History and Economic History

Can historians and economic historians understand each other? That was the subtext of a panel discussion on “Number Trouble” at last weekend’s meeting of the American Association for Slavic and East European Studies (ASEEES). On the panel were some of my favorite historians of Imperial Russia: David Darrow, Tracy Dennison, Steve Hoch, Yanni Kotsonis, and Katia Pravilova. In the audience were Evgeny Finkel, Amanda Gregg, Martin Kragh, Steve Nafziger, and I—economists and political scientists working on late Imperial and early Soviet Russia. And we talked.

Many historians, I believe, look at social scientists doing history and think: You don’t know where these data came from, you don’t know why they were assembled, and you don’t understand what they meant to the folks at the time they were collected. And yet you insist on sticking these variables into your econometric models and drawing inferences.

Many social scientists, in turn, think: You assume that because the data are noisy we cannot use fancy methods, yet it is precisely because the data are noisy that the methods must be fancy. Done properly, econometrics provides a means of bounding the bias, of asking whether the qualitative conclusions are sensitive to varying assumptions about the data-generating process.

What is to be done when both sides are right? Social scientists will rarely have the nuanced understanding of sources that historians do. At the same time, graduate training and disciplinary incentives imply that historians who truly understand social science will be rare exceptions. Gains from trade will be hard to realize.

One possibility, suggested Steve Hoch, would be for social scientists and historians to coauthor with each other. I like the idea, and I can think of a few good examples, but disciplinary differences imply that such collaboration will be the exception, not the rule.

Maybe we just need to take more opportunities to talk with each other. Martin Kragh suggested after the conference that we should organize panels of historians with social scientists as discussants, and vice versa. My colleague Giuliana Chamedes and I have been trying to do something similar here at UW, with a regular seminar series in History and Politics that draws speakers and audiences from both disciplines. We’re still talking past each other as much as we are talking with each other, but I do think we are at least starting to get used to each other.

One Hundred Years

One hundred years since a small group of extremists seized control of a major European state, thus launching a civil war, collectivization, terror, the complete reorganization of economy and society, and a geopolitical standoff that could well have ended in human extinction—all in the name of an untested and ultimately incorrect theory of human nature. When Keynes wrote of “Madmen in authority…distilling their frenzy from some academic scribbler of a few years back,” he must have had the Bolsheviks in mind.

You’re sure this is going to turn out OK?

In commemoration of this catastrophe, the Slavic Review—the journal of the Association for Slavic, East European, and Eurasian Studies—has published a special issue on the Russian Revolution. Evgeny Finkel, Dmitrii Kofanov (Ph.D. student at UW), and I chime in with a short note on peasant unrest from March to October 1917. Disturbances in the countryside were an important factor in the rapidly evolving political situation between the February and October Revolutions, but most academic work focuses instead on what was happening in the cities. There certainly has been little attempt to systematically examine the conventional wisdom that peasant unrest in 1917 was driven by demand for land and freedom (zemlia i volia). That’s what we do in this paper, using two province-level counts (one from the Provisional Government, one from the 1980s) of disturbances during the Russian Revolution.

What do we find? Land matters, but good land matters most: peasant disturbances were most pronounced in provinces with relatively good soil. And “freedom” matters, but it’s arguably the freedom of former serfs and their descendants that is most salient, as unrest in 1917 was concentrated in regions that sixty years prior were populated with serf estates. (On the eve of emancipation in 1861, approximately half of all peasants were serfs; most of the rest lived on state lands.) Peasants, in short, acted to seize good land and to burn down the manor house, where that existed.

Or so the geographic patterns documented in our note suggest. Unfortunately, much of the underlying event data seem to have been lost to history, so the sort of analysis that Evgeny and I do in our work with Tricia Olsen and with Paul Castañeda Dower and Steve Nafziger doesn’t seem to be possible. But it’s a start—more evidence, at least, than the madmen of 1917 had to justify their actions.

