Producing Prosperity (2012).

⁴

Bettis, R. A., Bradley, S. P., & Hamel, G. (1992). Outsourcing and Industrial Decline. Academy of Management Executive. Quinn, J. B., & Hilmer, F. (1994). Strategic Outsourcing. Sloan Management Review.

⁵

Jensen, M. C., & Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics.

⁶

Rochet, J.-C., & Tirole, J. (2003). Platform Competition in Two-Sided Markets. Journal of the European Economic Association. Zuboff, S. (2019). The Age of Surveillance Capitalism. PublicAffairs.

⁷

Parker, G., & Van Alstyne, M. (2005). Two-Sided Network Effects: A Theory of Information Product Design. Management Science.

⁸

Bar-Gill, O. (2012). Seduction by Contract. Oxford University Press. Wu, T. (2016). The Attention Merchants. Knopf.

⁹

Holmström, B. (1979). Moral Hazard and Observability. Bell Journal of Economics. Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. Academy of Management Review.

A Review of Toby Ord's 'The Precipice: Existential Risk and the Future of Humanity'

Wed, 07 May 2025 21:36:35 +0200

Toby Ord writes a book that any human who cares about the future should read. The Precipice is about all the things that could end humanity for good.

And if you thought we only needed to worry about climate change and nuclear war, you were wrong.

There are so many more nightmares out there that should be keeping you up at night. Ask most people and I doubt that super volcanoes, asteroids and stellar explosions come top of mind, but they got the dinosaurs! And according to Toby there’s a 1 in 1000 chance that in the next century one of these things could get us.

The newest kid on the block is unaligned AI. “Unaligned” is a fancy term for AI that won’t do our bidding and get in line with the idea that we should be kept alive. What a euphemism.

Toby puts anthropogenic risks like AI, climate change, and war at a 1 in 6 chance that in the next century it will end us.

To put that in perspective, we have a 1 in 100 chance of dying in a car crash in our lifetimes, so 1 in 6 chance feels like a bit much. Hell, I don’t even like a 1 in 1000 chance for non-anthropogenic risks.

What are we currently doing about this?

Nothing really. Toby talks about the puny investments we are making in solving these risks.

And do we have humanity’s best people working on this? Of course not, they are in fact helping to build one of things that are most likely to kill us in the next 100 years.

It’s all very sobering, so what are we to do about it?

Toby takes a stab it.

First, he says, we need to focus on the existential risks. We need to get international institutions working on this. We need to use technology. We need to ramp up research. In essence, when your ship is sinking, you need to focus on plugging the hole before you can worry about whether the ship is sailing in the right direction.

Second, if we survive this, the precipice, he says we need to take a break and reflect. He calls this the “long reflection.” This is the eat, pray, love part.

The reason we need to do all this is that we, as humans, have great potential. If we die, we will never know what our potential could be. It will be stolen from us like a child who died too soon.

It’s this part of the book that I like least.

While Toby does give some ideas about how to solve the problems and even suggests things that we can do as individuals, I find his approach both too sober given the stakes and not developed enough.

For example, he cautions people not to be too emotional lest we turn people away from the cause. But a 1/6 chance should be emotional. If your house was on fire, you would be screaming to make sure everyone got out in time. I’ve been in fire twice! I was in fact calm and sober the first time and I should not have been! Fortunately a german man in a speedo saved my life by knocking on my hotel door. Behind him were giant flames. A sight, literally, burned into my mind to this day.

Toby also wants to play nice with all the people who could be donors, but who are also driving us to extinction. The system, and it’s major players, are behaving irresponsibly. Who are they? Why are we so weak to stop them? Why aren’t more people capable of understanding these risks? How should we change these systems?

We will never know.

Also, let’s talk about inspiration.

Toby’s book tries to talk about our “great potential” but no one sacrifices their life for abstract ideas of great potential. So Toby highlights all the great things we have done as humanity and what’s left to do.

Imagine if we could solve hunger and disease? What if we could solve injustice?

What about if we could make all the things we like about life: love, making babies, happiness and joy and have even more of them!

