Friendflation is real. The numbers everyone quotes for it are not.

Somebody in your group chat has stopped coming to things. Not dramatically. They are just busy, or tired, or they will catch the next one. And the next one.

There is a word going round for what might be happening: friendflation. It is a portmanteau of friend and inflation, and it describes the rising cost of simply having a social life. Meals out. Drinks. Tickets. The wedding, and the flight to the wedding, and the group trip before the wedding.

I wanted to know whether it was a real thing or a nice piece of wordplay. So I went and pulled the actual price data. The short version: the price squeeze is real and it is specific, and I can show you exactly where it bites. But two of the statistics that appear in almost every article written about friendflation do not survive being checked, and the whole framing quietly skips the most important fact about why people stopped seeing each other.

First, where the term came from

Worth clearing up, because most coverage either gets this wrong or avoids it. The earliest datable use in print we could find is a Betches article from 25 November 2024, and even that describes the word as already circulating rather than claiming it. No article we read names a person, an outlet or a post as the origin. It seems to have surfaced online with no author, the way tipflation and shrinkflation did.

It reached a wider audience through a Financial Times piece in August 2025, which is the source almost every subsequent article is quietly downstream of. One widely-shared write-up dates the term to 2025, which is simply wrong by about a year.

What actually got expensive

Here is the part nobody has published. Every figure below is US Bureau of Labor Statistics Consumer Price Index data, not seasonally adjusted, running to June 2026, which is the most recent month available.

US city average, CPI-U, not seasonally adjusted. Note that October 2025 is missing from every CPI series, because collection lapsed during the 2025 shutdown.

Three things fall out of that table.

Eating out has genuinely outrun inflation. Food away from home is up 36.5% since February 2020 while the overall price level rose 29.1%. That is a gap of about seven points, and it is the clearest single vindication of the friendflation idea.

The cost of drinking with people has risen well over twice as fast as the cost of drinking alone. Alcohol away from home is up 27.0%; the same alcohol bought for your fridge is up 11.6%. This is the number I found most striking, because it isolates the social part of the spending from the substance itself. It is not that drink got expensive. It is that the room got expensive.

Live events are the pressure point right now. Admissions is currently the fastest-rising category in the whole set at 5.6% over twelve months, well above the 3.5% headline. Concert tickets have their own arc: the top 100 tours worldwide averaged $96.17 a ticket in 2019 and $132.62 in 2025, according to Pollstar, roughly 38% higher.

But only one of these actually beat inflation

Here is the finding that surprised me, and I have not seen it anywhere else. If you deflate each category by general inflation, so you are asking not "did it get more expensive" but "did it get more expensive than everything else", most of the friendflation story disappears:

Real change is the category index ratio divided by the all-items ratio, over the same period.

Eating out is the only part of a social life that has genuinely outrun the cost of living. Drinks out, cinema and live events have all risen roughly in line with everything else, which means they feel more expensive for the same reason a car and a haircut feel more expensive, not because socialising has been singled out.

That is a narrower claim than "everything about seeing your friends got expensive," and it is the true one. It also explains why the squeeze bites the way it does: the default adult social occasion is a meal, and the meal is precisely the thing that outpaced your wages.

Note the two windows are telling different stories and both are real. Admissions is the fastest-rising category right now, at 5.6% over the last twelve months. Over the whole period since 2020 it has merely kept pace. Events got expensive recently; restaurants got expensive and stayed that way.

Sources: BLS Consumer Price Index, June 2026 (released 14 July 2026), series CUUR0000SA0, SEFV, SEFX, SEFW, SS62031 and SERF02. Ticket prices from Pollstar's 2025 year-end analysis.

And if you are in Atlanta, it is worse

We are an Atlanta company, so this one is personal. In the twelve months to June 2026, food away from home in metro Atlanta rose 5.9%, against 2.8% for Atlanta prices generally and 3.4% for restaurants nationally. Eating out here rose more than twice as fast as everything else here, and noticeably faster than eating out elsewhere.

Source: BLS Southeast Information Office, Atlanta-Sandy Springs-Roswell CPI, June 2026. Atlanta CPI is bi-monthly, has a smaller sample than the national series, and measures change over time rather than price levels, so it cannot tell you whether Atlanta is dearer than anywhere else. Next release: 11 September 2026.

The statistic everyone repeats, and why we are not going to

If you have read anything about friendflation, you have probably met this sentence: Gen Z spends $433 a month on friendship. It is in most of the articles on the first page of Google. You will also meet a claim that 65% of people have cut back on socialising, sometimes attributed to housing costs or debt, sometimes broken out as 67% of Gen X and 49% of boomers.

Both trace to the same place: a single survey of 995 people, published by a site whose stated purpose is serving readers "with subpar credit" and whose experts rate financing products aimed at the subprime market. The page carries no publication date, no panel provider, no field dates, no weighting and no margin of error. Its entire methodology is two sentences.

That would be reason enough for caution. But the numbers themselves do not hold up either.

• The $433 is not what people think it is. The same page gives a Gen Z annual figure of $6,181, while twelve times $433 is $5,196. Roughly $985 a year of friendship spending sits outside the monthly number, because it counts recurring socialising only and puts weddings, travel and holidays elsewhere. Every article quoting "$433 a month on friendship" is describing the wrong number.
• The 65% has had its premise swapped. The source says people reduced social activities "to focus on saving and cutting spending." The version in circulation attaches it to housing costs and debt, which is a narrower and more alarming claim that the survey did not ask.
• The 67% and 49% do not appear in the source at all. We looked. The only generational split published on that item is a figure for Gen Z. Those two numbers are also implausible against the source's own adjacent figures.

