The Case Against Education: Why the Education System Is a Waste of Time and Money

by Caplan, Bryan

Published: February 13, 2018

Read: October 27, 2021

★★★★★

Review

This book shattered my world view. Bryan Caplan asks the very important question: what is the value of education, personally and nationally? And he answers with an amazing command and breadth of research, rigorous modeling, and nuanced recognition of uncertainty. And despite this academic rigor, its still very readable. He builds a strong case not just against the current education system, but the fundamental value of education in transforming students into anything besides what they already were. He doesn't shy away from hard questions and looks at all the potential benefits of education from job skills and work ethic to marriage, fertility, values, and crime. I watched three debates about this topic and really think Caplan handily won all of them, the case against education is very strong and shocking. Bryan Caplin is now one of my favorite writers and I look forward to reading his other books. "A cynic isn’t someone that puts a price on the sacred, a cynic is someone that puts a low price on the sacred"

Notes

Summary: Most of what schools teach is not useful for work, schools fail to teach most of what they intend to, adults forget most of what their taught. And transfer learning is veryyy weak.

#Book by [[Bryan Caplan]]
Chapter 1
Author is Berkeley, Princeton, and tenured professor. Want to give more money to education system, but the real deficit is there is way too much of it. Besides CS, premed, you learn history, art, or, higher math, foreign language which is irrelevant for most people. Russian history is not even try to be professional skills. Broadening horizon is great, but ask how often that happens. “Learning doesn’t have to useful or inspirational, but when it is neither how is that anything but wasteful”.
Signaling
- financial payoff record high, 50% higher
- Even if useless, companies will pay high for indications of productivity even if nothing you learned applies
- MIT at 3 years vs flunked out of high school
- Perhaps you already were what you, employers won’t believe you if you just drop out
- Reject Human capital purism - all education teaches job skills and these job skills are the educations value in the job market. Hinting rducation is investing in skills or future
- 1/3 of time in school and financial rewards are signaling, author believes it is higher than 50%
- Society would be better if we all get less, if everyone had fewer degrees then to get top third Elton would just get one less degree
for any single person education pays. But if education is signaling then lowering education level of society would leave wealth unchanged.
- Fallacy is what is true for a part isn’t true for a whole.
- If one person becomes educated better job, doesn’t mean if all are educated better job. Like one person standing at concert vs everyone standing
- If education reduced, less people will get it. If it’s less affordable, you will need less education.
Humanistic benefits mostly wishful thinking. History teachers are basically the only users of history. Few jobs require math, but high school does geometry. But it works right for one right, but ask if system is good.
Chapter 1 magic of education
- Author is economist. Professors are in ivory tower, companies care about their judgement. As a rule, education pays. 10% raise per year and less unemployment.
- Most skills are gained on the job.
- Human capital theory- education as skill.
  - many thousands of hours on poetry or advanced math, art, physics, Spanish
  - Educators tech what they know and don’t know modern workplace
- Most high school coursework won’t be used on the job. Really about getting into college, so super useless if you don’t go to college. Broaden horizons means expose to another subject they won’t use.
- 1. Professors teach about something they find interesting like economics of mafia
  - Student learn something
  - Magic???
  - Prospective in banking, consultations, coding improves
- Academic excellence is signal of skill, mastering these skills REVEALS traits you already have.
- Basics of signaling
  - Requirements:
    - Different types of people, intelligence consciousentous
    - Types not easily discoverable
    - Types must differ on average in signal
  - Employers don’t know which employee will be the best, but can see which signals
  - If something signals intelligence like haircut, then everyone would do it right
  - Signals must be less costly for those in higher demand. Impersonators on average such a chore they stop
  - Clear signals usually have strong stigma, weak signals weak stigma
  - Statically discrimination, elderly pay more in life insurance
- What does education signal?
  - ideal student is well behaved, docile, unaggressive, patient, agreeable to teach
  - One signal overpowers rest, education.
    - Could say education signals intelligence
    - Mark PhD, Steve HS; but both aced SAT. Mark shows education shows conscieousness
    - Jen biggest Ball of Yarn, Jane College Degree, both smart; Both sought difficult tasks and consciousness, but Jane toiled on socially approved goals not personal vanity project. Team player, conformity. Jen works hard when she feels like, but will she for boss.
  - Conformity says Hs for job and bachelors for good job
  - Companies want conformed model workers
    - Team players
    - Dress and groom conservatively
    - No sexual harassment and racism
  - Education shows intelligence, conformity, and consectiousness. Also does prosperous family
  - If you lack these skills to succeed in school, probably lack skills for companies. There are many ways they don’t finish education, but this isn’t about logic but prudence
- Locked in syndrome
  - education is only signal of skill for some jobs
  - Education indicates package of skills. People at top of class have total package, graduating shows prob strong in one and not defecient in one
  - If your so smart, why didn’t you go to college? If you were smart it should have been even easier, to go so can predict you are below average in conformity and consciouteniousness.
  - Therefore other signals of just one of the traits without a college degree are a negative signal for the other traits
  - Signaling conformity, catch 22
    - Alternative signals of conformity signal nonconformity, catch 22.
    - putting sat scores on resume can signal intelligence, but if others don’t do it then also shows socially in adept.
