Jung Made It Weirder and We Still Use It

Carl Jung believed in synchronicity, the collective unconscious, mandalas as psychic maps, and a shadow self that lived inside every human being like a feral twin. He thought personality could be divided into attitudes and functions that operated in paired opposites, and he laid this out in a 1921 book called Psychological Types that reads like someone trying to build a periodic table of the soul. It is dense, speculative, and strange. It is also the intellectual foundation for the most widely administered personality test in corporate America.

The Myers-Briggs Type Indicator sorts 2.5 million Americans per year into sixteen personality types using four binary dimensions derived from Jung’s typology. Introversion or extraversion. Sensing or intuition. Thinking or feeling. Judging or perceiving. The test produces a four-letter code; INTJ, ENFP, ISTP; and the code becomes an identity. People put it in their dating profiles. They put it in their email signatures. Fortune 500 companies pay consultants to administer it during team-building retreats, and the results get used in hiring decisions, leadership development programs, and organizational restructuring. Eighty-nine of the Fortune 100 companies have used the MBTI at some point. The test generates roughly $20 million a year in revenue for the Myers-Briggs Foundation.

None of this would bother anyone if the test worked. It does not work.

The Test Fails Its Own Standard

The most basic requirement for a personality test is test-retest reliability. If you measure something real about a person on Monday, the measurement should hold on Friday. The MBTI’s test-retest reliability is roughly 50 percent over a five-week interval. Half the people who take the test twice get a different four-letter type the second time. Not a different shade of the same type. A different type entirely.

This is not a minor statistical quibble. A pregnancy test with 50 percent reliability would be pulled from shelves. A blood pressure cuff that gave you a different reading half the time would be medical malpractice. The MBTI’s reliability numbers are publicly available, have been published in peer-reviewed critiques for decades, and have changed nothing about the test’s market penetration. The CPP (the company that publishes the MBTI) responds to reliability critiques by noting that the test is “not designed for selection” and should be used for “development.” This is the equivalent of a compass manufacturer saying the compass is not designed for navigation.

The deeper problem is construct validity. The MBTI assumes that personality traits distribute bimodally; that people are either introverts or extraverts, either thinkers or feelers, with a clean break in the middle. Actual personality data distributes normally. Most people cluster in the center of every dimension, not at the poles. The MBTI takes a bell curve and cuts it in half, then assigns everyone on each side a categorical label. A person who scores 51 percent toward introversion and a person who scores 99 percent toward introversion get the same letter. A person at 49 percent introversion and a person at 51 percent introversion get different letters despite being statistically indistinguishable.

The Big Five personality model; openness, conscientiousness, extraversion, agreeableness, neuroticism; distributes traits on continuous scales and has robust test-retest reliability. It has been the consensus instrument in personality psychology since the 1990s. It is not popular in corporate settings. The Big Five tells you where someone falls on a spectrum. The MBTI tells you what you are. One is accurate and boring. The other is inaccurate and satisfying.

Two Women With No Credentials Built This

Katharine Cook Briggs and her daughter Isabel Briggs Myers developed the MBTI starting in the 1940s. Neither had any training in psychology, psychometrics, or statistics. Katharine was a homemaker with an intense interest in Carl Jung’s work. Isabel was a novelist. They were smart, motivated, and completely unqualified to build a psychometric instrument, which did not stop them from doing exactly that.

The origin story matters because it explains what the MBTI actually is. It is not a clinical tool developed through empirical research. It is a mother-daughter interpretation of a Jungian theory that was itself speculative and untested; translated into a questionnaire format during World War II, when there was enormous demand for personality sorting instruments to place women into wartime factory jobs. The test caught an institutional wave. It met a need. The need was not scientific accuracy. The need was sorting humans into categories quickly and confidently so that organizations could feel like they understood their workforce.

Isabel Myers spent decades refining the instrument, and her dedication was genuine. She believed she was giving people a framework for self-understanding. The problem is that genuine dedication does not produce valid psychometrics. You need empirical validation, factor analysis, longitudinal studies, and the willingness to throw out your instrument when the data says it does not measure what you think it measures. That willingness was never part of the MBTI’s development. The instrument was the conclusion. The data was there to support it.

Jung Would Probably Hate This

The irony is that Jung’s original typology was never meant to be a sorting system. Jung described psychological types as tendencies, not fixed categories. He explicitly warned against using typology to put people in boxes, writing that “every individual is an exception to the rule.” His framework was clinical and interpretive; a tool for a trained analyst working with a patient over months or years, not a questionnaire that produces a result in twenty minutes.

Jung also believed that the goal of psychological development was individuation; the integration of all functions, including the ones that were undeveloped or repressed. In his framework, being “an INTJ” would not be an identity to celebrate. It would be a description of where you were stuck. The developed person would not fit neatly into any type because they would have cultivated access to all the functions. The MBTI freezes Jungian typology at precisely the point Jung said you should be moving past.

But none of that reached the corporate market. What reached the corporate market was a product. Sixteen types, each with a flattering description, none of which tell you anything negative about yourself. The MBTI does not have a “difficult to work with” type or a “probably shouldn’t manage people” type. Every type is presented as having unique strengths and valuable contributions. This is not psychometrics. This is a horoscope with a licensing fee.

Why Corporations Keep Buying It

The MBTI persists because it solves a real organizational problem. The problem is not “we need accurate personality data.” The problem is “we need a shared vocabulary for talking about interpersonal differences without anyone getting upset.”