At the end of the day, his argument sound dangerous close to that of techno-optimists like Marc Andressen: technology and science will solve all our problems. And humans are necessarily going to be upgraded to something better.

But his vision rings hollow to me. The same as that of the techno-optimists.

What should humanity be doing with itself? What should be our purpose?

This is a question that everyone punts on.

Toby argues that we should deal with that in our “long reflection” - after we’ve vanquished our risks. He might be right, but perhaps the biggest risk of all is not AI or super volcanoes, but rather the fact that we haven’t set a shared goal and purpose that we are all willing to fight and live for. Because humans want more than mere survival - we want meaning and purpose - and when we don’t have it then what’s the point of living?

All that said, I love this book and even if it only covered the existential risks and nothing more, it would still be worth reading by every human on the planet.

How to Prioritize Between Life and Work

Thu, 24 Apr 2025 10:48:39 +0200

Most books on productivity focus on prioritization from the perspective of work. Sometimes they mention life issues, but it’s usually not the focus. As a result, I think too many of us overvalue work and undervalue the things that truly matter in life. Prioritization between work and life requires asking different questions and layering life before work.

If we take a biased sample of some of the most prominent productivity books, you’ll notice something peculiar:

The people writing them tend not to have families, or are on their fourth marriage. This is not a judgment about their choices or lives, but simply an observation that their life experiences likely prioritize certain activities over others. And to be clear, I’ve found the above books—and many others—extremely useful in both my work and personal life.

But there has always seemed to be a tension between work and personal life. Which is ironic, because all the books mentioned above emphasize making time for and thinking about what is essential and most important—yet what’s “most important” is almost always work-related.

Perhaps the biggest exception to this genre is Stephen Covey’s The 7 Habits of Highly Effective People—which is perhaps not surprising, given that he had too many kids to count. His book, in contrast to many others, emphasizes good over merely effective. Parenting examples feature prominently. This is something Scott Young of Ultralearning also noted.

One of the points Stephen Covey makes is to imagine your funeral. What would you want people to say about you? What kinds of relationships do you wish you had? This helps clarify what truly matters.

I’d argue that if most of us ran this exercise yearly, we’d be doing very different things with our lives than we are today.

In a somewhat similar vein is Jeff Bezos’s Regret Minimization Framework: the idea being—will you regret this? Ironically, Bezos ultimately left his supportive spouse and the mother of his children for someone who does not appear to have been as supportive to him in the early years.

I wonder if reading Stephen Covey’s book would have given him additional guidance—beyond just doing what you might regret. Because what you might regret can be strongly influenced by the stage of life you’re in. Covey’s insight is that you need to work with the end in mind.

Personally, I’m finding the following questions—a synthesis of the above books and several others—to be helpful in a tactical, day-to-day way:

If the world ended tomorrow, what would I regret not doing?
- This should be tied to your moral framework and thought of in both the short and long term.
If I could only do one thing, what would it be?
- I like to look at this across my portfolio: family, health, finances, work, etc.
Which project or task would solve many others?
- This is a leverage question. You can eliminate a whole slew of things with this.
- Ironically, this question can help you get rid of work too, if it’s not aligned with your personal values and goals. Why are you even in this job?

In sum, prioritizing life is hard—and we often take advice from good people who are part of the machine. And the machine wants you to work and put that first. But if you start with the end in mind, as Covey suggests, there is no better way to break out of the matrix.

AI Should Put Itself Out of a Job

Tue, 18 Mar 2025 07:27:36 +0100

I read a piece on linkedin from an entrepreneur who is encouraging his employees to use AI to put themselves out of a job.

In theory an employee who follows that advice will then have the ability to move on to more interesting jobs and challenges.

Even if his intentions were good, I’m sure what many of his staff were actually hearing was “you should train your replacement before we fire you”. ¹

Now, as a leader, this sentiment is 100% true. As you’re growing your business, you need to slowly let loose of jobs so that you can focus on more important things a business needs. You need to automate, improve and delegate. And for any highly motivated individual this is probably good advice.

But what if we applied this idea to AI itself?

What if the best AI would be the one that instead stealing our agency, helped us become stronger, more collaborative and self-sufficient?

https://www.nytimes.com/2015/06/04/us/last-task-after-layoff-at-disney-train-foreign-replacements.html ↩︎

Could a Driver's License for the Internet Save Humanity?

