The Economics of Friendship

“The ferry crossing from Chappaquiddick to Edgartown, Mass., is only 527 feet long. But no matter how rich or important you are, you have to wait to get on the boat. Sometimes you wait in your car. Sometimes you get out. Someone complains about the line, the weather, our politics or the UPS driver who goes to the front of the queue. Someone else laughs. A conversation begins.

 

This isn't a defense of bad maritime logistics.

 

But spending time around that ferry has made me wonder if modern life has undersold inconvenience. We treat waiting as wasted time. Often it is. But sometimes waiting does something useful: It forces people into the same place, with nothing to do, long enough for conversation to begin.

 

That may sound trivial. It isn't. Loneliness has become one of the stranger problems of modern life. In 2023 the U.S. surgeon general called loneliness and social isolation a public-health problem. About half of American adults have reported experiencing loneliness. The World Health Organization describes social disconnection as a global health challenge, affecting roughly 1 in 6 people worldwide. In several Western countries, self-reported loneliness among young adults appears to have risen over the past four or five decades; young adults report more loneliness than middle-aged or older adults.

 

In 2010 research, Julianne Holt-Lunstad and her co-authors pooled evidence from 148 studies and found that weak social relationships predict mortality at a magnitude comparable to smoking and greater than obesity or physical inactivity. Loneliness has been linked to higher risks of cardiovascular disease, dementia, stroke, depression, anxiety and premature death. These estimates aren't from randomized trials, and loneliness isn't a pill with a known dosage. But they make one intuitive point hard to escape: Social connection is part of a healthy life.

 

Which brings me back to that ridiculously inconvenient ferry and what I call the friction theory of friendship. The idea is simple: Some inconveniences aren't merely costs. They are the hidden scaffolding of social life. If I had to estimate my own production function for friendship, it would look something like this: friendship = (proximity x repetition) + (idle time x low stakes) + shared irritation.

 

Proximity matters most when it recurs. Idle time matters most when nobody is trying to impress anyone. Shared irritation helps, but with diminishing returns: A delayed ferry creates conversation; a broken ferry creates rage.

 

This is why friendship is often a product of something else: work, school, church, children, sports, errands, waiting rooms. It is produced not by misery, but by enough common friction to make conversation natural. Modern life has spent decades eliminating that friction. We can work without offices, shop without stores, exercise without gyms and communicate without looking anyone in the eye. Each improvement is defensible, some phenomenal. Together they have made interaction with other people increasingly optional.

 

I am the last person who should be moralizing about convenience. I almost never go to the grocery store; there is too much waiting. I order food delivery as often as my wife will tolerate; our daughters love takeout and I get to keep working. I gave up the gym for a Peloton. Even my walk home from work has to be productive: I don't begin until I have started a deep-research query on whichever artificial-intelligence assistant I am in love with that week.

 

The joke is on me. I used to meet interesting people in line, at the gym, on the walk home -- in the small delays that interrupted my day. I don't anymore. I have optimized away the moments in which friendship used to appear. The last few friction-filled places in my life -- the only places I have made friends in the past couple of years -- are that annoying ferry line, fishing and on the sidelines of my kids' soccer games. Even there, I tried to work when my kid was subbed out. The other parents kept interrupting, so eventually I gave up.

 

Can we prove that the pursuit of frictionless living has contributed to an unintended loneliness epidemic? Not easily. But the data fits better than I expected. Using state-level loneliness measures from the U.S. Census Bureau's Household Pulse Survey, I compared loneliness with three proxies for modern convenience: working from home, broadband-enabled living, and the share of commuters who still move through public or pedestrian space.

 

States where more daily life can be conducted privately -- at home, online, on a screen -- tend to report somewhat more loneliness. States where more people commute by public transit or walking tend to report less. Places built around private convenience look lonelier than places where daily routines force people to bump into one another. After accounting for age, income and who lives alone, the relationship weakens but remains statistically significant. This isn't definitive, but it points in the direction the theory predicts: When the ordinary logistics of life no longer require us to encounter one another, the ties that begin in those encounters become harder to form.

 

Why hasn't the market solved this? Economics has markets for seemingly everything -- marriage, labor, education, housing, therapy, entertainment, even sex. But there is no clean market for friendship. That isn't because friendship lacks value. It is because friendship is hard to sell directly.

 

A friend isn't simply a person with similar interests. Friendship requires trust, recurrence, reciprocity and time. It is difficult to verify in advance and awkward to purchase explicitly. The moment one says, "I am here to buy friendship," the thing being purchased changes. Markets are powerful when the good is clear, the price is observable, and the transaction is socially acceptable. Friendship fails on all three margins.

 

That is why friendship is usually produced indirectly. People go to school to learn, to the office to earn a living, and to the ferry line to get to the other side. Yet these settings solve the hardest problem in friendship formation: proximity multiplied by repetition. They put the same people in the same place often enough that a conversation can become a relationship.

 

Search theory offers a useful way to think about this. In labor markets, workers and firms don't match instantly, because information is incomplete and search is costly. Potential friends are everywhere; good matches are hard to find. And unlike a labor market, there is no resume for a good friend. Institutions reduce those search costs. The office hallway lowers the cost of a first conversation. The school pickup line does the same. So do the church basement, the neighborhood barbecue, the youth-sports sideline and the ferry queue. They put us near one another often enough that trust can accumulate without anyone having to announce that trust is the goal.

 

The usual response is to blame phones. That is too easy. The deeper force is the war on friction. Remote work, delivery apps, online shopping, online banking, self-checkout, telemedicine and private exercise platforms each solve a real problem. But they also remove tiny moments of shared dependence.

 

The solution isn't to make life harder for its own sake. Pointless hassle is still pointless. But as technology redesigns schools, workplaces, neighborhoods and commerce, we should ask what social infrastructure is being lost.

 

Schools, for example, aren't only human-capital institutions. They are social-capital institutions, so making them hyperindividualized and efficient has costs. Workplaces are similar. The goal isn't to abolish remote work, but to recognize that company has value. Cities should build and preserve libraries, parks, playgrounds, walkable streets and mixed-use spaces as amenities and places where low-stakes contact can happen. It is a mistake to think of social life as something individuals can solve alone. Much of what we call "community" used to be produced by institutions as a byproduct.

 

That is why I've come to love the Chappaquiddick ferry. It is inefficient in almost every measurable way. It creates lines. It wastes time. It forces strangers into proximity. It gives people nothing to do for a few minutes except notice one another.

 

Convenience has saved us time. The hidden price is that it has also saved us from one another.

 

---

 

Mr. Fryer, a Journal contributor, is a professor of economics at Harvard, a founder of Equal Opportunity Ventures and a senior fellow at the Manhattan Institute.” [1]

 

1. The Economics of Friendship. Fryer, Roland.  Wall Street Journal, Eastern edition; New York, N.Y.. 11 July 2026: A13.