  - If nonschool great people are rare, then companies can ignore and any ambitious do well in school
  - If most of the people that show you strengths have a credential you need one too
  - Higher education isn’t onlllyy signaling, but especially liberal arts
  - If signaling was just intelligence, then one year or iq test would be enough. Could save money with one year, but need signal of conformity and intelligence along social range l. Education is war of attribution
  - Measuring employee productivity can take years.
  - Phonies where credentials overstate, diamond in the rough where credentials understate. Very costly for both
  - Group identity and pity means hiring is costly and employers will often keep subpar workers. Best to dehire them by pushing to seek other jobs
  - Harvard grads are great workers, if you have a fake diploma you can ride that gravy train forever. You get hired do subpar are friends with coworkers and references of company will be positive to get rid of you.
  - Higher Ed is only product where you try to get a little off it as possible . Easy A and little hw is goal. Cheer on canceled class. Rate my professor is about hotness and easiness not marketable skills
  - Anyone can show and start attending classes if you just show up, if you want to sit in teachers will prob let you. Guerilla education, no one does this so human capital purism is wrong
  - Would you rather have a Princeton education or a Princeton diploma? Not no brained
  - Failing and forgetting a class would be equally consequential, but no
  - Cheating would be futile if not about signaling. Detecting cheaters preserves signaling value of diploma
  - Rejoice when class canceled but you can skip class. It’s becomes you relative performance doesn’t change when class canceled l.
  - Students focus on grades rather than learning because they follow the money
  - Do teachers transform waiter to consultant or just evaluate if waiter could become one? Sculptur and appraisal both raise the value of a piece of art
Chapter 2 the puzzle is real
- Why does labor market reward useless education?
  - Learning how to learn or how to think
- This chapter is about learning, learning to oayoffss
- Content of curriculum
  - Useful skills and filler
  - Useful for college, but for job
  - 9-12
    - High Literacy and numeracy, but geometry and lit analysis rarely important
    - Medium career technical programming, cooking
    - Low arts, foreign language, history, pe
  - College
    - High health, agriculture, engineering, challenging so unpopular 25%
    - Medium business, administration, education; non majors regularly compete 35%
    - Low arts, philosophy, historian, liberal arts; few even pretend to prepare for job market
  - You might need X, but don’t hoard all this trash
- Measured learning
  - Human capital purists should think you learn and remember a lot of knowledge
  - Signaling it doesn’t matter, just learn enough to get a grade then not retain
  - Could you just look at increase over years, no cuz could learn outside/mature
  - Instead check value of surplus education,
  - Human beings struggle to remember knowledge they rarely use
  - Most people who study hs algebra and geometry forget half in 5 years and everything in 25. If you go to calc retain algebra and geometry better
  - Upper bound is what adults now know given that education could be up to 100%
  - Average American recall next to nothing about everything but literacy and numeracy. Barely half know saving 5 cents per gallon on 140 gallon is 7 dollars
  - Dropouts after 9 years remain functionally illiterate and inumeriate
  - American revolution 83% adults failed grades. 38% failed test of America citizenry
  - Failing basic history doesn’t mean you don’t know half of history it’s like not knowing letters you can’t read. If you don’t know 3 branches of gov you don’t understand politics.
  - Barely half of us adults know the earth goes around the sun
  - .7% of Americans report learning foreign language very well in school, 2 years in high sccool
  - Better to say schools teach of subjects ie they exist rather than they teach the subjects
  - If we learn so little, why do employers value it? Because they grade on a curve
- Relevance of relevance
  - Maybe standardized test miss most of learning
  - How to think?
  - Transfer of learning,
    - education is narrow they learn what you teach them if your lucky not more
    - People apppy knowledge in the context you teach them to
    - Tell two stories back to back analogous, 30% transfer. Get 70% if you bluntly tell them to apply. This is ideal conditions
    - Time delay and less similar impedes transfer.
    - “Despite importance of transfer of learning, research findings over past 9 decades clearly show that as individuals and instructions we have failed to achieve transfer of learning on any significant level”- Robert haswell leader in transfer learnin
  - Education fails to durably improve critical thinking out the classroom
  - People reasoning skills barely change between 1 and 4 years. Credentialed people are better though
  - Teach physics and math problem. Only 10% of physics used what they learned to solve math, but 72% of math solved physics. Easier to go from abstract to specific than specific to abstract
  - Most students from a very elite uni showed no improvement for everyday stats in the final week of their class, imagine years after
  - Different majors get better at the reasoning they are taught, barely change in other reasonings. If all goes well, they get better in what you teach them
  - Much of knowledge in inertia, so not applied outside class. Physics class honor students, still get many qs wrong if phrased differently. How many forces on coin while in air? One just gravity, but many think 2. 12% get right then after physics class 28% get it right
  - Teachers start with idealism that you want it to be difficult so are teaching discovery of principles for transfer, but eventually experience and research wears you down and graduate education is learning of information and now you want it to be as easy as possible. Teach people exactly what you want them to learn as close as possible to the situations you want it to be applied.