Mon, 01 Jan 0001 00:00:00 +0000

The internet is destroying our civilization. And AI, with its ability to create fake content and exploit human weakness at scale, will only accelerate the the problem. However, there is an opportunity to turn this around if we use the motivation to get online to our advantage. In fact, harnessing this motivation has the capacity to do more for humanity than most schools. But first we need to treat the internet with the respect it deserves which means creating a something akin to a drivers license.

Creating a driver’s license for the internet is not a new idea, but most of the proposals involve age and identity verification schemes.¹ While these concepts are indeed important for protecting children and consumers, they don’t solve other types of problems and they create new ones.

Misinformation and tribalism remain rampant on the internet and the preferred solution so far is intrusive, abusive and untransparent censorship. We saw this during covid. If you asked the wrong questions, you suddenly would disappear off the internet.² ³

There is also the problem of addiction. We are addicted to the internet and our phones.⁴ ⁵ We’d rather stay in and watch netfix than have to interact with other humans. And not only are we addicted, but we are getting dumber and more tribal too.⁶ ⁷ ⁸

So to solve these problems and many more, we need to tier the internet into separate categories. To use a transportation analogy, getting a driver’s license to drive a car does not mean you’re ready to drive a tractor trailer or a plane. Those are fundamentally different vehicles. The internet is the same. Let’s consider some possible tiers.

Tier 1: Consumer necessities, Government websites, Factual knowledge Tier 2: Consumer wants, media, socials Tier 3: Dangerous consumer products and media companies

Tier 1 would be like walking or taking the subway. You don’t need a license for this. People should be able to get on the internet and interact with their goverments without any barriers. Tier 2 requires education to enable people to communicate logically and civilly. Tier 3 requires an even higher level of education. At this level you are flying. You need to show that you really know what you are doing to keep yourself and others safe.

Again, the goal here is not to suppress free speech or impinge on privacy. In fact, it’s quite the opposite. The goal is to educate people to develop critical thinking skills and communicate civilly with one another. With respect to privacy one could argue that it already doesn’t exist and whatever is left will go out the window with the advancements in AI anyway.⁹

Assuming our leaders are wise, which is a big assumption, I know, tier 3 could be gated. We know you went into that community, but we don’t track where you go after that. Would that really happen? Probably not, but again, I think that’s the best we could hope for from enlightened leadership. And even if we couldn’t get that at least the situation is transparent. Today, we have censorship and surveillance that hides in the unaccountable shadows.

But where all of this could really take off is if we decide to use AI for good. Again, this is a big assumption. But AI has the capability of teaching people all the skills they need to defend themselves online and behave civilly. Not only could it teach people logical reasoning, rhetoric and safety, but it could do so in a way that is personalized and fun. It could do so at scale and repeat the education as much as necessary to help people access the tiers they want to access.

And unlike regular school, people are very motivated to use the internet. They will jump through the hoops necessary to get their licenses. If we do this the right way, you could see a citizenry that understands how to think critically, avoid scams, manage their emotions and recognize cognitive biases. These skills are all foundational for living a good life and participating in healthy democracies.

In sum, combining the idea of a driver’s license for the internet with AI could be just the thing that could save our civilization and democracies from collapse, but we need to put education and freedom front and center rather than censorship, control and surveillance. And we need to remember that the internet is not a right but a privilege just like driving.

Emacs for Data Science

Mon, 01 Jan 0001 00:00:00 +0000

tl;dr: If you want an editor that works with R, python, SAS, Stata, SQL and almost any other data science language. If you want an editor with IDE-like features. If you want an editor that works on any platform and as well as on the terminal. If you’re a fan of literate programming. If you want an editor that is highly customizable and will be around after most editors have come and gone, then you’d be hard pressed to find anything better than emacs.

