Every single health-care cost is distorted and meaningless by itself
September 13, 2014, 11:03 pm
Filed under: Uncategorized

Arnold Kling is convinced by some health care research:

I would say that the Dartmouth studies changed my mind about health care spending in the U.S., convincing me that much of it is “wasted” (I prefer “spent on procedures with high costs and low benefits”).

A commenter raises a similar concern:

What we really want to know is why a comparable developed country with high-quality health outcomes, well-paid medical professionals and similar track records can do the same procedures for so much less.[...]

A difference of 20-50% isn’t too bad. A difference of a few hundred percent without any apparent bang for the extra bucks is what needs explanation.

I would say there is an intrinsic problem with trying to suss out ‘health care waste’ defined thusly and I fear that no one is getting close to the truth.

It’s true enough, I imagine, that an MRI in the U.S. ‘costs’ some multiple of what the same procedure ‘costs’ in [insert country here]. It’s also true that represents or at least is symptomatic of something wasteful going on, and the commenter is right that it needs explanation.

The problem is that the true explanation can’t be found the way people are looking for it.

The true explanation is that every single health-care ‘cost’ of every single line-item in the U.S. (and, perhaps even more so, every other country it’s being compared to) is utterly distorted beyond recognition in a way that bears no relationship whatsoever to inputs and outputs.

Yes, sure, your MRI ‘cost’ $1k in the U.S. if by ‘cost’ you mean, “the number the hospital wrote down on the invoice next to that particular line-item, which they sent to your insurance company and asked them to pay some gross, far bigger amount”. But what does that have to do with anything?

The reality is that health care is basically (give or take) at least half socialized in every modern country, and that includes the US. There is not a free market for health care line-items in the US. Individual line-item ‘costs’ are not haggled over and comparison-shopped to find the equilibrium point of anyone’s supply-demand curves. All that’s really going on is that a bunch of socialism is taking place – stealth taxes, means-tested, and money being garnished from people in a ‘from each according to his ability [to pay]‘ sort of way, laundered through insurance companies or government agencies.

This ‘cost of an MRI procedure’ people obsess over has nothing to do with anything by itself. It’s just part of the gigantic formula that has been developed over time to slosh money around in ways that Smart People find ‘fair’ – completely distorted by politics, regulated by bureaucracies, prices being set line-item by line-item by committee (“Medicare codes”, whether Medicare is the payer or not). No single line-item means anything by itself. Yes you’re being ‘overcharged for an MRI’ if you focus on that line-item. Perhaps, likewise, you’ve been undercharged for the 20 minute visit from the nurse. Overcharged for the band-aid? Undercharged for the X-ray diagnosticiation time. I dunno! Who the hell knows?

And meanwhile, the person in the room to the left to you is ‘paying’ three times as much because they don’t have insurance but do have cash. The person in the room to the right of you, meanwhile, is paying $20 for the same exact thing because they have Medicaid (or gave money under the table to a mid-level hospital bureaucrat lady to fake paperwork to get them on Medicaid). And someone with no insurance and no money is in the ER waiting room waiting to see someone and the hospital will have to treat them free, if only because of EMTALA.

All different ‘costs’ for the same exact stuff. Because? Because it’s (sloppy) socialism. You’re paying for that person to your right, and the person in the ER, of course. It’s just been laundered through your taxes, plus some bogus ‘cost of an MRI’ (or whatever), in a way that is hopelessly vague and impossible to unravel. So yes the hospital has billed you that $1k ‘for an MRI’ but it’s really just cuz the hospital, well, needs $1k and that’s what they’ve been politically able to arrange for billing you for, under the ‘for an MRI’ mantle, for someone in your category. If they could’ve convinced the Medicare price-setting committee to set the price of that particular code to a number such that their blanket ‘we get Medicare’s rate plus $X from you’ agreement with your insurance company would allow them to bill you 10x or 100x as much they would, but as it stands this is how the numbers have shaken out so $1k it is.

There’s no way to unravel any one of these (supposedly individual) ‘costs’ from any other. There’s no way to audit them or track down why they are what they are or why they are ‘more’ than in some other country. Indeed, comparing any particular procedure’s ‘cost’ to that in some other country (which is also heavily distorted, and subsidized by taxes, etc.) is a stupid, apples-oranges exercise. All these costs are distorted and sloshed together and have no individual meaning. All of them contain (implicit) socialism taxes and/or subsidies of various kinds.

Again, that doesn’t mean they don’t reflect waste and it doesn’t mean explanation isn’t needed. It does mean that any effort at explanation that is focused on individual procedures and line-items rather than being holistic and apprehending the socialism embedded in the entire system as a whole, is doomed.