  - If it doesn’t look like real world, then when the knock of geometry in real world knocks it knocks too faintly to hear
  - “Teachers are better at teaching what can’t be measured than what you do” - hahah so confident, you can barely teach them what you are trying to
  - Mental muscles like exercise. No you exercise the muscles you specifically want to grow and atrophy
- Make you smarter
  - Iq, education matters in study
    - Dropping out and some early childhood programs improve iq by up to 30 points
    - Extra education usually seem to boost iq
  - Iq test practice gets better
    - 50% student could expect to score in 60% the. 70% on their next try’s
  - School is vague form of iq test prep with literacy and numeracy
  - School teach kids what answers are required
  - In Sweden, find school days raises comprehension but not logic or spatial. Raised crystallized intelligence but not fluid
  - Is iq raise we see even lasting? Fade out
    - Early education raises iq by 32, by 14 age declined to 10 points
    - 13 iq that vanished by age 8
    - Headstart raised by a few points but gains disappear by end of kindergarten
  - 1 month decline after summer vacation
  - Once you graduate, imagine the decline
- How people get good at their jobs
  - Practice, doing specific tasks over and over again
  - Novices improve if they are
    - Given task with well definined goal
    - Motivated to improve
    - Feedback
    - Chance to retry
  - Deliberate practice to get to higher level
    - Specific and sustained effort you can’t do well, not what you already know
    - Don’t become skilled by dabbling, but by doing one job better
  - Students learn discipline and socialization
    - Boring work in hierarchical org
    - Teens working, work also teaches though and even better because more specific
  - Colleges and work ethic don’t overlap much
    - 50 years ago college was full time job with 40-50 hours in class/studying, since 1996 work collapses and now 27 academic hours and only 14 hours of studying. Fell for all choices of majors.
      - With extra free time, play. 85 hours for other activities.
      - Teach them life is easy
    - Grade inflation, 4 year easy vacation
  - Year of work raises income by 2-3% and prob teaches better discipline and socialization, yet education increases your premium dramatically…
- Who you know
  - Half of all workers used contacts to get their current job
  - Wins friends, better school and better connections
  - But would be better to get friends or family in your narrow domain
  - Stanford Cs or frats could be great funnels, but many find it afterward in vast economy
- Most of what schools teach is not useful for work, schools fail to teach most of what they intend to, adults forget most of what their taught. And transfer learning is veryyy weak.
- So they learn something? Ok but is the huge rewards all due to this??
Chapter 3 the puzzle is real
- earnings gap is enormous
  - High school earn 30%
  - College get 73%
- Did college cause this or would it have happened anyway?
- Is it contradictory to say Education causes big effect on earnings? No signaling explains why despite the weak growth of job skills
- Signaling model is best if you learn nothing in school
- Ability and education are the package, must compare equal ability to see value of education
- Dropouts are below average students, so if they finished they should think about being Below average
- Ability bias is a sepa rate story besides signaling and human capital model
  - Issue 1: visibility of skill
    - Human capital, skill is obvious
    - Signaling, must be inferred from resume
  - Issue 2: Education effect on skill
    - Human capital, education effects skill
    - Signaling, you forget most
  - Issue 3: Education on income
    - Human capital, boost by boosting skill
    - Signaling, boost by certify skill
    - Ability, no effect on income since skill is obvious to employers
- Both signaling and ability say employers like skill, education don’t do much on skill, ability bias employers see skill but in signaling employers don’t. Ability says you can rise without diploma
- It’s probably a mix, you can rise with just ability but diplomas make it easier
- Study control of iq and education
  - Correcting for iq reduces education premium by 18%
  - 1 iq increase income by 1%
- Education premium falls by 50% for lots of things like iq, self esteem, family, attitude, personal behavior
  - Reverse causation could overstate ajbility
- Reverse causation, if you control for X but education increases X then could underestimate. Doesn’t seem to be problem
- Missing ability, controlling should show all abilities
- Family background counts 0-15%. Rare studies that correct for Intelligence and family background conclude that intelligence alone is sufficient. It is heredity
- Cognitive ability bias 20% cautious, 30%
- Noon cognitive bias 5%, 15%
- Raw education premium doesn’t account for inherent differences, but education still pays
- Card consensus
  - Some labor economics say ability premium is almost nonexistent
  - Value of college often ignore ability bias
  - Just quasi experimental studies like twins with different education, but author has problems as not well controlled
- Name any ability that well educated tend to lack, good across the board
Wheat vs chaff
- high paid wheat math and lit
- Useless chaff of history and art
- Could be idea, But chaff definitely pays too
- Ignoring preexisting ability, extra classes in math, foreign language modesty increase earnings but extra classes in English and fine arts decrease earnings
- A in math in high school earn 10% more, advanced math 21% cuz more likely to go college and goes away if not more increase ambitious
- But, this wheat chaff theory is overblown
- Major matters
  - Business average
  - Steep declining by correcting by ability
  - Strong students tend to major in high earnings, need to correct by this
  - After correcting for sat and stuff, major premium falls by 50%
- Wheat chaff check worst majors to see the chaff
  - Lowest 20%
  - Anthropology, history, communications earn 30%
  - Economics get almost as much as engineering, professor says economics is easy and mostly impractical fluff
- How much degree related to job?
  - 55% closely
  - 25% somewhat
  - 20% not
  - You earn less if less related
  - Market mismatch greater for more vocational subjects like engineering or health, roughly 0 for others like English
- Average philosophy earns 30% then equal worker
- is credentialling from the state?
