Listen: https://serve.buildtall.systems/0713f731ff421eda34565f738c89be0fbad91aab0a561993d6cdcdf78407ba04.mp3
Welcome to marginal revolution read out, covering the post Coffeehouses and the Scientific Revolution. In this episode, we look at a single linked article with the same title, which presents new evidence on a classic claim: that coffeehouses, as venues where strangers could meet and talk across social ranks, helped drive the Scientific Revolution. The author of the post highlights a recent paper by Monir Bounadi and Jens Oehlen, and adds that Bounadi is on the academic job market this year.
The paper's abstract sets up the question directly. Historians have long credited coffeehouses with fostering scientific contacts during the Scientific Revolution, but this paper actually tests that claim using seventeenth-century London as a natural experiment. The researchers rely on newly geocoded records and a staggered event study across parishes. Their key finding is striking: when the first coffeehouse opened in a London parish, that parish saw a persistent increase in the number of applied scientists and scientific instrument shops. The effect did not wash out quickly; it lasted over time. Importantly, the increase was concentrated among applied scientists who had weak family ties to Oxford and Cambridge, meaning people outside the formal academic networks were the ones newly drawn into scientific work. The authors also find that social connections linking craft skill to theory became more dense, and new shops shifted toward making experimental instruments rather than just traditional trade goods.
To understand the magnitude, it helps to know what an applied scientist meant in seventeenth-century London. These were not professors in a modern university sense, but often instrument makers, surveyors, mathematical practitioners, and skilled craftsmen who used theory to build tools or solve practical problems. The paper's parish-level data allow the authors to compare places that got a coffeehouse early with similar places that got one later, and to track what happened in the years after each opening. That design is the staggered event study, and it gives the estimate of a persistent increase.
The paper then expands beyond London to a sample of European cities in the seventeenth century. There, the numbers of applied mathematics titles and scientific instrument makers rise after a city's first coffeehouse, relative to cities of similar size and growth. This cross-city comparison helps rule out the idea that coffeehouses simply located in already-thriving scientific centers. If the effect disappeared once you controlled for city size and growth, the coffeehouse would look like a symptom. Instead, the relative increase after the first coffeehouse suggests something causal.
The authors go further and calibrate a quantitative urban model to their parish-level and city-level estimates. From that model, they conclude that coffeehouses account for 31 percent of London's applied scientists in the half century after the first one opened in 1652. That is a large share for a social institution with no formal curriculum, no exams, and no degree. The first London coffeehouse opened in 1652, and within fifty years, according to this estimate, nearly one in three applied scientists in the city owed their presence to the coffeehouse effect.
Why does this matter? The paper's final sentence in the abstract says it plainly: social institutions can raise the production and use of knowledge by broadening access to intellectual exchange beyond formal academic networks. Coffeehouses were not schools or laboratories, but they lowered the cost of conversation between people with different skills. A craftsman who knew how to build lenses could meet a scholar who knew optical theory, and that combination produced new instruments and new applied knowledge. The evidence that the effect is concentrated among people with weak family ties to Oxford and Cambridge supports this mechanism: the coffeehouse mattered most for people who did not already have elite academic connections.
The author of the Marginal Revolution post does not add much beyond pointing to the paper and noting that Bounadi is on the job market, but that pointer itself is a signal that this is fresh, potentially important research. The paper is coauthored by Jens Oehlen, and the combination of detailed geocoded parish data with a staggered event study design gives the findings real credibility. The 31 percent figure is large, and the cross-European replication makes the story broader than one city.
One interesting tension in this literature is whether coffeehouses were themselves a cause or a symptom of urban intellectual ferment. This paper tries to address that with the staggered timing of coffeehouse openings across parishes and cities, and with the comparison to similar cities. The use of an event study means the authors can look at what happened before and after each opening, not just overall correlation. The persistent increase after the first coffeehouse, and the shift of existing shops toward experimental instruments, suggests a causal channel rather than mere sorting. If coffeehouses had merely followed scientists, you would not expect the shops to change what they made after the coffeehouse arrived.
For listeners interested in economic history, the paper connects to older debates about the Republic of Letters and the role of informal institutions in knowledge creation. It also speaks to modern discussions about coworking spaces, online forums, and other venues where strangers with different expertise meet. The core idea is portable: if you lower the cost of talking across disciplinary boundaries, you may get more applied innovation. That is the through-line of this post: a single link, but one that packs a lot of intellectual history and careful empirical work.
That wraps up this episode of marginal revolution read out. We covered one article, Coffeehouses and the Scientific Revolution, showing how seventeenth-century coffeehouses increased applied scientists and instrument making, especially among those outside elite academic networks. The author's pointer to Bounadi and Oehlen's job market paper suggests this is research worth following as it moves toward publication.