Technology Made Us Slaves, Technology Will Set Us Free

I’m back. Vacation turned into the mad rush before the start of the semester turned into the actual start of the semester turned into…Anyway, it seems as good a time as any to return to the theme of distractions and how to manage them. My friend John Ahlquist alerts me to a piece about tech innovators going off the grid. It’s the latest installment in an escalating backlash against the smartphone and its unfair demands on our scarce attention. I touched on this earlier this year when discussing Cal Newport’s excellent book Deep Work. Suffice it to say that I am sympathetic to the idea that smartphones are draining our cognitive reserves and making it difficult to get any real work done. Nonetheless, I am reasonably confident that we are not permanently mired in some Zuckerbergian dystopia. Let me explain why.

The costs imposed by smartphone (ab)use are not just personal. They are imposed on numerous businesses, whose workers are distracted by the device on their desk even when the screen is off. Sooner or later, employers are going to figure out that they can get more out of their employees if the smartphone is stowed away, just as instructors have learned that they can get more out of their students if electronic devices are banned from the classroom. (Those who are forced to go cold turkey eight hours a day may find it easier to abstain at home as well.)

Now, leaving behind your smartphone isn’t yet practical for many of us, who rely on the damn thing to stay in touch in the myriad ways that make knowledge work possible. What is needed is a way to stay on the grid without getting sucked in. Those who are especially disciplined can remove the social-media apps, or even the browser, from their phones. I suppose employers could demand the same, but I don’t know many people who would be happy to work for somebody who demanded that they cripple their shiny new device. Somehow it seems easier just to be told you can’t have it.

What the market will reward is a crippled smartphone that people feel happy using and that employers feel comfortable allowing. Something, say, that you can wear on your wrist. Something that can tell you that a file has been uploaded to Slack or that your child’s school is closing early due to a snowstorm. A device from which you can make an emergency phone call or even respond to a text message. Maybe even something that tracks your movement to help you to stay in shape. But definitely not something that you would want to stare at for hours on end.

Come to think of it, we have these devices already. And they are just getting capable enough to replace a smartphone for short periods. The workplace of the future may ban smartphones but allow smartwatches. Sound fanciful? Maybe. But when was the last time you saw a doctor or nurse with a smartphone in her hand? I personally can’t remember one—but I do see lots of medical staff wearing Apple Watches and Fitbits.

The economic costs of constant distraction are too high to be absorbed indefinitely. And Silicon Valley is too smart not to provide a solution with so much money on the table.

(Not) Correcting Acemoglu and Robinson

When is an error not really an error?

Daron Acemoglu and Jim Robinson recently posted a correction to the key proposition in “Why Did the West Extend the Franchise? Democracy, Inequality, and Growth in Historical Perspective,” the seminal paper in what has proven to be an enormously influential research enterprise. That proposition characterizes equilibrium in terms of the parameter q, which measures the probability of future unrest in an undemocratic regime. When q is large, then promises of future redistribution are fully credible and democratization is unnecessary, whereas when q is small the elite democratize to prevent revolution the first time that the poor pose a credible threat of unrest.

Those results still hold in the corrected proposition, but it turns out that for intermediate values of q, the unique equilibrium is in mixed strategies: the elite democratizes with probability strictly between zero and one, and revolution occurs on the equilibrium path. Technically, this correction is driven by a failure in the original analysis to check for all possible deviations. Substantively, the issue arises because institutional change in the Acemoglu-Robinson model is treated as a discrete choice: democratize/not. This discreteness implies that democratization, when it takes place, leaves the poor with strictly more than their payoff from revolution, thus creating scope for the deviation that Acemoglu and Robinson discuss in their correction.

Models are abstractions from the real world, and at first blush treating democratization as dichotomous seems like a reasonable stylization. In reality, of course, elites choosing to surrender political power to some previously excluded group have a wide range of options available, such that we can think of “liberalization” as a continuous choice—any level of representation between 0 and 1. And generalizing the Acemoglu-Robinson model in this way resurrects the original insight.