  - Good gov require credentials
    - Egypt guaranteed college graduates job
    - Government positions in us 3% haven’t graduated high school
    - If gov inflated salaries, then private shouldn’t hire them
    - But gov flattens income scale, highly educated paid more in private and lowly paid more in gov
    - 80% more in private than public
    - Therefore gov undervalues education so not true
  - license
    - Raise income by 10-15% as restricts competition
    - Some licenses require education and more educational premium
    - Could say education matters because government makes it matter, but Education premium dwarfs license premium
  - Iq laundering
    - Human capital purists say iq tests 3 hour vs 4 years, but about more than intelligence and intelligence and no degree signals low consciousness and conformity
    - Iq test illegal, because disparate impact on minorities. Colleges are allowed to use these considerations. Must be business neccisity to use iq test
    - Kernel of truth, 10-30% use cognitive tests but the law makes it not more common
    - Total test tax(lawsuit cost) is under 500M a year, but college graduates get 40% more. Could profitably break law and get better talent. 100B in labor costs
    - And iq test ban didn’t really affect premium
    - Intellectually challenging interviews and more easy to give covert iq tests
- Underrating benefits of education
  - Unemployment, college 2% and high school 4%
    - 50% effect disappear if correct for attitude and intelligence
  - Miss measurement of like bad survey data cuts both ways, need to correct for mismeasurement on all variables and ends.
    - Author argues mismeasurement underrates premium
- Real rewards of education
  - Skillful do a good job, successful have a good job
  - Correct for everything and education premium doesn’t go away, even anthropology
  - Signaling, going to school to certify skill
  - Irrelevant education is valuable
Chapter 4 the signs of signaling
- 4 research literature to detect and measure of signaling
- Sheep skin effect
  - Study for final for last class but are prevented from taking it, should you take last semester or just take all the degree skills
  - 12th grade pays more than 9-11, 3.4 regular years
  - Senior year pays more than freshman-junior, 6.7 regular years
  - Finishing sends a much better signal than not finishing
  - if you somehow you correct for degree not years in school, 60% of amount
  - Correcting for ability still has the relative value of the last year and degree
  - Maybe it will banish if we had better ability measures, would have to cut payoff for degree but not years of schooling
  - One way can see human capital gets payoff for years of schooling and signaling for degree effect, so 40-60 for hs
    - But in reality, signaling should get all the payoff for degree and some for the years of education too
    - This view sets lower bound for signaling
- Malemployment and credential inflation
  - Overqualified too good for job, malemployed is their job isn’t good enough for them
  - 3 ways to measure
    - 10-20% malemployment if you look at a atypical education for occupation
      - Overlooks if average is over educated
    - Asked if they have too much education, 35%
    - Job analysis decides what job needs and check, 20-35%
  - After 2008, reached 40% for New college grads
  - Average education is also rising without jobs needing it, one estimate says 20% of rise in education is do to more complex jobs
  - Human capital would say credential didn’t teach them to do a better job
  - Signaling explains this average rise, because too many competing workers have more signals
    - Pay for skills unused, high school grads out earn dropouts in 93% of occupations . Same for college grads
    - Security guard, bartenders, waiters, are all paid for diplomas
    - Ability bias? Shrinks but doesn’t eradicate premium.
    - Builds character? Does year of work really build work ethic worse than education
- The speed of employer learning
  - Statistical discrimination, use real ideas on average to make decisions
  - Credentials are statistical discrimination, but eventually employers will figure out how good you are and give raises etc
  - As you work, education premium should disappear and be replaced with ability as employer learns. Takes years and decades not months
  - 10years education premium only falls 30%, and then plateu
  - On study found Sheepskin effects take 2 decades to disappear
  - One study says college graduates get near perfect skill assessment, others like hs dropout get almost no payoff for ability until 10 years in labor force
  - Perfect knowledge is one plateu and so is continuous ignorance
  - Even if you stop learning about spouse, doesn’t mean you have perfect knowledge
  - Employers don’t like demotion or cuts instead give bad raises,
  - could also use signaling to get the job then gain marketable skills on the job that justify job
  - 60-40 one study, 86-14 one study, 20-80 author could be split between human capital and signaling
- National vs personal education
  - If education is signaling, then National increases in education should lead to only personal prosperity
  - If education is teacher, then it should lead to both personal and national prosperity
  - Education premium in us has highest college and high school premiums in developed world. But matches world as education premium is lower in richer countries
  - Personal education premium is high and real. 10% for extra year
  - National education increases do something between slight decrease and decent increases in wealth. 1-3% for extra year.(J national income is prob different than personal income, harder to change ?)
  - Could richer countries just have more schools? Reverse causation.
  - No clear effect of education on growth
- Test scores
  - Good teachers can have life long effects on students
    - Could be real human capital increase
    - Could be schooling inspiration that inspires future conformity to system
    - Either way author is only arguing education is mostly signaling
  - One study shows National test scores lift rate of growth, math and science
    - Prob about intelligence
    - Education years doesn’t really explain test scores
  - Blah blah labor economists are dumb for rejecting signaling
Chapter 5 who cares if it’s signaling
- the selfish return to education
  - Signaling says education redistributes the pie
  - Human capital says it makes a bigger pie
- “A cynic isn’t someone that puts a price on the sacred, a cynic is someone that puts a low price on the sacred”
- Educational rate of return, it’s has time and money costs but many benefits
  - Could look at just time, money cost, and benefit is not just money but also health happiness satisificstion etc
  - Let’s look at good student, who is average on anything of BA graduate.
  - Welfare state decreases effect of education with unemployment
  - Job satisificstion, more educated are marginally happy but could be because of income
    - Compare to peers, so let’s say 0
  - Cost of boredom
    - On average, students are bored. 66% of high school day they are bored in class everyday. 82% say material isn’t interesting.