Teams have conflict. Managers need frameworks for explaining why certain people clash and others collaborate. Organizations need language for diversity of thought that does not invoke race, gender, class, or any of the categories that make HR departments nervous. The MBTI provides a clean, unthreatening taxonomy. You are not difficult; you are an INTJ. Your boss is not a micromanager; she is an ESTJ. The framework converts interpersonal friction into type differences, and type differences are nobody’s fault.

This is genuinely useful in an organizational context. The problem is that the usefulness has nothing to do with accuracy. A made-up vocabulary would work equally well, and in a meaningful sense, that is what the MBTI is. The value comes from the shared framework, not from the framework’s correspondence to psychological reality. Any system that gives people a non-threatening way to discuss their differences would produce the same organizational benefits. The MBTI just got there first, built the infrastructure, trained the consultants, and locked in the contracts.

The consulting ecosystem around the MBTI is self-sustaining. Thousands of certified MBTI practitioners have built careers on the instrument. Entire consulting firms exist to deliver MBTI workshops. The certification process itself generates revenue. Once an organization has invested in MBTI training, the switching costs are high; not because the MBTI is irreplaceable, but because replacing it means admitting the previous investment was based on an instrument that psychologists have been critiquing for decades.

The Barnum Effect in a Business Suit

The MBTI’s type descriptions work the same way horoscopes work. They are written to be universally flattering and vague enough that almost anyone can recognize themselves in them. This is the Barnum effect; named after P.T. Barnum’s observation that a good con includes something for everyone. “INTJs are strategic thinkers who value competence and can appear aloof.” Name a professional who would not recognize at least part of that in themselves on their worst day.

Forer’s 1949 experiment demonstrated this with surgical precision. He gave every participant the same personality description, told them it was individually generated from a test, and asked them to rate its accuracy. The average rating was 4.26 out of 5. The description was pulled from an astrology column. The MBTI operates on the same mechanism, wrapped in just enough psychological vocabulary to feel scientific.

The difference between the MBTI and astrology is not one of kind. It is one of institutional legitimacy. Astrology never got a corporate distribution channel. The MBTI did; through the Educational Testing Service in the 1960s, through Consulting Psychologists Press (now CPP), through thousands of HR departments that adopted it because other HR departments had adopted it. The diffusion was social, not empirical. Nobody adopted the MBTI because the research was compelling. They adopted it because everyone else was using it.

The Internet Made It Worse

The MBTI experienced a second life online. 16Personalities.com, the most popular free MBTI-adjacent test, has been taken over 200 million times. The site is not officially affiliated with the Myers-Briggs Foundation and uses a modified model, but the four-letter codes are identical and the cultural effect is the same. Reddit has dedicated subreddits for each of the sixteen types. Tumblr had a years-long MBTI subculture. TikTok serves personality-type content with billions of cumulative views. The types have become identity categories on the same level as zodiac signs, and the comparison is not a dismissal; it is a structural observation. Both systems provide a taxonomy of selfhood that feels revelatory upon first encounter and becomes a filter through which all subsequent experience is interpreted.

The online MBTI culture has a feature that the corporate version lacks: tribalism. Type communities develop in-group language, stereotypes about other types, and hierarchical prestige structures. INTJs and INFJs are treated as the rarest and most interesting types. ESFJs and ESTJs are treated as basic. The system generates a social hierarchy based on a categorization that does not survive psychometric scrutiny, and the hierarchy feels real to the people inside it because identity categories always feel real to the people who have adopted them. The mechanism is the same one that makes astrology communities vibrant and self-reinforcing. The system provides a language for talking about yourself, and once you have the language, the language shapes what you see.

What Actually Happens When You Sort Humans

The practical consequence of widespread MBTI use is not benign. When organizations use type indicators in hiring; and they do, despite the MBTI Foundation’s disclaimers; they are making employment decisions based on an instrument with the predictive validity of a coin flip. When managers interpret team dynamics through the MBTI lens, they are pattern-matching against categories that do not hold up under measurement. When individuals internalize their type as identity, they are constraining their behavioral repertoire to match a label that may not have applied to them last month.

The constraint is the real cost. A person who believes they are “an introvert” because the MBTI said so may avoid leadership roles, public speaking, or collaborative work that would actually suit them fine. A person labeled “a feeler” may discount their analytical capabilities. The types become self-fulfilling prophecies; not because they captured something true about the person, but because the person adjusted their behavior to match the type. Identity is sticky. Give someone a label with institutional backing and a flattering description, and they will wear it.

The MBTI is not the only offender here. StrengthsFinder, the Enneagram, DiSC assessments; the corporate personality market is crowded with instruments of varying validity, all solving the same organizational need for a shared vocabulary. But the MBTI is the most entrenched, the most profitable, and the most obviously descended from a theoretical framework that its creator would barely recognize.

Jung was trying to map the psyche. He was wrong about many things; synchronicity, the collective unconscious, the particular structure of his typology. But he was doing something genuine. He was sitting with patients for years and trying to describe what he observed. The MBTI took that attempt, froze it into a product, stripped out the complexity, added a licensing model, and sold it to organizations that wanted certainty about something that is genuinely uncertain.

That is the story of the diagnosis industrial complex in miniature. A speculative theory becomes a standardized instrument becomes a revenue stream becomes an institution that resists correction because too many careers depend on it. The science was never the point. The infrastructure was the point. And the infrastructure, once built, does not care whether the science holds up.