Bryan Caplan still making the wrong kind of argument
September 13, 2014, 10:34 pm
Filed under: Uncategorized

The case for open borders

But Caplan continues: “So why, exactly, is it that people who are born on the wrong side of the border have to get government permission just to get a job?”


Once again:

Even if we stipulated once and for all that government should not prohibit employing a foreigner – indeed, should get out of the business of regulating who employs whom whatsoever – that wouldn’t “open the borders”. Immigration restriction and employment-law are different things. One could make the most wonderful argument in the world for eliminating the latter entirely, but it needn’t budge anyone’s position on the former.

This argument about why-restrict-employment may be compelling, but unfortunately for Caplan it is simply not a ‘case for open borders’. And he knows it, because I have told him so.

UPDATE: Gene Callahan making the same point well.

It is not the “getting a job” part they need permission for, but the “moving to America” part. So once the situation is rightly understood, the question becomes, “Why should anyone not currently part of the polity of the United States of America need to get the permission of its government, the body recognized as holding sovereign power in that polity by the vast majority of its members, just to join that polity?”

But once you phrase the question sensibly, the answer becomes pretty obvious: a group has the right to control who can become a member of the group. If it loses such control, it will soon cease to be a coherent group.

STEM is truly outrageous. Truly, truly, truly outrageous.
September 7, 2014, 6:32 pm
Filed under: Uncategorized

I don’t know who first coined the acronym ‘STEM’ but as a case-study in why it’s dangerous to let faddish buzzwords disseminate to the public, notice that after some mysterious process of handwringing and op-ed pieces, we’ve reached the point where public schools now all seem quite proud of ‘teaching STEM’ in a special ‘STEM program’ of some sort or another.

That is, in addition to the standard subjects of Science and Math, your public school principal lady will probably explain proudly this Back-To-School Night that they also teach ‘STEM’. (Which, um, stands for Science something something Math.) Hmm. Maybe more Logic instruction is what we need?

Dig a little into what this ‘STEM’ is, and I think you’ll find it’s kind of like a dumbed-down, more fun version of Science & Math. You know, with groups, doing hands-on things, with construction-paper, and whatnot.

That’ll teach the Russians to get into space first! Damn Sputnik!

But I’ll bet anything it’s more fun to teach ‘STEM’ than Science or Math. Which brings me to my theory: education reforms are driven mostly by what is fun for schoolteachers to teach.

This also appears to be the case with ‘core’ math instruction. (Where do I stand on ‘common core’ controversy, the world was clamoring to hear?) Not that I really know what ‘common core’ is supposed to be, because depending on who you’re listening to it either

a) just means there are national standards! What’s wrong with that!, or
b) means teaching math in a bunch of weird, unrecognizable ways that puzzle parents when forced to help kids do their homework.

I don’t know how/why a) and b) got intertwined but focusing on what b) appears to involve in particular, it doesn’t really seem any different from the umpteen other iterations of ‘new math’ we’ve gone through. More word problems, more exercises involving drawing things, more problems with vague ‘no wrong way to do this’ answers.

All of which – again – are largely driven by what is fun for teachers to teach.

After all, what is the standard rap against ‘traditional’ math? The main complaint is that it’s “just” teaching “rote” memorization. But what’s wrong with rote memorization? Speaking as someone who got pretty far in math, I’d say that when it comes to the basic arithmetic kids are trying to absorb at the grade-school level, rote memorization is just fine. Arithmetic is one of those things that’s utterly boring once you know it, and once you absorb the patterns. But until that happens, “rotely memorizing” it is just as fine a method as any other. “Rote memorization” isn’t a bad way to teach, it’s just a dreary way to teach. So teachers refuse to do it, and will work up whatever education theories they need in order to not have to. Even if it works.

I mean, I am here to state that to this day the reason I know that 7 x 8 = 56 is because of a stupid memorizing device I was once taught:

Start with 5678. Put a vertical line between them: 56 | 78. So now you know: 56 = 7 x 8!

I am not lying when I say this is how I am able to remember that 7 x 8 is not 54 (which must be 9 x 6 then). I don’t visualize that fact. I didn’t discover it on my own through a journey of playful exploratory self-discovery in a mixed-skills group of four. I simply memorized it and now it’s part of my brain’s muscle memory. Another one, albeit not ‘math’ per se: the way I’m able to remember left vs. right is by sticking my thumbs out and looking at the backs of my hands, the one forming an L is my left hand. This extended upward to something like the ‘right hand rule’ in physics and vector calculus. Two cheers for memorization!

