the one with the argument about human capital theory
Classic debate concerning the school/income correlation: Does education actually give you skills that make you valuable? Or is it merely a signal that you are smart and hard working? Here’s another thought experiment: Say person X went to MIT and got a computer science degree. Then say person Y logged into Academic earth, the web site of free university courses, and listened to every free lecture in the engineering curriculum at MIT. She then also downloaded all the free lecture notes from the MIT CS department. And she did all the homework problems. Who would make more money, in the long run, after graduation? Why? If you believe Y will make more or just as much, why don’t you quit school and just self-educate?
My answer: In the short term, X is still more valuable in the labor market. She may know the same basic raw material, but being in a dense engineering environment add more value. She knows the practice of engineering better than the isolated self-educator. In the long run, Y would probably have more labor market value. Being able to self-educate at such a high level, Y probably has enormous discipline and motivation. She can probably catch up on the informal dimensions of engineering. Your answer?
Definitely an interesting point. I think that you are right with one caveat: the ability to break into the field.
I believe that person Y is probably more valuable in actuality due to the ability to self-teach, but person Y may not be able to get their foot in the door of the relevant labor market.
The KSA-heavy job requirements tend to all but disqualify person Y, whereas the less valuable X will be able to fulfill them by virtue of their diploma. HR departments have relied more and more on these KSAs to filter through their applicant pool due to an overwhelming number of applications, and this leads Y to be discarded from consideration before they can showcase their value and talent.
Sure, there are occasionally ways for person Y to stand out and get a shot when they would normally have been overlooked, but it is much more difficult and much less likely. Society tends to mistake credentials for education or ability. It is the reason that many people believe that dropping out of college after 3 years means that their tuition money is wasted, as if the education they received during that time was worthless. To them, the only value of higher education is the piece of paper you receive at the end, not the education itself. Person Y will suffer because they don’t have that piece of paper, only the learning.
So I agree with your assessment, but Y needs to be able to get hired in the first place in order to surpass X in human capital value, and that is far more difficult than it should be.
Kevin Bondelli
February 11, 2009 at 4:59 am
I think that more often than not the person with the degree will do better. It is hard to assess people’s skills, so top firms don’t want to take a chance on someone without one. The better initial job often results in more opportunities for on the job learning, better networks, etc.
I don’t believe the education-income correlation is driven mostly by signaling, but signaling is probably part of the story, at the margin. Therefore it makes sense to create more optional exams and other ways (I have ideas) to signal ability without having to earn expensive degrees.
A related interesting question is, what type of improvements in society’s educational achievement/attainment would have what sort of effects? As someone who is deeply interested in education policy, I’m convinced that aiming at improve people’s educational attainment and test scores can have important positive effects, greater economic growth, reduced inequality, and more. But I think that the benefits from increasing college access are much smaller than those from improving elementary and middle school education.
Michael Bishop
February 11, 2009 at 5:48 am
I love this quote from Emerson: “You send your child to the schoolmaster, but ’tis the schoolboys who educate him.”
So much of learning how to be a Z (here, computer scientist) comes from the informal interactions with peers as well as mentors.
I totally agree that the self-discipline needed to self-teach the MIT CS curriculum would be very bankable, but for a fair human capital comparison, you’d want the equivalently self-disciplined counterfactual learner at MIT as well. In this case, not only do I think the person at the institution gains more, but MIT is banking on it. Their Open CourseWare initiative seeks to make as much instructional content freely available as possible.
brubineau
February 11, 2009 at 5:54 am
There is a problem in this thought experiment (I think). After all, it is possible to *fail* out of the MIT computer science program while it is only possible to *opt* out of the self-taught alternative. X has “succesfully completed” the program (and we can even grade her performance in greater detail).
Y can ultimately only *claim* to have completed her program. For the purposes of the experiment we have to assign her either a hypothetical grade or (though this is really imponderable, I think) a hypohethetical “degree of mastery” of the subject matter. Either way, I think, we beg the question.