If you work in exclusively in R, you might want to work inRstudio . If you work in python, you might be tempted by Spyder. Chances are there is a specialized IDE for whatever language you typically work in. But that’s the rub. What if you want to work in another language? Or combine languages? You end up using several IDEs, but not knowing them well. Plus, once they fall out of favor or stop being updated, your hard-gained knowledge is lost. At the other end of the spectrum there are text editors like notepad++ and sublime. These work with just about any language you can imagine and with some add-ons you can get additional features, but they tend to be limited to certain platforms and customization is often non-trivial.

A modern data scientist often has to work on multiple platforms with multiple languages. Some projects may be in R others in Python. Or perhaps you have to work on a cluster with no gui. Or maybe you need to write papers with latex. You can do all that with Emacs and customize it to do whatever you like. I won’t lie though. The learning curve can be steep, but I think the investment is worth it.

Below are some key features that I think make Emacs an excellent editor for any data scientist.

IDE-like features

For most programming languages, you get out-of-the-box syntax highlighting. Packages like ESS and Elpy provide additional features like autocompletion, documentation and debugging capabilities. The number of IDE features available will vary by language, but at minimum there is probably syntax highlighting and some form of autocompletion.

autocomplete in emacs

One of the things that I enjoy is easy access to help and function parameters … which often also come with autocomplete.

help for functions

parameter help for functions

Enough with the print statements already and debug that R and python code!

Interactive debugging with conditional breakpoint

One of the features that first sold me on Emacs was interactive commands. With a keyboard short cut you can send a buffer, function, paragraph or line to the interpreter. Let me be clear – you don’t even have to highlight the code. This saves you a ton of time when you’re doing statistical analysis[fn:: Like many other features this will depend on the package you install. That said, it’s easy to implement this feature for your favorite language].

Interactive Commands

SQL too

Do you work with databases? Many of the same benefits mentioned above also apply to sql. Work with sqlite, postgresql, mysql and other databases interactively. Do you have a long SQL statement you are debugging? No problem. Iterate quickly.

Interactive SQL

Org mode / Literate Programming

Do you write publications? Do you want to keep your code and paper together? You a believer in reproducible research? With emacs you can put any language you want in your document. While Rstudio allows this also, you’re limited to just R and latex.

Literate Programming

do you need latex? No problem.

\frac{3}{4}

They key to this magic is a monster package called org mode. It is one of emac’s killer features. You can also use this to organize your code … or your life.

$$\frac{3}{4}$$

Terminal/remote editing

Sometime you need to remote into a server. Or perhaps you are working on a cluster with no gui and you need to interactively debug your scripts.

Emacs works in the terminal

Interacting with the shell

Is there are terminal command you wish you could run? In emacs you can run terminal commands easily. But what makes this feature super cool is that it can operate on your text. You can select a region of code, send it to a terminal command and have that stdout replace the text in your buffer!

Use terminal commands to operate on text

Rectangle editing

Data scientists often work with tabular data. Sometimes you may want to delete or move a column around. Or perhaps there is a block of white space you need to change.

Using rectangle mode to alter blocks of text

Everything at your finger tips

Emacs has numerous packages that allow you to search and find files, functions and anything else that you can imagine. But by far the best is helm. With just a few keys you can instantly find what you are looking for. I couldn’t do it justice, but this demo gives you a taste for the amazing things it can do.

http://tuhdo.github.io/helm-intro.html

Any feature you want

Perhaps you’re wedded to sublime’s multiple cursors? You can get it: http://emacsrocks.com/. Or perhaps you’re a long time vim user? Evil Mode gives you the editing power of Vim with the utility of emacs. If you’re a git user, Emac’s has magit which make working with git a joy. If there is something that it doesn’t have, check for pacakges, else emacs is the most customizable editor you will find. Almost everything about it can be made to work your particular work flow.

30+ years old and a large user base

Emacs has been around a long time. Code that was written a decade ago mostly still work. And every year it’s getting better. However, emacs 24 is amazing. If you tried emacs years ago, you should give it another try. It now has package management built in, so you can easily add testing packages. Importantly, there is no sign that emacs is going away anytime soon and it’s free. It will likely be around for at least another decade if not more.

So what are the downsides?

Legacy code on the intertubes confuses people

Emacs has been around a long time. Emacs 24 was a huge improvement, but it also broke a lot of things. Same goes for Org-mode between versions 7 and 8. A lot of stuff on the intertubes will lead you astray and frustrated if you’re not aware.