There is plenty of truth to some of the things common-core defenders say though. It’s (probably) true that different people learn in different ways. It’s true that when it comes to a typical arithmetic problem, there are multiple ways to attack it, none of them ‘wrong’. If you get the right answer, using right logic, the method cannot have been ‘wrong’.

The problem is that this sort of observation – like the buzzword ‘STEM’ – is dangerous. Once it trickles down into mainstream educational usage it becomes an elementary schoolteacher telling her class that this or that math problem ‘has no right answer’. Which is totally wrong! Of course there’s a right answer! There are even right and wrong (false logic/incorrectly-reasoned) methods! In the great game of telephone that is apparently schoolteacher theory, the (correct enough) view that ‘there’s no single correct algorithm, algorithms that use correct logic are all equivalent and must necessarily lead to the same right answer, so one should use whichever algorithm works for them’ has gotten all garbled and reinterpreted to mean something like ‘all algorithms are equally ok and there’s no single right answer’.

That’s not math. That’s not good instruction. But I’m sure it’s more pleasant to be able to say such things when standing in front of a classroom full of puzzled looks than to simply instruct kids how to carry the 1 over and over again. And teacher comfort/enjoyment is the important thing in our educational system, so that’s the criterion that prevails.


UPDATE: Commenter (and I think actual teacher) Mark says,

…I’ve never heard any teacher in my district describe it as fun. It’s a giant pain in the ass for everyone, not least because I have to defend it to parents even when it’s crap.

But who are you gonna believe, a real teacher, or the butt from whence this theory came i.e. mine? I rest my case.

Every one should go to college, but not everyone should go to college
September 4, 2014, 8:55 pm
Filed under: Uncategorized

The NY Fed is releasing a series of posts (Part 1 here) about the returns to a college degree. The lede you may have heard from this work is something like: the NPV of a college degree is ~$300k, which is an all-time high; the years it takes to pay off a college degree is at an all-time low of ~10. So yay college!

But if you ask me, this is the real lede, down near the bottom of the post:

Why has the value of a college degree remained so high even while tuition has been rising and the wages for college graduates have been falling? The primary reason is that the wages of high school graduates have also been falling

‘Also’. As if, by some sort of coincidence. In other words, it’s less that going to college gives you a bunch of extra money, than that not going to college gives you less money than ever before. But why is that? Because fricking everyone is going to college! Because of studies like this!

The broader trend is pretty clear. A good sizable chunk of people who would have been non- or marginal-college candidates back in 1970 the first year graphed at that link, most of whom would have fallen into the HS Diploma-Only bucket, now fall into the College Degree bucket. But (aside from being 4-5 years older) do they really know any more actual things, or have more skills, than their 1970 counterparts?

I have some doubts. It also doesn’t matter. The cold calculus of competition, and the retarded rigidity of corporate HR departments and their sorting software, means that nowadays if you’re one of those college-marginals, you have to go to college just to keep up with the others, and just to get the analogous level of job you might have gotten in 1970 with only a HS Diploma. So while sure, that may be ‘worth it’, you are not getting ahead when you do this, you are just treading water to stay afloat against the greater and greater drowning heaped upon non-college-grads. That’s not me talking, that’s the NY Fed, according to the real lede of their study.

And don’t get me wrong. It’s not that the individual calculus isn’t correct. I’m not going as far as Bryan Caplan does in advising giant swaths of people that going to college is a bad decision for them. Sure I guess that’s gotta be true of some marginal percentage of people, but it’s possibly not a very big percentage. Lacking other info, if a HS student drawn at random asked me ‘should I go to college?’ I’m gonna guess yes and I’d bet they’d indeed collect a net wage premium.

Indeed the wage premium might even be underrepresented by the NY Fed’s study. They use a generic 5% discount rate on future cashflows (which are CPI-adjusted), but in this day and age of low interest rates that may be too punitive. It’d be perfectly defensible to say well actually one should discount that future wage premium less and thus weigh it far more positively against the cost of paying for college today – making it seem like an even better deal to just shell out or borrow to do it. In the other direction, maybe this ‘wage premium’ is an illusion driven mostly by the top-earners, an artifact of the growing inequality in society – the top-earning tail is self-selected to be the type of person who automatically goes to college, and their increasingly-high wages (against everyone else’s stagnant wages) brings up the average. There are also plenty of simplifications a model like this necessarily makes and potential quibbles (for one, worryingly, I can’t tell if their calculations of wage premia are based on after-tax dollars – who cares if I get a ‘wage premium’ in before-tax dollars if the higher income is almost-all taxed away by the progressive income tax? And what if I project that progressive income tax is only destined to get even more progressiverer?) But these are quibbles; again, I don’t dispute the basic point.

