Arnold Kling  

Bryan Caplan Joins the Pigou Club

The Reform Mindset... Tyranny Without Law Enforcemen...

He writes,

Question: If signaling has negative externalities, could government action make matters more efficient?

Answer: Yes - government could tax education. Then everyone could get half as much education and still get the same job offers.

No, this is not Greg Mankiw's Pigou Club, which wants to tax carbon fuel usage. But it is a proposal to tax an activity with a negative externality. Bryan argues that pursuing a degree as a signal to employers (the courses themselves having little intrinsic value) creates a negative externality, in that it forces others to do the same.

Another negative externality is what I call "segregation equilibrium," in which parents want their children to attend schools with the children of affluent parents, so they select high-tuition schools because they are high-tuition schools.

Eight years ago, in one of my first essays on school vouchers, I wrote,

it is conceivable that we would want to enact a "luxury estate tax" on private school tuition. If the tuition is above, say, $20,000, some of the tuition might be taxed. The revenue from this tax could be used to increase the money that is available for vouchers for low-income parents.

Comments and Sharing

COMMENTS (10 to date)

If only there were some mechanism for assigning the authority to tax these types of negative externalities to the private institutions that had the best information about the magnitude and impact of the externalities...

oh wait... that's what the tax to fund school vouchers does.

I get it.

Horatio writes:

Does he propose to quantify the portion of an education that is mere signalling and only tax that?

I am earning a Ph.D. in physics and plan to pursue a career in either quantitative finance or management consulting. If I go into quant finance, much of my graduate education is producing "real" value, but is mostly signaling if I go into consulting.


I know you're not asking for advice from me here, but can I recommend that you stay away from the Martingales and study accounting theory instead?

The best quantitative finance starts with the information provided in financial statements. Seems like many of the quants I know have focused on price charts. Price is too rich a phenomena to be worth studying quantitatively until we understand how firms work better.

Jody writes:

As it's every bit as much about avoiding certain children, why not tax the children (or their parents) that are being avoided?

Personally, I'm not a big fan of Pigovian taxes. It's too hard to quantify the externalities because there's generally an infinite # of interacting effects.

I think the only real "benefit" of Pigovian taxes is they give economic cover to those of the "Reform Mindset".

Someone from the other side writes:

[Comment removed pending confirmation of email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog. Mailinator email addresses do not work and cannot be considered as valid.--Econlib Ed.]

Phil writes:

Is there any way to reduce the benefit of the signal? For instance, by making it illegal for the employer to consider which school an applicant graduated from?

Okay, that particular example wouldn't work, but why not encourage employers to use selection methods that don't depend so much on the signal?

giesen writes:

lol! I can't believe you posted that! You have my vote if it's ever needed.

Rob writes:

I digress a little. From your article of 8 years ago you mentioned having cities own major league baseball teams. Really? Have you ever written in more detail on this? Sounds to me like a perfect way to destroy a great game by having a government entity get involved. Maybe you were just proving a point using sarcasm?

aaron writes:

Phil, the government actually does that. A degree is a degree.

Phil writes:

Aaron: Thanks, I didn't know that! So what does the government do instead to distinguish between candidates with the same nominal degree? Just interviews?

Comments for this entry have been closed
Return to top