Paul Castañeda Dower, Evgeny Finkel, Steve Nafziger, and I recently developed just such a generalization of the Acemoglu-Robinson model, by way of setting up the empirical work in our forthcoming APSR article on “Collective Action and Representation in Autocracies: Evidence from Russia’s Great Reforms.” As we show in an accompanying note, the equilibrium level of representation granted by the elite in our generalization leaves the excluded group with precisely their payoff from revolution, thus eliminating the incentive for the elite to deviate in the manner that Acemoglu and Robinson identify.

The key empirical prediction of “Why Did the West Expand the Franchise” thus emerges intact from a generalization of the Acemoglu-Robinson model to allow for a continuous institutional choice. Moreover, the prediction itself generalizes. Not only does the elite not “liberalize” when the excluded group poses a frequent threat of unrest, but conditional on some representation having been granted, the equilibrium level of representation is decreasing in the probability of future unrest. This is the prediction that we empirically examine in “Collective Action and Representation in Autocracies.”

What Trump Knows (apropos of Korea, Putin, and Charlottesville)

Donald Trump is a little man who knows one big thing: the power of political rhetoric lies in not sounding like other politicians. He is not the first to understand this—Bill Clinton won the Democratic nomination in 1992 by playing against (1980s Democratic) type—but few have honed the tool to such perfection.

During the presidential campaign, this furthered two goals. First, by saying what no other Republican candidate would (that the Iraq War was a disaster), or even what no decent American would (that John McCain was not a war hero because he was captured), Trump could signal that he was not a typical politician. Second, he could convey to key constituencies that he would provide more than lip service to their causes. Unseemly to release a list of potential Supreme Court nominees? Unheard of to attack a federal judge for his ethnic heritage? Unprecedented to call for a ban on entry into the country based on religion? Well, that was precisely the point.

As the saying goes, once you have a hammer, everything starts to look like a nail. And so three times in in the past five days Trump has gone out of his way to sound unpresidential. On Tuesday, responding to an escalating crisis with North Korea, he seemingly threatened a nuclear attack with his "fire and fury" comments. On Thursday, he responded to Vladimir Putin's order that the U.S. slash its embassy and consulate staff by expressing gratitude, saying that "we're trying to cut down on payroll." And today, after white nationalists brought bigotry and violence to Charlottesville, Virgina, Trump condemned the "hatred, bigotry, and violence on many sides—on many sides" (emphasis in the original: watch the video).

In each case, Trump left just enough room that one could explain away his comments, if so desired. "Fire and fury like the world has never seen" need not actually refer to nuclear weapons. He might have just been joking when he thanked Putin for helping the U.S. government to save money. And he did tweet that "We ALL must be united & condemn all that hate stands for."

Nonetheless, those accustomed to norms of presidential rhetoric heard the discordant notes in Trump's speech—and if they didn't, the media let them know. And so Trump signaled to his base that he is "tough," even if the North Korean leadership may be getting a different message. Simultaneously, he assured Putin that he is "understanding," for reasons that Bob Mueller is presumably investigating. And to white supremacists…well, just read what the white supremacist website Daily Stormer had to say: "Trump's comments were good. He didn't attack us. He just said the nation should come together…When asked to condemn, he just walked out of the room. Really, really good. God bless him."

My guess is that Trump will eventually say something approximately like the right thing about Charlottesville, as when during the presidential campaign he eventually disavowed former KKK leader David Duke (who said Friday in Charlottesville that "We're going to fulfill the promises of Donald Trump" to "take our country back"). This will constitute renewed permission not to break with the president, for those who desire that. But he will have already made his allegiances clear.

A Good Conference

What makes for a good conference? The opportunity to see old friends and make new ones. Quality panels with work that challenges and crosses intellectual boundaries. Outstanding plenary sessions. Pleasantly situated receptions and dinners.

And what sort of conferences satisfy these criteria? To my mind, those that are small enough that everybody can fit into a banquet hall, that bring together adjacent disciplines, and that are held in interesting places. These are the meetings that people mark on their calendars a year in advance.

I have a couple of conferences that fall into this bin, and I’m just back from one of them: the annual meeting of the Society for Institutional and Organizational Economics (SIOE), formerly known as the International Society for New Institutional Economics (ISNIE). I think of this as the conference that brings together anybody who can trace a line to the Nobel Prize winners Ronald Coase, Douglass North, Elinor Ostrom, and Oliver Williamson. That’s a big chunk of three disciplines—economics, political science, and law—with smaller parts of several more.