What I dispute is the conclusion we should draw. The national drive to send everyone to college has clearly led to degree-inflation. A college degree today just doesn’t mean what it did 40 years ago. It merely means, at minimum, that you kept up with the others in your cohort who got one. You might basically just have a 1970-equivalent high-school diploma in your hands, but you wasted** 4 non-earning years of your life (and spent possibly tens of thousands) getting it.

**well ok, not wasted. Those years are usually fun; such people may partake of a bunch of neat drugs, play computer games, enjoy a bunch of hookups, meet future spouses, generally consume a lot of valuable leisure, etc.

In any event, it’s entirely fair to wonder if that journey is worth it on a societal level (even if it we all agree it’s likely worth it for this or that given individual).

It’s an arms race. I thought we all agreed that arms races are wasteful? They lead to a deadweight-loss. If everyone in the college-marginal bucket – however one might define that – could somehow tacitly simultaneously agree just not to go – “but I have to play musical-chairs or I won’t get a seat?” “let’s just not play musical-chairs” – then employers having to fill the same slots wouldn’t be able to discriminate against them, and they’d get the same jobs (and mostly be perfectly able to do the same jobs, despite not having taking Sociology 2B). All would be fine and functional, I claim. There is no problem whatsoever envisioning such a societal equilibrium. In that world, the wage premium for those people is gone – but they’ve saved the cost of college. It’s a win-win. But as things stand, we can never get there, because all the incentive in this prisoner’s dilemma is for each particular college-marginal to defect, and so they do, and understandably, correctly so.

Now, I’ve noticed it’s almost impossible to get in a real conversation about this with a regular ol’ Smart Person without being misunderstood. “You hate knowledge!” “You’re trying to keep people down!” So: no, I’m not advising against any particular person going to college. I’m not saying college is ‘bad’. I’m not saying we should ban it or even discourage it per se. However, can we at least recognize this arms race we’ve created, and apprehend that it can be wasteful at the margin, that it may have set up a collective action problem. If so, then although maybe there’s nothing we can ‘do’ about it per se, we might at least look for areas and things to fix where we are needlessly adding to the prisoner’s-dilemma incentive. Some that come to mind are changes in government policy and some aren’t:

  • Government could stop giving and guaranteeing student loans
  • Or at least discriminate by major/school; make at least some attempt to price them accurately
  • Firms could stop ‘requiring’ a bachelor’s degree for the (I suspect many) positions where one really isn’t needed
  • HR departments could rely less on degree as a filtering mechanism for candidates
  • Culturally-imposed automatic expectations of ‘where is your little Billy going to college?’ could shift
  • Any taboos against not going to college could be removed

Some are going to be natural responses that we don’t even have to actively seek out, like people discovering and taking advantage of various forms of the community-college arbitrage. One could then publicize those, like I just have.

And of course, there’s this:

  • When people bandy about studies like the NY Fed’s that ‘prove’ college is worth it, you could say the story is actually more subtle, and explain all of the above.

Hey, I can dream.

Star Wars 7 Whiteness Update
September 4, 2014, 7:26 am
Filed under: Uncategorized

When an initial casting list was announced for the upcoming Star Wars 7 movie, right-thinking Smart People reacted the only possible way they could: by complaining it was so disproportionately dominated by white males!

This was clearly the first thing we all noticed about the cast list. There was nothing else to even notice, really. It goes without saying that what proportion of the human-like characters in this long-ago galaxy far, far away are ‘white’ (=have skin hues similar to those found in European descendents here on earth), and female, is pretty much the only important fact to be adduced about a Star Wars movie, entertainment-wise.

See, we go to movies in order to see a demographically-correct (according to some ill-defined standard) slice of human skin hues depicted onscreen for us doing whatever. What other reason could there even be? When a movie contains lots of nonwhites and females in it, it’s a good movie; when it doesn’t, it’s not. What other criteria for film quality could there even be?

So any good Star Wars, sci-fi/fantasy, and/or action-movie fan would agree that a Star Wars 7 with, say, N characters played by nonwhite females sitting silently around a conference room for 2.5 hours in front of a static camera would be preferable to a let’s say hypothetical Star Wars movie in which 2/3 of the (talking) main characters are ‘white’ guys, 16% is indeterminate but revealed to have been implicitly ‘white’ in a later film, and the 16% female main-character contingent is also ‘white’. (Where again we note as an aside that ‘white’ here comes with an asterisk, since none of the characters we’re talking about are even earthlings let alone descended from Europeans. It’s just the skin hue that concerns us here (because it’s so important).)