    - Gates ranked boredom as highest reason people drop
    - Let’s compare job agony vs Ed agony, work has slight edge. Equally bored, work has more pain but more happiness
  - Health
    - After correcting for intelligence and more, don’t look like sheep skin as smooth on years
    - Still year of education increases life expectancy by .1-.2 years
    - Healthier lifestyles? Gets rid of 1/3 of effect
    - Smoking and alcohol decreases as education increase though knowledge is widespread
    - Set value equal to income you miss out on
  - Cost
    - Average cost minus aid is 3.6k per year
    - Triple price on campus housing, can assume you live with family for free so no real cost
    - But oncampus have more social, but can say the net value is worth cost so no impact on these calculations
    - Could have job instead? And miss income and job experience value. 2-3% increase mostly near beginning
    - You could fail, so get cost without prize
      - 25% high school fail to finish in 4 years
      - 60% of full time college fail to finish in 4 years
      - Half of advanced degrees aren’t completed
  - So conclusion, good student(73rd percentile). Let’s look at adjusted for inflation return rates
    - 7.4% return per year for high school
    - Master has first year negative return, but last year gets equal to stock market
    - 4.9% so corporate bonds in college return per year
  - If you ignore ability bias and completion probability, then the education rate of return is huge for everyone over 20%, college 15%, master 12%
- Return on education by type of typical student(background intelligence etc)
  - Good student education looks pretty good, ba without graduate
  - Excellent student, typical masters
    - 82 percentile
  - Fair, typical high school
    - 41st percntile
  - Poor, typical high school dropout
    - 24th percentile
  - absolute return is higher for better students, but they ha more earning potential across board so could still be worth. Excellent loses 65k per year not going to masters from junior high, poor 40k per year
  - High school is great for all types
  - College is only good for good and excellent because academic performance has to do with completion rate
  - Masters soso for excellent and bad for good and money pit for fair and poor
- Major and return to education
  - Electrical engineer pay strongly especially for strong students
  - Fine arts pay poorly especially for poor
- Where you study
  - Less important than what you study, local college engineering is better than Harvard
  - Some studies found after ability bias that prestige is almost worthless
  - Students who apply to many great colleges enjoy great success even if they don’t attend
  - Another study Top schools vs bottom schools, 20% increase
  - Another shows bottom to top raises by 12% for men and 8% for women
  - Students at top schools have very high graduation rates,
- Students choices: Go to top school that accepts you, top state school pretty good or don’t go
  - Masters just great for excellent students that love school
- Sex and selfish return to education
  - Women’s rate of return is higher as they are more likely to finis what they start
  - Women 4 year graduation rate is 33% higher than men
  - Men make more from masters despite completing less
- Marriage and return
  - Like marries like, you marry education level
  - Mating requires meeting
  - Sheepskin effects of sheepskin effects
  - American marriage is education caste system
  - How much will another year in school help odds of marrying money?
  - Marriage is zero sum though as lower earners gain is higher earns loss, but there is benefits of sharing consumption
  - One two person household saves 20-40%, 35% with house and transport and more
  - Marry pumps up return by 1% for men and 2% for women, if you look at marrying at 25
  - Harvard puts you in exclusive dating pool for life
  - “Don’t marry for money, go where the rich people are and marry for love”
- Workforce participation
  - Rises with education but always below 100
  - Workforce and ability bias is pretty sparse, but…
  - Participation into account wipes away educational edge
  - High school stops being no brained, because less than half of these women get full time jobs
- Conclusion
  - Go to high school unless your a terrible student or don’t want a full time career and are poor
    - Don’t get a GED just signal intelligence without consciousness and conformity
  - Go to college only if your strong student or special case
    - Rules
      - STEM, economics, business, engineering, politic
      - go to state school
      - go full time job
    - Special case
      - Major in engineering
      - Elite school miraculous gave scholarship
      - Women who firmly plans to marry
  - Even excellent students shouldn’t get masters, only top 5-10% can expect to cross finish line. Women who marrys can expect good return
  - “Some call this sexist, the correct label is candid”
  - I advise my kids the same way
  - Education decisions currently clouded by conformity and inexperience
  - Doubts-Why are my estimates the best,
    - I am the only one I know of to calculate effects of ability bias, sheepskin effect, and completion probability
    - Where data is sparse I give the best guess, we need a conclusion here
    - Completion probability
      - Very few modelspo
    - How work and school feel? Unclear
    - Health effects on education, he only values a year at income level
    - Masters sheep skin effect less studying so he makes it same as bachelors
    - AA, PHd , vocational etc ignored
  - You have to make a decision here
- Spreadsheet of you
  - Could put value everywhere like
Chapter 6 we care if it’s signaling, social value of education
- Let’s put dollar value on everything everyone values now
- Signaling is back, wasn’t in last chapter for personal reward
- Signaling payoff is illusion, comes from someone else
- Most economists look at human capital purists and take selfish value then tweak
- Social value of education
  - Marriage zero sum for society so remove
  - There are so purely social effects like decreased cost of crime and jails
  - Ranking students has social value, but extra ranking social value ends
  - Marginal social benefit of signaling is zero, even if total social benefit is substantial
  - So how much is education signaling? 80
    - 57% for high school
    - 45% bachelors
    - 25% masters
  - As signaling increases, spread between actual productivity and pay increases. If you are Ba ability and have masters your overpaid, just high school underpaid
  - Social cost all about productivity, loss of time and gain is what you learned to produce
  - Employment risk, signaling is zero sum. Lowers your risk by raising it from another
  - Welfare and tax. What really matters is productivity
  - Job satisification? Mostly wealthy, education could even have negative effect. Status is zero sum
  - Health? Is it status? They are related, 20-60% of health effect disappears correcting for status. Status means it’s zero sum on not to be counted for society benefit
  - Full cost of K-12 is 12136 all inclusive, capital. Let’s exclude food and special Ed, 10769.