But schools, in any case, do not just offer employers smarter people they can employ. They offer them *graduates* (i.e., “graded” persons) to select their workforce from. There is a great difference (to employers who have to select from among many applicants) between being smart enough to pass an exam and actually having passed that exam.
One reason that you should (and no doubt do) earn more when you have a degree is that it costs your employer less to decide whether or not to hire you. If employers had to test people themselves (in practice) a good portion of what would have been your salary would have gone to the salaries of people who turned out to be of no value to the company.
Thomas Basbøll
February 11, 2009 at 9:19 am
In India, pursuing higher education through distance education was pretty popular during the 70s and 80s owing to the fact that the number of full time seats available was far less than the demand for those. Hence, for the purpose of government jobs, a person who gets a degree from a certain university as a distance education student was treated equal to a person who gets the same degree from the same university in a full time mode. However, the time it takes to accomplish this degree was very different for the two modes. While a full time masters would take 2 years, the normal allocated period for a masters in the distance learning mode would be 3 years. So, the motivation factor could also be discounted here. This model is no longer popular, because there has been a revolution in education industry. The number of educational institutions offering higher education has multiplied over the years. At present, there is a clear rank ordering of educational institutions themselves and also the mode of education they offer. Part time or distance learning degrees are considered less legitimate and it worsens if it is from a no name uni. This shift in evaluation criteria is not surprising. When the demand for good quality higher education of a preferred form far over shot the supply for that form, alternate forms of getting comparable education was accepted. However, when the demand for good quality higher education of the preferred form increased substantially, a person not getting his education from the preferred form was evaluated as being incompetent.
To summarize my argument – yes, there is a social value for education, but it is contingent on the evaluators’ frame of reference.
Rajiv Krishnan Kozhikode
February 11, 2009 at 10:12 am
Erratum:
Error: However, when the demand for good quality higher education of the preferred form increased substantially, a person not getting his education from the preferred form was evaluated as being incompetent.
Correction: However, when the supply for good quality higher education of the preferred form increased substantially, a person not getting his education from the preferred form was evaluated as being incompetent.
Rajiv Krishnan Kozhikode
February 11, 2009 at 10:16 am
Agreeing with Thomas above. Degrees do more than signal commitment/hard work/etc. They signal that recognized experts in a given field acknowledge that the degree holder has performed to a satisfactory minimum standard in his or her field. The auto-didactic merely has his or her own approval. It would be costly for every firm to administer extensive aptitude tests for all applications, so degrees may act as a stand-in. Note that there are some exceptions — Google is well-known for their challenging aptitude tests. I suspect that candidate Y may be able to do nearly equally well at a company that administers its own checks of ability.
Trey
February 11, 2009 at 12:59 pm
Y will always be at a disadvantage because of the social network and social skills that develop through in-person interactions. Maybe Y makes up for it, maybe not. But doing all that learning and studying alone would require a personality that many (most?) of us might find odd. The fact that it’s engineering that you picked for this hypo gives Y much better odds, however.
Michael F. Martin
February 11, 2009 at 5:36 pm
A lot of these comments have centered on the fact that both X and Y are educated but only X is evaluated. In principle it might be possible to find another way to evaluate Y. For instance, a system of certification exams open to anyone so that people could get educated anyway they wish (including self-education) and still have a credible credential on the labor market. I think this is more an interesting thought experiment than a (short-term) practical proposal, but it’s interesting to think about both what would such a model be lacking as both education and evaluation (and whether this varies by field) and also even if we assume it would be more efficient, would it nonetheless fail to be adopted because of institutional and political factors.
Gabriel Rossman
February 11, 2009 at 8:18 pm
I have had friends who were self-educated (i.e. graduated from high school but skipped college), although quite intellectual. They were very well-read, but hard to engage with in debates and discussions. I think this has a lot to do with the lack of social interactions and networks mentioned above. When you mainly read by yourself, and have no one to challenge you or present other interpretations (even of scientific texts), your thinking probably will not develop as much. That’s not to say everyone gets worthwhile intellectual stimulation out of college — but the opportunity is there, and employers know that. They also want the prestige that comes with having their employees be university grads (preferably from top schools).