This year’s SIOE meetings were held in New York, as Columbia’s Bentley MacLeod is the incoming president. My friend Sergei Guriev is the current president, and he gave an outstanding address at the gala dinner on inequality and well-being in economic transition—the subject of this year’s report of the EBRD, where Sergei serves as chief economist. (Spoiler alert: The postcommunist transition had an impact on average height—a common measure of developmental stress—analogous to being in a conflict zone.) We had keynote addresses by Matt Jackson (“New Developments in Network Economics”), Duncan Watts (“Computational Social Science: Exciting Progress and Future Challenges?”), and last year’s Nobel Prize winners in economics, Oliver Hart and Bengt Holmström (“Assets, Contracts, and Organizations—Past, Present and Future”).

Ownership ties among Ukrainian oligarchs circa 2004 (preliminary)

My own panel was SIOE in microcosm. Organized by Dean Lueck of Indiana’s Ostrom Workshop, we had four papers, spanning two disciplines, on governance and institutions. I presented work on “oligarch networks” in Ukraine with my longtime collaborator John Earle, Anton Shirikov (an outstanding Ph.D. student in political science at UW), and Solomiya Shpak (an equally outstanding Ph.D. student in public policy at George Mason). We are interested in how powerful businessmen in countries with weak institutions balance the desire to obscure ownership in key enterprises (the better to protect them from competitors and the state) with the need to maintain control over those assets. Ukraine is a rich setting in which to explore this question, with an abundance of both oligarchs and data.

I was joined on the panel by my UW colleague Nick Parker, who is exploring the impact of the fracking boom on an increasingly scarce resource: fine-grained silica sand. It turns out that Wisconsin is home to much of the sand used in the industry, and all of that extraction—and the associated transportation of sand by truck along rural roads—has led to battles with local regulators. Gustavo Torrens of Indiana followed with a revisionary look at the American Revolution. If the American demand was no taxation without representation, he asks, why not grant representation to avoid war? The answer lies in British politics—moving the political locus of the revolution from the New World to the Old. Finally, Werner Troesken of Pittsburgh discussed the origins of municipal-segregation laws in the early twentieth century in the American South. Contra the conventional understanding that increased housing demand by African Americans was responsible for such laws, Werner argues that it was the inability of whites to enforce segregation through vigilantism that motivated city councils to pass discriminatory ordinances.

A bit of this, a bit of that—but we’re all studying institutions and organizations. Sound like a good conference? Mark your calendar for next year’s meeting: June 22–24 in Montreal.

Game Theory and Medicine

It’s been radio silence for a few weeks as I have raced to beat various deadlines. The last of these lifted as I gave a talk on Thursday in the Internal Medicine Department at the University of Iowa. This was, needless to say, my first presentation to an audience in scrubs. I owe the rare privilege to my brother Brian, who is a faculty member at Iowa in internal medicine (pulmonary, critical care, and occupational medicine) and neurology.

What can a social scientist say that would be interesting, and on topic, to a group of physicians? Well, strategic interaction is everywhere, and the folks at Iowa were interested in learning some game theory. There are limits to what one can do in an hour, but we managed to cover the basics and to talk through a few applications, including to vaccination. The decision to vaccinate—oneself, or one’s children—is typically characterized by strategic substitutability: the more that others vaccinate, the stronger is the incentive to rely on herd immunity and not bear the (real or perceived) cost of vaccination. Equilibria will consequently be “interior,” with some but not all individuals choosing to vaccinate. Whether this is sufficient to eradicate a disease—well, you should check out my slides.

Relying on the herd

Putting this together was a lot of fun. I learned some medicine (including the morning of the talk, when I got to hang out with the lung-transfer team) and also some network games. (A big shout-out to Matthew Jackson and Yves Zenou for their excellent paper on “Games on Networks” and to Daron Acemoglu and Asu Ozdaglar for some exceptionally clear notes on network effects.) And, for a day, I got to pretend that I was a real doctor.