  - Special Ed is 13% and twice as costly. That seems high, but schools have reason to inflate
  - College have huge price tag, but offer huge discount to price discriminate. So minus grants piblic 4 year so 8279.
  - little masters data so let’s say the same
  - Completion rate, is failure rate and everyone can’t pass
- Social return to education purely social
  - Holistic? What kind of society do we want to live in?
  - Ideas, education is foundation right. Adds funding to leading researchers right. But national education premium is already studied not clear it helps progress at all
  - College 10% in engineering and bio
  - Extra education seems to decrease voluntary unemployment, less welfare then. But you also get more offers so could just be signaling
  - Crime
    - 65% of inmates never earned hs degree
    - 15% of white male dropouts and 70% of black male dropouts spend some time in prison by 30s
    - 2/3 lower for high school grads and even less for college
    - Count for ability bias here, they were already low Iq troublemakers and antisocial. After correcting, 1 year cuts lifetime jail time by one week. And only the last year counts
    - 30k year for inmates
    - 3-5% of rape, robberies, and aggravated assault leads to jail time
    - Cost of crime 7300 per American per year
    - 1950 crime rate didn’t change though way less college
  - Educated more politically engaged. Is more participation better? Unclear.
  - Children of educated more educated right
    - Adults after 1950, 37% of children of dropouts finished high school, 98% of children of college graduates finish high school
    - “Successful parents might have successful kids because of the genes they gave them.”
      - Adoptees across families. Strong pervasive effects of nature and weak sporadic effects of nurture
      - Genes your parents have you more important than everything afterwards
      - Both adoption and twin studies show an adopted parent with an extra year of education increases child education by about 5 weeks. Shrinks by factor of 10.
      - Zero effect of upbringing on grades. Talent, attitude, and behavior revolve around genes
      - 10% increase in parents income leads to 0-1% increase in kid income
      - So near zero effect on their children
- Quantity of children
  - Family size goes down as education goes up, high school dropouts have 50% more kids than those who finish college
  - Less child means less environmental damage but also less advancement and economic
  - Author doesn’t conclude
- Computing social return
  - Consensus by boosting productivity and workforce participation and cuts crime. But this is meager, better ways to invest
  - In world where everything but sheepskin effect is human capital. Social return less than half of individual benefit. Masters return is negative test is low
    - Best return is poor high school because crime really high for youth and these most likely at 6.7%
    - College is a bad investment if you will do good, awful if you will likely fail
  - If education is 80% signaling, high school is 2% rest negative. Not no benefit but so costly
  - 1/3 signaling high school is good, college pretty good, masters a waste. But this for good students. High school just good for poor students. Even here education system is questionable
  - At 2/3 signaling, even for excellent students it becomes a waste . High school claim to fame is preventing poor students from crime
- Searching for social return
  - Major, selectivity
    - Major all about engineering means skill, but half of engineers don’t use their specialized training cuz employers impressed. Cautious signaling engineering good, even 2/3 reasonable signaling even engineering at best schools with best students bad investment
  - Sex
    - Women get higher return from college
    - But women less likely to apply skills because less jobs
    - Women do less crime by far, men college grads commit more crime than female hs dropouts
    - Crime means men investments less wasteful
- Best guess
  - Reasonable is education is 80% signaling
  - Crime cost is scarce and sheepskin effect,
  - “Educations power of transformation is galatically overrated” Value of education is value times completion rate times not due to ability bias times not due to signaling. Yes well educated are amazing in our society, but education is not the pa th ther
Chapter 7 the white elephant
- The white elephant impractically with much expensive and yielding little profit.
- Every gov supports education. Free and compulsory education across the globe and political systems. Clea majority in every country want bigger education budgets.
- 74% support more education in us. 60% of republicans want to spend more.
- No need to explain gov funding
  - “We need to invest in people”
    - We usually use free market for crucial important
  - Nothing is more important than education,
    - Food and we use free market
  - We need to help poor kids
    - Means education vouchers would solve this without gov running schools
- Who can oppose investment in children
- Popular arguments are rarely the best
  - Irrationality - free market can’t convince people to make good investments. Too young and don’t know enough, short change themselves. Giv makes education benefits to massive
  - Credit constraint - free markets fail to exploit are socially valuable education systems. Students can’t afford education especially minors. Lack of collateral could mean bad interest
  - Positive externalities - little incentive for social advantages, but not personal advantages. Fund social story
- Counters
  - Student irrationality goes with long shot degrees and overestimate value. Signaling produces negative externalities
  - Government might overshoot driving social return below market rate
- Soul craft
  - How could you put a price on admirable values and uplifting humans? Unless budget is infinite you have to put a price on everything
- Cost
  - 1.1 T for education, 700 B for military
  - 3600 for every person in America,
  - half is signaling means half is being wasted. Education spending is mostly private
  - Gov 2010 census estimates state and local spending, 861 B besides direct spending or assistance to students (loans)
  - Government provides 4/5 of education spending, 6% of economy
  - Cut education!