Bedhaya
February 11, 2009 at 9:13 pm
“Does education actually give you skills that make you valuable? Or is it merely a signal that you are smart and hard working?”
After thinking about it some more, it seems to me that the right answer combines elements from these two options. Getting an education signals that you have the skills that make you valuable.
But I just thought of another thought experiment. What if you paid a company the equivalent of tuition at MIT to let you work for them for four years? Would you then be more or less valuable to that company (or its competitors) upon “graduation” than applicants coming straight out of the CS department at MIT?
Thomas Basbøll
February 11, 2009 at 9:21 pm
I agree with Bedhaya’s point about the value of social interaction in a university setting. I personally found the social experience I gained from college to be the most valuable to me. However, I think that there is an aspect of the educational system that ends up dimming the creativity of many students.
The structure of many academic programs leads many students to co-opt the views and ideas of their instructors in order to receive better grades. They do not want to risk their standing in the credentials race and tend to avoid arguing critically against those who hold their evaluations at their discretion. In this case, person X is learning how to figure out what the instructor wants to hear and how to give it to them. This leads to university programs graduating a lot of yes-men and women. This seems most prevalent among students of the humanities and social sciences.
On the other hand, person Y is more likely to have retained their creativity. Their ideas are not necessarily weighed down by the need for approval or the mental barriers of how things have always been done. I think this is why many of the most successful innovators have been those that have not attended or dropped out of college.
Bedhaya’s closing sentence, “They also want the prestige that comes with having their employees be university grads (preferably from top schools),” reinforces my argument in my earlier comment about the perceived value of higher education is the degree, not actually the education. I think this makes Gabriel’s idea about independent certification an interesting one. It would not be perfect, of course. Certification exams would have the same shortcomings as other standardized tests, but it could provide an alternative.
Michael’s notice of the fact the thought experiment involved engineering, and that person Y would be more likely to succeed in it, has me thinking. I think one reason why this is the case comes back to Gabriel’s idea for certification testing. The computer science and engineering field seems to offer more certifications than any other discipline. The skill set is also relatively easy to quantify: can person Y code? Can she work out these equations? It is much more difficult to evaluate people in other fields.
Kevin Bondelli
February 11, 2009 at 9:56 pm
[...] discussion is taking place on orgtheory.net about a thought experiment posed by Fabio Rojas on human capital theory: Classic debate concerning the school/income correlation: Does education actually give you skills [...]
A Positive Example of Blog Comments and Discussion - Kevin Bondelli’s Youth Vote Blog
February 11, 2009 at 10:11 pm
It is hard to measure skills, but a fairly basic math test score is predictive of earnings across all workers, not just those who use math on the job. Interestingly, the estimated effect of years of schooling on earnings declines with job experience whereas the estimated effect of test scores increases. See Altonji & Pierret
I believe there are at least some legal restrictions on giving employees tests because it was argued that such tests were sometimes used for discriminatory ends.
Michael Bishop
February 11, 2009 at 10:11 pm
In theory, they both have the same value, but what we have here is a knowledge problem.
Person X has an MIT degree that vouches for her says “Don’t worry about checking how smart and knowledgable I am, these other people with specific knowledge have already done it and you can trust them because they have a long reputation.”
Person Y may have the same knowledge and skills, but he has no way to communicate this to potential employers.
Josh
February 12, 2009 at 1:44 am
There seem to be two different issues here. The first is related to human capital problems of distance education. The second which appears to be the real issue is the problem of the self-educated versus the formally qualified. Much of the long-term success of the self-educated is moderated through government regulation. In highly regulated markets, the self-educated may not be able to compete without appropriate qualification, no matter how talented. In Taiwan, this has given birth to the underground industry of ‘rented diplomas’. Unqualified individuals who feel they can successfully operate a regulated business, such as a drugstore or building design firm, pay for the use of a qualified person’s name or diploma to register a company and conduct business.
Scott Sommers
February 12, 2009 at 2:14 am
Your site does not correctly work in safari browser
HenIdealtlany
February 15, 2009 at 9:12 pm