We are pointing this out for a self-interested reason as much as a principled one. We are asking you to believe things we tell you about how friendship works. If we passed along a number because it was convenient and everyone else had, you would have no way of knowing which of our other numbers got the same treatment.

A statistic that everyone repeats and nobody checks is not evidence. It is a rumour with a decimal point.

What better-sourced research actually shows

There is real data here. It is just less quotable, because it is more careful.

LendingTree surveyed 2,000 US adults in July 2025 with published fieldwork dates and a named research vendor. It found that 69% of Americans have opted out of a social outing because it felt too expensive. Gen Z was 71% and boomers were 67%, which is worth sitting with: this is not only a young person's problem, and the four-point gap is much smaller than the coverage implies.

The finding that stayed with me is a different one. Of the people who skipped something over money, 39% did not tell the person who invited them the real reason. The CFP Board found something similar in January 2026: 67% had declined an event over cost in the previous two years, and 56% of them did not say why.

So the visible symptom of friendflation is not a conversation about money. It is a friend who has become unreliable for no stated reason. From the other side of the group chat it looks like drifting apart. Sometimes it is just a bill.

The same LendingTree survey found 36% have had a friendship end over money. On weddings specifically, 34% have declined an invitation because of the cost, and 12% of those said it damaged the relationship. Which is not surprising when The Knot puts the average cost of attending a wedding at $610, and the average bachelor or bachelorette party at $1,400 per attendee.

The Knot's guest figure comes from a survey of 1,000 US adults fielded in October 2024; its bachelor and bachelorette figures were fielded in July 2023 and have not been updated since, so treat them as 2023 dollars. Bank of America's 2026 Better Money Habits study has the strongest methodology of anything we read here, using a probability-based panel with a published margin of error: 24% of Gen Z passed on events with friends in the past year, and 42% practise "loud budgeting", meaning they say out loud what they can and cannot afford.

Now the part the other articles leave out

Here is where I have to argue against my own headline, because the honest version of this story is more complicated than the price data alone.

Americans were already seeing much less of each other before any of these prices moved. The American Time Use Survey has tracked this since 2003:

BLS American Time Use Survey, population aged 15 and over. Collection was suspended from 1 October to 12 November 2025 and BLS states the effect of that gap on the 2025 estimates cannot be quantified, so treat 2025 as provisional.

Two things are obvious once you look at it. About 70% of the entire two-decade decline had already happened by 2019, before the pandemic and years before the price spike. And through 2021 to 2025, the worst inflation in forty years, the number is flat and the share of people socialising on a given day actually went up slightly. If price were the thing stopping people, that is precisely the window where you would expect a collapse. There isn't one.

The decline in friendship itself is older still. The Survey Center on American Life found the share of Americans with no close friends went from 3% in 1990 to 12% in 2021, and the share with a best friend fell from 75% to 59%. That was documented before the inflation spike, not after it.

And the most widely read account of this whole phenomenon, Derek Thompson's The Anti-Social Century in The Atlantic last year, does not mention prices once. He blames the car, the television and the phone.

So what is actually true

Both things, and they are not the same size.

The squeeze is real and it is specific. Nominal wages rose 32.0% between February 2020 and June 2026, and food away from home rose 36.5%. The average worker's overall basket is roughly covered. The socialising basket is not. If you feel like going out costs more than it used to in a way your pay has not matched, you are reading your own bank statement correctly.

But friendflation did not start the retreat. It arrived on top of one that was already two decades old and considerably larger. Which leads to the conclusion that almost every article on this subject avoids, because it is not a tidy one: cheaper hangouts will not fix this. If price were the binding constraint, the last five years would have looked very different from how they actually looked.

The bit that is actually fixable

Look again at the drinks number: out is up 27.0%, home is up 11.6%. The venue is doing most of the damage, not the thing you consume in it.

So why does adult friendship default to venues? Not because anyone prefers paying. It is because an open invitation has no organiser. When you say "we should get together" to five people with different schedules, somebody has to pick a time, a place and a thing to do. Nobody wants that job, so the group falls back on the option that requires no decisions: a restaurant or a bar, where the venue supplies the structure and the bill is the price of admission.

Every article we read on friendflation lands on individual behaviour. Budget better. Host at home. Suggest a walk. That advice is fine and it will work for about three weeks, because it asks one person to keep doing the coordinating job that nobody wanted in the first place.

The alternative is to change the unit, not the price. That is the whole idea behind NexSpark Friends. We place you into a Circle: four to six compatible, verified people near you, who meet in person on a schedule. Because the group is fixed and recurring rather than assembled fresh each time, the venue question gets settled once instead of every single time, and a group that already meets can set its own norm about what things cost. A standing commitment also survives a squeezed budget in a way an open invitation does not, because "are we still on for Thursday" is a much easier question than "does anyone want to do something".

It is free to join and free to meet, which is not a coincidence given everything above. And if the pattern in this article sounded familiar, the friend who quietly stopped coming, we wrote separately about what a one-sided friendship actually looks like and about why four to six turns out to be the number that works.

Money is a real constraint and I am not going to pretend otherwise. But most of the friendships that quietly ended in the last twenty years did not end because of a bill. They ended because nobody scheduled the next one.