- Cutting education
  - Wasteful spending will get people to share, but they don’t want to reduce spending but redirect
  - Stop wasting then wait until you find ways to spend it
  - Don’t say there must be a cure
  - Cutting fat
    - Social studies, arts, history, so many tragic useless subjects. Breadth
    - End school day
    - Don’t just make it optional, there will be competition. For unimportant subjecg, raise standards and cull bottom 80%. If you survive you might have a future
    - Shut down impractical majors funding
    - Math is great for quants, cut straightforward and then be more careful
    - People will find other ways to signal, work harder in real classes
  - Shift cost to taxpayers,
    - “Attendance could fall” that is what we are going for
  - How will know when we cut enough?
    - Reasonable signaling, says when high school grads earn twice as much as dropouts, ba earn 3x hs grads, ma earn twice as much as ba
    - Completion rates 20% too high, attendance must plummet
  - Raising tuition by 5-10 will plummet it by 20% but overstock of existing grads will last a while
  - Subsidizing education is worse than just subsidizing children and fertility
  - Status quo is so wasteful; we can but kills by omission
- High tuition and student debt
  - Tuition increases improve mix of majors
  - Debt gives incentive to weigh options
  - Loans are a rounding error of cost because it is not easy to get rid of in bankruptcy
  - Loans inflate credentials though
- Raising completion rates
  - Chiefly about their performance. Retention won’t make port students good students
  - If there’s good wonderful programs, just cut it’s been too long to try. Obviously system doesn’t use such low hanging fruit
  - Grants has mixed evidence of effect on completion, performance based does better
- Signaling and social justice
  - Most curbed are the poor to not go
  - If it was skills, Education would exacerbate giving rich more skills too. But it’s wrong, one poor youth is helped sure but all poor youths changed what educational signaling means and means affluent pursue more
  - Poor NEED more credentials to get job
  - Diamond rings right everyone gives so it doesn’t mean anything, rich now give even more
  - I’m world where 50% didn’t go college, I couldn’t afford it becomes real excuse
  - Subsidize does help fee Greta poor, but think of all the pots affected on the other side by credential inflation
- What I really think?
  - All political philosophies should agree for cuts
  - But some change burden of proof, author is Libertarian but puts burden on taxation. Libertarians like vouchers
  - Failure of colleges is not market failure but market correction, public education has awful track record
  - State supported religion has bad track record, just separate school and state don’t cut to 95% cut all the way
  - He wants means tested voucher
  - Private charity in 19yh century 95% of 15 years were already literate before school mandatory
- Why not tax education?
  - Pc is negative
  - Besides politicals, anything wrong?
  - Don’t give real government task of moderating what major or type of education, it’s 20% skill so not everything
- False savior of online education
  - Internet teaches more effectively so schools are obsolete
  - If it’s obeskete, then start edtech right?
  - Education isn’t a bubble, but stable waste as it receives so much stable funding
  - Edtech may get niche but that is all,
  - if ed was just skills, then online education would have case as online better with async content at self paced
  - Students aren’t about skills, it’s credentials. Moocs are great with unbeatable price not kind that can get you a job
  - So you will soon get a better signal right from online schools,
    - but testing traits offline has been dirt cheap for decades.
    - Major test companies make one test spread across thousands of students.
    - Employers don’t take standardized tests from uncredentialed applicants seriously
    - Why would there be online testing Revolution if there wasn’t an offle one?
    - Credentials signal conformity, other ways signal not conformity
    - Video tapes could have replaced mediocrity 40 years ago, but they didn’t and non gained momentum
  - Online education growing inside brick and mortar schools so some courses at home or countless educational website
    - Educational websites are edtainment, saying you took an online course is like I read lots of blogs
    - If you think 4 year college will drop by 10%, feel free to bet author on even odds
- Politics of social desirably
  - Everyone converge on this same folly, complaining about collective resources spending is a waste but will make you enemies unlike saving yourself money on individual folly
  - Truth has to beg for spare change, popular ideas will propagate even if false
  - People don’t like ugly truth, we like what sounds good
  - No kid is stupid sucks compared to 10% are stupid
  - If you avoid call thought, you can believe almost anything
  - “Education is most important investment”, emotionally appealing
  - Selfish returns individual vs social, part is the whole eight
  - Western elite liked education and then everyone liked it, just like Christianity really doesn’t mean it’s true
  - Politicians pander
Chapter 8 one is greater than 0
- we need more vocational schools
- Could take human capital model to be idea of education and try to turn schools into useful skill factories, amplify 20%
- Instead author likes vocational education that teaches job skills by doing
- Social desire ability infects debate, late bloomers?
- Why vocational education rules?