Is Trump Tipping? An Update

In what seems like eons ago, but in fact was early March, I posed the question: Is the Trump presidency at a tipping point? My conclusion then was that it was easy—perhaps too easy—to tell stories in which actors have incentives that lead to tipping points. Let’s not get carried away, I argued: one could just as easily tell stories that work the other direction.

Still, for the sake of argument, I laid out four tipping models, each involving a set of actors on whom the Trump presidency depends. Somewhat to my surprise, my imagination proved to be a pretty good guide to how things would play out over the last few months. Let’s take stock:

Disaffected insiders: This is a classic case of safety in numbers: it’s harder to get caught leaking if everybody around you is also talking to the media.

Um, that seems right. At this point, I wouldn’t be surprised if Ivanka is trash talking the president off the record. It’s the most transparent administration in American history.

The media: No legitimate news organization wants to get scooped. The more news organizations are investigating a story with legs, the more that others want to jump in.

Yep. The Times and the Post in particular seem to be in a race to win the Pulitzer for the Trump-Russia story.

Republican policy experts: Well-connected Republicans smell scandal and failure, discouraging them from taking positions in the Trump administration. This in turn discourages other potential nominees from seeking out such positions.

Is there anybody in DC who hasn’t already taken a job in the Trump administration who wants one now? As David Brooks writes, “even if the Trump presidency survives, it will be staffed by the sort of C- and D-List flora and fauna who will make more mistakes, commit more scandals and lead to more dysfunction.”

Republican members of Congress: Trump is still quite popular among Republican voters, so cutting loose the White House—say, by calling for an independent investigation of Trump-Russia ties—poses a political risk. That risk, however, may be smaller if other Republican members take the same position.

Yes, well…I think the most we can say is that there is a conspiracy of silence. But let’s check back on the GOP in another few months.

Second Pancake (on Ukraine after the Euromaidan)

A big shout-out to Grigore Pop-Eleches and Graeme Robertson, who organized a great conference at Princeton on Wednesday. It was the latest opportunity for scholars and policymakers to come together to discuss the future of Ukraine, which—I don’t think this is an overdramatization—is at a critical juncture in its history as an independent nation.

There is a saying in Russian: первый блин всегда комом, which translates roughly as “the first pancake is always a flop.” Novel endeavors often fail—but the second attempt may be successful. Postcommunist Ukraine’s first attempt at regime change was certainly a flop: the Orange Revolution of 2004 resulted in a change of government but little real reform. The Euromaidan Revolution of 2014 (aka the Revolution of Dignity), in contrast, has the potential to set Ukraine on a new path.

The difference is in Ukrainian society. Ukrainians learned from their first revolution that it’s not enough to replace those in power. Power corrupts, and power attracts the corrupt. Government turnover is therefore insufficient to produce real change. If you want to transform the political culture, you need to organize to monitor those in power.

Civil society at work

Ukraine today has a vibrant mix of NGOs, watchdog organizations, independent media, and other organizations dedicated to holding officeholders accountable. Credit for unveiling Paul Manafort’s corrupt dealings in Ukraine, for example, goes to Serhiy Leshchenko, a prominent investigative journalist and member of parliament (and participant in the Princeton conference). Major policy achievments include an asset-declaration law with teeth and a world-class electronic procurement system.

Inexperienced cooks may be hesitant to push on after the first pancake, wondering if the problem is with the pan. If Ukraine is to succeed, it is because members of the Ukrainian public drew a different lesson from the failed Orange Revolution. The future of Ukraine is in their hands now, not those of their political leaders. Let us hope that it is enough for the second pancake to come out right.

Going Deep

Consider me a convert. After listening to Ezra Klein’s interview with Cal Newport, author of the popular Study Hacks blog, I read Newport’s Deep Work: Rules for Focused Success in a Distracted World. The book is not written specifically for academics, but I haven’t read anything better about how to be one.

