Hannah Love, she researches cities and public safety at the Brookings Institution.

What we overwhelmingly heard from folks was fear of crime in downtown business districts.

It was two years into the pandemic, and in the middle of what felt like an explosion of American crime.

Crime spiking on the streets of cities across the country.

Crime has increased.

Nationwide spike in violent crime in the U.S. is rising.

Homicides increased by 36 percent.

But since then, the story has gotten a little more confusing.

Sometimes the news says that crime is going down.

But other times, crime is turning cities into war zones.

FBI is under-reporting random acts of violence.

Two-thirds of Americans say crime in the U.S. is a very or extremely serious problem.

And more than three-quarters of us say that there's more crime than a year ago.

So which is it?

Crime up or crime down?

And if it is down, what exactly do most Americans actually think is happening with crime?

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

Crime.

So, my first question was, OK, when the news says that crime has decreased, where's that info coming from?

I asked my colleague Abdullah.

My name is Abdullah Fayyad.

I'm a policy correspondent at Vox.

The two areas that we get our crime statistics from are the FBI crime database.

Law enforcement agencies around the country voluntarily report their crime data.

The other is the National Crime Victimization Survey, which is administered by the federal government and asks people directly whether or not they have been victims of a crime.

And both of these sources have their own limitations.

The survey usually talks to around a quarter of a million people, which is both a lot of people and only about 0.07 percent of the country.

And the FBI's data is, by nature, only made up of crimes that get reported to police.

That's actually a big reason the FBI's data on homicide is one of the most widely studied and cited crime statistics, because murder is almost always reported.

Now, right now, this chart ends in 2019.

Add 2020, and we see murder spike.

There is a reason why people feel like crime is going up, because crime was going up.

But experts we talked to also recommended putting that spike into context, of the way that things were before this, what sociologists and criminologists refer to as the great crime decline of the 90s.

And if you extend the chart in the other direction, too, all the way to 2023, that 2020 spike starts to look temporary.

It's now falling.

Not yet to where it was before the pandemic, but down.

So that's homicide, but we can also look at the FBI's broader violent crime rate over that same period of time, which also looks like it's falling.

The national property crime rate is a similar story.

This doesn't mean that all crime is going down everywhere, but it does mean it's down in most places.

And yet, 77 percent of us say the opposite is true.

So let's look closer at that.

The polling organization Gallup has actually been asking Americans this question for about 35 years.

And their data on how people have answered it over that time gives us a clue on how to interpret it.

Because it turns out, with the exception of like two years, a majority of Americans have always said that crime is higher than it was last year.

What we're seeing is a consistent fear about crime, despite the fact that over the same period of time, crime had been steadily on the decline.

And it's also worth taking a similar look at this chart from earlier, where two-thirds of Americans say crime is a very or extremely serious problem in the U.S.

We can also chart that over time and see that it is also a pretty consistent belief.

But Gallup also asked another version of that question.

How serious a problem is crime in your area?

And only a small minority of Americans typically say that crime where they live is a very or extremely serious problem.

In other words, the vast majority of Americans feel safe.

But there's this kind of abstract perception of crime is out of control.

Crime is out of control.

But not where I am.

But somewhere.

That somewhere was partly the subject of Hannah's research interviews, many of which, remember, were about a specific fear in a specific place.

The fear of random acts of violence downtown.

And crime downtown in the parts of cities where people go to work.

Hannah and her team found that interesting because they also did a geographic analysis of where in cities crime was happening.

When we crunched the data, we found that there was a significant mismatch in perception and reality of crime, particularly as it pertains to where crime occurs.

In Chicago, for example, they found that most of the increase in gun deaths in 2020 happened in disadvantaged neighborhoods on the west-south and southwest sides, and barely at all in the dense downtown loop area.

In New York, they found that in the busiest parts of the city, violent crime only increased about 2% from 2019 to 2022, compared to an 8% increase in the rest of the city.