  - Selfish return
    - Academic underachievers before entering, raises pay and employment and hs graduation more than other education
    - In 50s, vocational retard employment but before is pretty worth
  - Social returns
    - Vocational sends negative signals, so students are thought to be worse by going
    - So skills and stigma are package, more social benefit than personal benefit
    - Only 3.8k historians employed way more electro and and plumbers,
    - skills are not zero sum
- Child labor
  - Wouldn’t it be better to learn on the job, that’s child labor
  - Cant employ below 14 years except for newspaper, family , and art
  - CA 16/17 can’t work more than 4 hours on school day
  - Defends of school say the toil and pain of school teaches you something, why not work then
  - Unpaid internship is beneficial
  - Some think giving students option is mistake as academic is so much better as a reliable path, work distraction
  - Stigma just hurts selfish return
  - Unpaid internship is where business gets no immediate benefit, ha business wouldn’t do that so just not enforced for college like jobs. McDonald’s would be shut down
  - Just let dropout go fulltime with parent permission
- School teaches no job, better to teach them one job
  - Don’t think there is liberal education as they just learned specific topics for jobs that don’t exist
  - Vocational about looking at market, educators just teach what they were taught plus their interests
  - Learning general skills isn’t what kids are learning that would be math and literacy
  - Careers without 4 year degree are “alternative”, many viable careers are stigmatized
- 16 year old have real job skills, look up to mentor instead of delinquents
- Achievement oriented spend 2 decades in school, then apply sliver of what they study
Chapter 9 Nourishing Mother, does school nurture your soul
- High social return disapproved, but humanist argument left and author takes it seriously
- Author had transformative experience with reading dictionary
- Education can be good for soul but is it setup this way
- Meritorious education
  - Practical Ed doesn’t need to be uplifting
  - 3 ingredients
    - Worthy content
    - Skillful pedagogy, inspired teachers
    - Eager students , can’t force feed
  - Really pockets of greatness and teachers are poor and students in eager
  - “The best students yawn at the best teachers”
  - Bryan Caplin teaches to lecture hall half empty but rated one of top professors
  - So what’s the cost of conversion, even nourishing soul can’t take all money
  - Free internet, the merit machine, free spiritual admission
    - If people don’t use for spiritual growth, damning that there are.
    - Enlightenment free
    - Lack of enlightenment is due to apathy not access or cost
- Social fallback
  - Does education shift society in right direction?
  - Education does shift values, but how?
    - Leadership from educators?
    - or through peer groups so just shuffle creating pockets of like more secular in educated and less in less educated so overal effect unclear
  - Peer effects seems to outweigh leadership
    - On dimensions where Ed seems to be trying intensely effect is mild
    - Education sways opinion on many issues not pushed by Ed
    - Individual and social effects rarely add up
- High culture falls on deaf ears
  - Ed Instill appreciation for high and classical culture with arts and lit
    - Raise status of ancient art not pop culture
    - Ed can’t be more than 100% of high culture appreciation, so upper band
    - 100 per family per year in reading, 5x on alc
    - Tolkien Rowling top best seller every charts more than classical
    - 1.4% of music market is classical
    - High culture takes more mental effort and hard
- Paper tiger of political correctness
  - 45% dem in teachers k-12, 25% inde
  - 52% dem, 35% ind in college professors
  - Elite schools even more and 8:1 in social scientists. 17% of social scientists self described as marxists
  - Well educated only barely more liberal, extra year seems to move someone on 1-7 scale .014 more liberal so 70 years required to shift one step
  - Education rises people grow less democratic, dem vs Republican extra year .071 steps more Republican!
  - Education reduces racism and sexism, correcting for income and background cuts in half
  - Education raises support for capitalism, globalism, free markets correcting for income and demographic cuts in half
  - Educators prob unpersuasive
  - Even extreme dominance doesn’t change students, but has peer effects
- History and politics
  - Most adults mostly historicalnillerate
  - Transfer from history to policy sucks
- Voting?
  - Voter turnout increases with education
  - Even after correcting, boost still there
  - But education has risen and turnout fallen
  - Turnout depends on relative education. Peer effects
- Modern lifestyle
  - Secular few children
  - Educated are more modern and traditionally
  - As educated, less traditional religious beliefs but church membership and attendance goes up!!
    - 0-6 belief, 0-8
    - Year Ed reduces belief in god by .04 by increases attendance by .06
    - Intelligence decreases both
  - Many students very disengaged
  - Marriage and divorce
    - College graduates more likely to be married because stay married longer
    - College Men slightly more likely to marry, but women less
    - effect from Ed solid, one year raises 0.7% marriage chance and less divorce by .3%
    - All about social equalsa with elite club
  - Fertility
    - Very strong correlation
    - Educated have less kids within and across countries
    - Low educate outbreeding educated by 1/3
    - Even after correcting, one year prevents .1 births
    - Wife educated matters 3x
    - Most impressive impact as job status better than kids, educated often below fertility and the smartest kids don’t happen
    - Developed below replacement, or overpopulation
- Broaden horizons
  - Rich possibilities shown to close minded kids
  - Schools are narrow minded too with many not career options or only classics
  - Trial and error better than academic standard
  - Slogan but curriculum didn’t even try to show bls jobs
- Merit of play
  - Careful exploration, school displaces
  - Kids have more fun and learn more lessons when given space according to one researcher
  - Right now Education dominates, 31 hours in 1989 to 38 hours. Play 11 hours week counts computer but not tv. Much less outside play
  - Leisure college us is actually a feature to give time for exploration
- Cynical idealist
  - I don’t reject transformative education, but cynical about students and teachers even when elite
  - Complusive enlightenment
  - How can we do better? Better pedagogy and marketing, better online
  - Adults could be better ready for enlightenment
chapter 10 5 chats on education and enlightenment
chapter 11 conclusion
- Education overrated, look for summary