The Whole Pie, Guessed First SOC 1113 · Trend Lab · Midwestern State University

Labeling theory makes a testable prediction: who ends up counted as a criminal is patterned by power, not just by acts. This page holds the American incarceration data behind that test, from the Prison Policy Initiative's Whole Pie report and the Bureau of Justice Statistics. The numbers are hidden until you commit a guess, because the gap between what you assume and what is measured is the day's lesson. Most people's picture of this system comes from television. The data disagree with television on almost every count.

How many people are locked up in the United States?

On any given day, counting every kind of confinement: state and federal prisons, local jails, youth facilities, immigration detention, and involuntary commitment.

Drag your guess, then lock it in.
1.0M
1.9M+
people incarcerated on a given day (Prison Policy Initiative, Whole Pie 2025)
7M+
times people went to jail in 2022. Admissions, not unique people: the door revolves.
2.9M / 671K
on probation / on parole, supervised but not confined
Scale check: 1.9 million is roughly the population of Nebraska held in confinement on an ordinary Tuesday. Add probation and parole and the total under correctional control passes five million, which is arithmetic on the figures above, not a separate measurement. No other democracy is close to this scale per person. Whatever theory of deviance you hold has to explain this number first.

Sources: Prison Policy Initiative, Mass Incarceration: The Whole Pie 2025 (March 2025). Jail admissions and probation and parole figures as reported there.

Where are they held?

Crime shows are about federal agents and federal prison. Before the chart appears, make a call: which system actually holds the most people? The options are alphabetical.

Which system holds the most people?

People held, by system

3,116
local jails in the United States, against 98 federal prisons
state + local
is where mass incarceration actually lives. The federal system everyone watches on television holds roughly 1 in 12 of the total, which is arithmetic on the rows above.
Why this matters for labeling: the labels are mostly applied close to home, by state legislatures, county prosecutors, and local police, not by the FBI. If you want to know who gets labeled in America, watch a county courthouse, not a federal courtroom drama.

Sources: state prisons 1,097,597 and federal prisons 156,627 at yearend 2023 (Bureau of Justice Statistics, Prisoners in 2023). Local jails 664,200 at midyear 2023 (BJS, Jail Inmates in 2023). ICE detention 44,000, confined youth 34,200, involuntary commitment 25,000 (Prison Policy Initiative, Whole Pie 2025). Reference dates differ by source and the pie has smaller slices not shown here, so the rows are close counts, not one synchronized census.

Three numbers most people get wrong

Guess each one before it shows. These three carry most of the myths about the system.

What share of incarcerated people are held in private, for-profit prisons?

50%
held in private prisons
8%
held in public facilities
92%
your guess
The system is overwhelmingly public. Abolishing every private prison tomorrow would leave 92% of mass incarceration untouched. Whatever drives the scale, it is not primarily corporate profit from owning prisons.

What share are locked up for drug offenses?

50%
drug offenses
~1 in 5
everything else
~4 in 5
your guess
Over 360,000 people, roughly one in five. Large enough that the drug war matters, small enough that ending it alone would not end mass incarceration. The system is bigger than any one label.

What share of people in local jails have not been convicted of the charge they are held on?

50%
unconvicted
70%
convicted
30%
your guess
467,600 people, mostly held because they cannot pay bail, and all of the net growth in jail populations since 1997 has come from this unconvicted group. Labeling theory in one number: jail time arrives before any verdict does, which means the label does not wait for the act to be proven.

Sources: private prisons 8% and drug offenses over 360,000 (Prison Policy Initiative, Whole Pie 2025). Jail unconvicted share 70% (467,600) at midyear 2023 (BJS, Jail Inmates in 2023).

Who gets labeled

Here is the test the whole page has been building to. If deviance were only about acts, the people behind bars would roughly resemble the country. Labeling theory predicts they will not, because the power to label is not evenly distributed. This one is not a guessing game. When you are ready, show the data.

Black Americans as a share of...

all U.S. residents
14%
the prison and jail population
41%
Reading the gap sociologically: a three-fold overrepresentation is not explained by differences in behavior. Decades of research find the gap opens at every discretionary step: who gets stopped, searched, charged, offered a plea, and sentenced to what. Each step is a labeling decision made by someone with the power to make it. Compare the other direction too: corporate fraud costs far more than street theft and almost never produces a prison label or a master status. The system does not just punish deviance. It selects which deviance, and whose, will count.
Keep the claim precise. This is a structural claim about how a system sorts, not a claim about any individual case, officer, or judge. That is exactly the kind of claim sociology exists to test, and the data above are how you test it.

Source: Prison Policy Initiative, Mass Incarceration: The Whole Pie 2025: Black Americans are 41% of the prison and jail population and 14% of U.S. residents. The white-collar comparison is Sutherland's, developed in the readings.

Questions to take with you

  1. Which of your guesses missed by the most, and in which direction? What had put that wrong number in your head?
  2. Most incarceration is state and local, yet most crime stories on screen are federal. What does that gap do to the public's picture of who the deviants are?
  3. 70% of people in jail are unconvicted. Run that fact through labeling theory: what does pretrial detention do to a person's job, housing, family, and plea decision before any trial happens?
  4. Use Merton, Sutherland, Hirschi, and Becker on the 41% versus 14% gap. Which theories can explain it, which cannot, and what does that tell you about the theories?
  5. If private prisons hold 8% and drug offenses are one in five, two popular one-cause stories of mass incarceration are too small to carry it. What would a sociological explanation have to include instead?
  6. Name one number from this page you want to see for your own county, and where you would look for it.