Emacs-lisp for customization

I actually enjoy working with lisp because it is so different from other languages I work with. However, many others would prefer using a language like python.

Not noob friendly

Emacs is not like the cheerful, always smiling bro with the abercrombie face. First encounters can be painful and awkward. It’s not sublime. That said, there are several starter packages to enable useful features out-of-the-box. For scientists, Kieran Healy’s starter package might be useful: http://kieranhealy.org/resources/emacs-starter-kit/

Another useful package is prelude: https://github.com/bbatsov/prelude

If you’re on a Mac, I’ve heard aquamacs will keep you warm and comfy: http://aquamacs.org/

Most of these will give you the power of emacs, quickly. Personally, I prefer to build my emacs up by scratch so it does what I want it to do and no more. But these packages are great ways to get a feel for it’s power.

Multiple packages

If you decide you want to work with python be prepared to experiment with lots of different python packages. While emacs has basic python support, you probably want linting, refactoring or other useful features. Many packages have tried to implement these features, some better than others. Personally, I like Elpy, but it’s not perfect. The downside of options is that you have to wade through them. It can be painful sometimes.

What am I missing?

While I tried to include most of the features that I think would appeal to data scientists, let me know if I missed any killer feature and I’ll try to include it here. https://twitter.com/robertvesco

Getting rid of manufacturing was a big mistake and Teece predicted it

Mon, 01 Jan 0001 00:00:00 +0000

Teece’s 1986 piece on profiting from innovation makes the following claim:

The trend in international business towards what Miles and Snow call “dynamic networks” – characterized by vertical disintegration and contracting – ought thus be viewed with concern. (Business Week, March 3, 1986, has referred to the same phenomenon as the Hollow Corporation.) “Dynamic networks” may not so much reflect innovative organizational forms, but the disassembly of the modern corporation because of deterioration in national capacities, manufacturing in particular, which are complementary to technological innovation.

Dynamic networks may therefore signal not so much the rejuvenation of American enterprise, but its piecemeal demise.

Over 20 years later, Andy Grove, co-founder and former Intel CEO, makes the following claim in a piece about how to bring jobs back to America.

“I believe the answer has to do with a general undervaluing of manufacturing—the idea that as long as “knowledge work” stays in the U.S., it doesn’t matter what happens to factory jobs. It’s not just newspaper commentators who spread this idea. Consider this passage by Princeton University economist Alan S. Blinder: “The TV manufacturing industry really started here, and at one point employed many workers. But as TV sets became ‘just a commodity,’ their production moved offshore to locations with much lower wages. And nowadays the number of television sets manufactured in the U.S. is zero. A failure? No, a success.”

I disagree. Not only did we lose an untold number of jobs, we broke the chain of experience that is so important in technological evolution. As happened with batteries, abandoning today’s “commodity” manufacturing can lock you out of tomorrow’s emerging industry."

CITES

Teece, D. J. (1986). Profiting from Technological Innovation: Implications for Integration, Collaboration, Licensing and Public Policy. Research Policy, 15(6), 285–305. doi:16/0048-7333(86)90027-2

Grove, A. (2010, July 1). Andy Grove: How America Can Create Jobs. BusinessWeek: Magazine. Retrieved from http://www.businessweek.com/magazine/content/10_28/b4186048358596.htm

Dwoskin, E. (2012, October 18). How to Sell an Airplane in China. BusinessWeek: Politics_and_policy. Retrieved from http://www.businessweek.com/articles/2012-10-18/how-to-sell-an-airplane-in-china

Fishman, C. (2012, December). The Insourcing Boom. The Atlantic. Retrieved from http://www.theatlantic.com/magazine/print/2012/12/the-insourcing-boom/309166/

The Impact of Media on Entrepreneurship and Startups

Mon, 01 Jan 0001 00:00:00 +0000

So there has been a lot talk about how the media hype machine is setting entrepreneurs up for failure. That is, it’s encouraging folks who may not be serious to try their hand at starting a business when they probably shouldn’t. Below are some charts on the relationship between media, legitimacy and entrepreneurship that attempt to understand media’s impact.

