Why Is Target Closing Stores?

Data casts doubt on the company's implication that it's due to rising crime.

Target announced this week that they would be closing 9 stores across 4 cities in the next month. The company’s carefully worded statement received widespread attention from national outlets that read more into the statement than was actually said.

CNN quoted an expert saying “It’s proof that store crimes are reaching a new level” (Target doesn’t say the problem is worsening), CNBC said Target cited “violence” as a reason for the closures (Target repeatedly cited safety, not violence).

The meat of Target’s statement reads:

In this case, we cannot continue operating these stores because theft and organized retail crime are threatening the safety of our team and guests, and contributing to unsustainable business performance. We know that our stores serve an important role in their communities, but we can only be successful if the working and shopping environment is safe for all.

Before making this decision, we invested heavily in strategies to prevent and stop theft and organized retail crime in our stores, such as adding more security team members, using third-party guard services, and implementing theft-deterrent tools across our business. Despite our efforts, unfortunately, we continue to face fundamental challenges to operating these stores safely and successfully.

It seems to be fair to come away from the above thinking that crime is getting worse at these stores and that is why they are being closed. A closer look at the above paragraphs, however, highlights how the statement does not say the crime problem is worsening, only that crime generally is “threatening the safety of our team and guests.”

It’s certainly plausible that I am reading too much into it, but the statement does seem to have been precisely crafted to avoid saying that crime is getting worse. But if crime is not rising at these stores, then why were these stores able to bear this suddenly unbearable level of crime for so long, and why do they all have to close so rapidly — within a month of the statement going out?

With this in mind I set out to see whether the stores that Target is closing are in fact experiencing increasing reported crime and how crime in those stores compare to other nearby Target stores that are not being closed.

Target announced that it will close stores in New York City, San Francisco, Portland, and Seattle. Unfortunately, incident-level crime data in NYC and San Francisco does not contain addresses, but the data from Portland and Seattle do. The addresses are anonymized to the block level, but that is good enough to gauge crime levels in and around these stores.

First, some disclaimers about the quality of our data specifically and crime data in general.

• Not all crime is reported to the police. This is especially true of property crimes which make up the vast majority of crimes we are examining. Only 30 percent of property crimes were reported to the police in 2022 according to the most recent National Crime Victimization Survey.

• Minor incidents at big stores like Target may not be reported to the police as a matter of policy. Of course, if Target wants to claim increasing crime is forcing them to close stores but isn’t reporting any of those crimes to the police then it’s a difficult to believe that employee safety is their overriding concern. It is tough to argue that you’ve taken every possible step to reduce criminal activity in your stores if you haven’t even gotten the authorities involved. So let’s assume that Target is, by policy, reporting most crime as they see them to the police.

• As I mentioned above, data is anonymized at the block level so we can’t tell for certain which crimes occurred in Target and which occurred near Target.

Ok, on to the data.

Both Portland and Seattle have publicly available crime data which neatly separates crimes into NIBRS Society, Property, and Person categories. Unsurprisingly, the vast majority of crimes at these stores (90+% at most) are property crimes such as shoplifting. Very few violent crimes have occurred on these blocks since 2016.

There are three Target stores in Portland being closed and two in Seattle. Both Portland and Seattle have three Target locations that are not being closed making it possible to compare crime trends at the closing and non-closing stores. How fun!

Below are graphs of both Portland and Seattle showing the average number of NIBRS crime incidents at each store each month rolling over 12 months. The lines in red represent the blocks with Target stores that are closing while the lines in gray represent blocks with the target stores that are not closing. Note that several of the stores were not open when the graphs start.

These graphs pretty clearly show that the closing stores have typically had less crime per month than the non-closing stores. What’s more, there is no clear surge in reported crime at the closing that would seemingly necessitate an immediate closure specifically at those stores.

There are reasons to be cautious about reading too much into these graphs.

One problem with analyzing the data in this way is that we don’t know how large each store is, so we can’t calculate the degree to which the rate of crimes relative to the number of shoppers or employees has changed. Having 50 shoplifting incidents out of 200 customers in a month could be seen as a much worse problem than having 100 shoplifting incidents out of 2,000 customers in a month even though the former had half as many crime incidents as the latter. A store may have opened an expansion or closed part of the store thereby changing the rate of crimes per customer.

We’re also assuming that the value of stolen goods between stores is consistent. Perhaps one of the closing stores is regularly having large TVs stolen while much smaller items are being stolen in slightly higher quantities from other stores. Again, we have no way of saying for sure from the available data.

It is also plausible that thefts are more organized at one store than at others. A large organized shoplifting of tens of thousands of dollars at a store would show up as one incident as would a single shoplifter stealing a $10 item. Again, we have no way of saying for sure from the available data though there is no reason to suspect any systemic differences in the types of theft at one Target versus another.

That said, the data does not point to any clear changing crime trend that would justify rapid closures of these five stores for specifically crime reasons. Given these trends I’m highly skeptical that rising crime at these stores is the main explanation for the store closures.

One of the stores being closed in Portland, for example, has been open for more than a decade and there were half as many reported crime incidents on its block in the last year than there were on average before COVID. What’s more, Bluesky user Aaron Gordon noted that Target is closing a store in Harlem citing crime while opening a new store…in Harlem about 1.5 miles away.

So, what might be driving these closures if it is not surging crime at the stores?

My theory is that the above graphs are actually telling a deeper story. Those stores may not be closing in spite of lower levels of crime but rather because of them.

Crime isn’t good — obviously — but shopliftings at a store imply something about that store’s performance. That the stores that are closing have lower crime levels may suggest that they are seeing fewer customers than the other stores. The number of crimes isn’t just a measure of criminal activity, it’s also a measure of customers (NOTE: A few readers have pointed out that some of these stores are physically smaller than normal Target stores which certainly fits the data and is good context to have).

You can’t have pickpockets on Bourbon Street if there are no tourists around, so a ton of pickpocket incidents being reported can be taken as a sign of a rollicking good Mardi Gras.

Thinking of it another way, the level of crime at the stores being closed — which is largely unchanged from previous years — may no longer be acceptable to Target given lower sales relative to other stores and other factors (hello online shopping!) that contribute to store productivity. If Target stores have become less profitable (Target’s stock price has fallen 58 percent since mid-2021) then the impact of crime on the company’s bottom line may be more severe even if the level of crime itself did not change.

I am not an executive with Target — call me if interested! — but this explanation fits the available data much better than a long press release decrying the existence of crime.

Unfortunately, crime makes an easy and simple boogeyman for complex decisions that may only be tangentially related to crime — if at all. We are fortunate to have open crime data that allows for a deeper dive into the factors that may or may not be impacting corporate moves such as this.