Downtowns were not driving any sort of citywide increases in crime, even though people felt as though they were.

So, even when violent crime was higher, it wasn't high where people thought it was high.

But why did people think that?

Hold onto that thought, because slight tone change.

This video is sponsored by BetterHelp.

The US election is just a few months away, which for a lot of people is stressful.

It is important to stay informed, but it is also important to take care of yourself.

And this month, while the stress of the fall is still kinda in the distance, is a good time to start.

The way that BetterHelp works is that you fill out a brief questionnaire and you get matched with a licensed therapist, although you can also switch therapists at any time for no additional charge.

Take a moment, visit betterhelp.com slash foxvideo to get 10% off your first month.

Alright, so we were looking at this data that showed us that while people thought urban downtowns were really dangerous, they actually weren't.

One reason people thought that might have been media coverage.

Crime continues to concentrate within neighborhoods where it's always been bad, but community or everyday gun violence often doesn't make the headlines as much as something that would happen in a wealthier or more tourist-based district.

Two violent attacks in Times Square.

Two separate stabbings in Center City.

Four robberies that have happened downtown.

But Hannah found that the people she talked to weren't just seeing stuff on TV.

They were also reacting to something they were seeing in person.

When we were doing our interviews, people often conflated crime and unsheltered homelessness within the same breath.

One thing that has changed a bit since the pandemic is in who you see on the streets in these downtown neighborhoods.

Before the pandemic, that would be overwhelmingly office workers, some tourists, but there were also always some number of people on the streets experiencing homelessness, addiction, mental health issues.

Then the pandemic emptied these places out.

And as office workers and tourists slowly returned...

When there's less people crowding the streets, you're more likely to see things like visible homelessness or visible drug use.

Today, more of us are back in offices.

But there are also, by some accounts, simply more vulnerable people on the streets, partly due to the cutting of social services.

For example, by 2022, New York State had 20% fewer psychiatric beds for people with severe mental illness than it had in 2014.

And from 2019 to 2023, the number of Americans experiencing unsheltered homelessness went up by more than 20%.

I live in Boston, where over the past few years, there has been a homeless encampment that does create this public perception that there is not just disorder, but chaos in how the city is running itself.

Even if, you know, those encampments don't necessarily pose, you know, an imminent threat or danger to the residents that live in those areas.

Changes like these can make a city feel different, even less safe.

But all the data we have tells us that equating vulnerable people with crime is not correct.

And here's the problem with that.

If people don't feel safe, then we're not accomplishing our task.

Stats don't matter if people don't believe they are in a safe environment.

What we're currently seeing right now is a lot of politicians and policymakers crafting policy based on perception rather than evidence.

The governor of New York sent the National Guard to patrol the New York City subway.

Crime on the subway had already been on the decline.

That's an explicit example right there of having perception-based policy take a lot of municipal resources and taxpayer money without having any results because it's in the wrong place.

A lot of what we're seeing are counterproductive policy solutions to crime in the long term.

Harsher penalties for petty crime that include eliminating eligibility for parole and probation.

In San Francisco, voters imposed drug test requirements on welfare recipients.

Tough on crime bills that we kind of felt like the country was moving away from, you know, over the last 10, 15 years and now are coming back in a very big way.

In a way, this is an impossible story to tell.

Try telling the victim of an actual crime that crime is down.

Try telling it to a country seeing real crimes on TV.

Last time I'll show you this chart and I want you to tell me which of the following two clips fits it better.

Last year, the murder rate saw the sharpest decrease in history.

Crime fell to one of its lowest levels in more than 50 years.

To fund the police, Democrats have turned our once great cities into cesspools of bloodshed and crime.

In April 2024, an ABC News poll found that voters trust Donald Trump over Joe Biden on crime and safety by a margin of eight percentage points.

I was honestly surprised it wasn't more than that.

Imagine that the homicide rate kept falling like this, and that by 2025 or 2026, you saw it get to its lowest in decades.

How many people would believe you?

How many people would believe you?