“Deep work” is cognitively demanding work requiring intense concentration. It is the sort of work that academics get paid to do, yet often find hard to accomplish because of the myriad other demands of the job. As with anybody who has been an academic for awhile, I have developed my own methods of going deep—parking myself where others can’t find me, leaving email to the margins of the day, that sort of thing. By and large, they have worked. Reading Newport’s book however, I realize that I could have been doing more deep work, and in less time.

I lived in central Europe for three years before moving to Berkeley for graduate school. When I left Budapest, I shipped 23 boxes of books back to the States. (Of these, nineteen ultimately arrived. Somebody in the Hungarian postal system has my annotated copy of The Wealth of Nations.) I read a lot during those three years, even as I worked often long hours at an NGO in Prague and in a research position at CEU.

The secret, I now understand, is that I quit work for the day when I left the office. I didn’t have my own computer, so there was no way to answer email (such as it was in 1997), clean data, revise a paper draft, or do anything else directly work-related from home. More often than not, I just picked up a book.

All of that changed when I got to graduate school; I have never since had a “normal” work schedule. I don’t bring work home so much as I never leave it. Through the miracle of Dropbox, my home machine looks exactly like my office one; my smartphone provides a portal to my job wherever I am. And so I work, even when I am not at work.

This is crazy, says Newport. If you bust ass during the day, if you close the door when you need to concentrate, if you minimize non-essential tasks—if you do all of that, there is no reason you cannot log off at 5 pm and go home. And before you say, that would never work in my job, please note that Newport is a (successful) computer scientist at Georgetown.

The social science behind Newport’s advice is straightforward: hard budget constraints promote efficiency. If you know that the workday ends at 5 pm, then you make the most of the hours that you have. And because you are not burning the midnight oil, you have the energy to work at a high level of intensity when you are “on.” Paradoxically, it may be easier to maximize hours of deep work when there are fewer hours in the workday.

The hard part, of course, is credibly committing to keeping the laptop closed after dinner and story time. But in my first week adhering to this schedule, it hasn’t been quite as hard as I expected. Reading more regularly means that I want to get back to the book I left on the bedside table the night before. At the same time, I know that if I take the evening off, I will wake up in the morning eager to get back to the deep work that I left at the office.

The Internet giveth, and the Internet taketh away. The ability to transmit data to anyone, anywhere, at zero marginal cost has created unparalleled opportunities for collaboration. Dropbox, Skype, Slack, ShareLaTeX—complex research projects with multiple investigators were far more challenging before we had such tools.

But the same revolution in communication has given us Twitter, Facebook, and continuously updated news sites. I have heard that submissions to political science journals are down since the November election. If true, this is testament to the ability of the Internet to distract us from productive activity. Speaking personally, I have found it especially hard to ignore Twitter on big news days—and lately, it seems like every day is a big news day.

Yes, following politics is part of the job description for a political scientist. But it is a slippery slope, and I know that I have slipped. The only time my job requires that I have up-to-the minute information is when I am doing media or a public event—and I don’t do a lot of either. The rest of the time, the firehose part of the Internet is a distraction. And so, I have decided to adopt Newport’s recommendation to remove the Twitter and Facebook apps from all of my devices. I still have the accounts, but to access them I need to log in from my web browser. That’s a speed bump just large enough to discourage frequent use of social media.

(Yes, I appreciate the irony that any readers of this post who follow my lead will be less likely to learn of future posts through Facebook or Twitter.)

The previous two recommendations help to ensure a productive workday free of distractions. But that workday can still be spent on “shallow” rather than deep work. To guarantee that it does not, I have implemented a version of Newport’s practice of keeping track of the number of hours devoted to deep work. On my watch, I now have a simple clicker app, which I advance by one every time I log 30 minutes of deep work. At the end of the week, I record the total and reset the clicker. Human psychology should encourage me to try to beat last week’s mark, creating a powerful incentive to maximize the time available for deep work.

Enforcing a hard end to the workday, quasi-disconnecting from social media, keeping a tally of hours engaged in cognitively demanding effort—taken together, these represent a fairly sharp departure from how I had previously organized my day. I’ll see how it goes and report back. In the meantime, what works for you? Have you read Cal Newport’s book? How do you avoid distraction and go deep?