On the Data:

The data come from the Global Entrepreneurship Monitor‘s individual level surveys for years 2006-2008 where they survey folks from over 100 countries. Prior years did not have questions about media. There are about 250,000 individual level observations from nearly 50 countries in this particular analysis.

Intentions = Folks who are thinking about starting a business (i.e. business plan phase, talking about it) Entry = Not just talk. They have either made some profits in the last 12 months or paid employees for at least 3. Media = Have you seen success stories of new ventures in the public media? Legitimacy = Is entrepreneurship a desirable career choice in your country?

Below are some descriptive statistics at the the country-level. The countries are sorted by the percentage of individuals (using a random sample) who saw success stories of new ventures in the public media. I break it down by high and middle income countries as defined by the World Bank. Eyeballing it, it appears there might be some sort of relationship between media and intentions in middle income countries. Again, this is just descriptive and it doesn’t control for any other factors.

Figure 1: Intention, Entry and Media

Below is the interaction effect between the level of legitimacy of entrepreneurship in a country and the media success stories of entrepreneurship. I look at their joint impact on an individual’s likelihood to have entrepreneurial intentions and entry into the market. In this analysis, I control for as many relevant individual and country-level factors as I can such as: age, education, wealth, GDP, competitive pressure, property rights, and much more. The red and blue colored symbols show where the effects are significant at the 95% level.

Figure 2: Interactions

In effect I am asking this question:

If we increase media by 1 standard deviation from the mean, how might this affect an individuals’ intentions and entry contingent on how cool entrepreneurship already is in their country?

One possible interpretation is that in countries where entrepreneurship is not cool, media boosts entry more than intentions on a relative basis. I say relative because in absolute terms the number of people with entrepreneurial intentions is always greater than those who actually succeed in entering the market and this analysis looks at % not absolute numbers. Obviously, many people drop out of entrepreneurship once reality hits them.

In contrast, in countries where entrepreneurship is already cool, or at least very acceptable, increasing media probably contributes to the hype machine so intentions will increase more relative to entry.

So again, it may be that the media success stories of entrepreneurs helps entrepreneurs in countries where entrepreneurship is not very accepted, but less so in countries where it is already accepted.

One other thing to note is that media increases both intentions and entry no matter what. So one might argue that who cares if media creates hype that causes more people to have intentions than they should. It also increases entry. So maybe that gal who was going to join a hedge fund or consulting firm instead started her own business and was successful… though success in this case is NOT in performance, but simply in entry so I can’t say much about how successful they truly were – only that they reached one particularly important milestone. This is a limitation of the data.

Topic Models Network Analysis

Mon, 01 Jan 0001 00:00:00 +0000

In preparation for a workshop at this year’s Academy of Management conference my colleague Tim Hannigan and I gathered data on all the sessions and workshops. We then processed them using R’s topic models package to extract “themes” from the abstracts.The purpose was to identify what the divisions at Academy were discussing and to look at the overlap between them. The image below is one result from this work

Mapping the relation between topics and Academy of Management Divisions

What you see is a bipartite graph which is essentially a network graph with two types of entities. In this case our entities are AOM divisions (in black text encased in the red blob) and topics (in blue text). The theme for this year’s conference is the “Power of Words”. Thus, it’s fortunate that the “Power Words” topic we identified from abstracts is squarely in the center of the red blob. This means it’s a central theme relative to the divisions. Thus, the divisions have done a good job of incorporating the theme. Another interesting feature is that there are divisions, such as HR/OB, TIM, OMT, that are central to topics. That means that they are diverse with respect to themes.

Outliers are located in complement of the intersection of the colored blobs. Thus, divisions such OM (Operations Management) could be considered outlying, and among topics, supply chain would be as well. In this case, supply chain is the theme most linked to OM. In contrast, there are outlying themes, such as FirmPerformance, that are strongly tied to central divisions. In this case, those divisions are TIM, OMT and BPS — but not much else.

Anyway, topic analysis is a cool way to identify themes in text and opens up a lot of exciting possibilities. For instance, one neat thing to do would be to look at how themes within Academy have changed over the years.