All of that goes without saying. But as more and more of the cast has been announced I kept thinking back to that initial reaction, and wondering, ‘how much more correcter and correcter is that reaction getting?’ Because obviously that reaction was correct. It’s just the question of how correct that concerns me here. With that in mind let’s take a look at the announced cast for Star Wars 7 shall we? And break down their race/sex, because again, that’s the only important thing about this movie. When in doubt, in determining race I will be using the all-important metric, ‘how the person looks according to my initial knee-jerk reaction and perhaps how I initially feel about that’, since that’s the tried-and-true metric used by the critics linked above to the first cast announcements.


Harrison Ford: White. Guy.
Gwendoline Christie: White. Amazon-Giant Girl.
Carrie Fisher: White. Girl.
Andy Serkis: CGI specialist/doesn’t matter? But ok maybe he has a live role: White. Guy.
Mark Hamill: White. Guy.
Domhnall Gleeson (sp?): looks like White. Guy.
Adam Driver: White. Guy.
Lupita Nyong’o: White guy. Just kidding. Black. Girl.
Kenny Baker: Robot/doesn’t matter. Exclude.
Oscar Isaac (nee Hernandez): Hispanic. Guy.
Peter Mayhew: Giant doggy suit. Exclude.
Max von Sydow: White. Guy.
Anthony Daniels: Robot. Exclude? But he talks like a damn White. Guy.
Daisy Ridley: Who? White. Girl. I guess
John Boyega: Black. Guy.
Christina Chong: looks like Asian. Girl.
Dixie Arnold: White. Guy.
Crystal Clarke: Black. Girl.
Roman Bloodworth: Puppets I think. Exclude
Pip Andersen: White. Guy.


Male – 11 (65%)
Female – 6 (35%)

CONCLUSION: Is the ratio supposed to be 50-50? Or is ‘more females = better’ just a rule that always applies? Let’s go with the latter. Either way we see that this movie is sexist and probably being made by people who hate women, to have cast only 6 of them.

White – 12 (71%)
Black – 2 (12%)
Asian – 1 (6%)
Hispanic – 1 (6%)

Now, Star Wars 7 is an American movie. One intuitive criteria to use when solipsistically and parochially complaining about the skin hues of play-actors in a sci-fi/fantasy film not even set on earth or about humans descended from earth, as we Smart People are wont to do, is to compare the skin-hue breakdown to that of US society. So, for comparison the current United States of America here on earth is said to be roughly 63% white, 13% black, 17% Hispanic, 5% Asian.


Whites are indeed overrepresented in Star Wars 7. 12 out of the 17 characters we know of are white (counting the sissy robot) and 11 out of 17 would be most accurate. That one-too-many is an egregious numerical disparity. No wonder everyone reacted the way that they did!

Blacks are about right, to within rounding. No wonder everyone reacted the way that they did!

Asians are technically slightly overrepresented percentage-wise but with one character, you have to round and there’s nothing else to do, so they’re also about right. No wonder everyone reacted the way that they did!

Hispanics are underrepresented. You need 2 more to get par. THERE WE GO

Unless they switch a white cast member to a Hispanic actor, or say Carrie Fisher discovers some Hispanic blood in her heritage, I suggest you boycott STAR WARS 7 as a crime against humanity and against Smart People everywhere. After all,

WHITE GUYS – 9/17 (53%)
NON-[WHITE GUYS] – 8/17 (47%)

Or, to summarize,

“Mostly white guys” –Forbes Magazine, 4/29/14

UPDATE: The IMDB page is stale, not updated for the announcement of Syrio Forel the swordfighting teacher from Game of Thrones. Dang it this screws up all my percentages! I think he’s a Greek dude. Is that bad? Good? Oh I give up, I’m not nearly Smart enough to do these calculations.

My recline rant
August 27, 2014, 8:25 pm
Filed under: Uncategorized

Next time you buy an airplane ticket check the fine print. What you probably won’t find: language to the effect of, ‘the purchase of this ticket fully and without restraint entitles the ticketholder to the recline function of his seat for the duration of the flight’. That doesn’t mean one can’t recline. (It also doesn’t say you can breathe while on the flight…) It does mean however that claims like “I paid for the right to recline!” are made-up. No, you paid for an airplane ticket. There are some things explicit (we’ll take you from point A to point B, at such-and-such time, we kinda-sorta promise) and many things implicit. It didn’t specify a ‘right’ to recline just like it didn’t specify a ‘right’ to occupy such-and-such volumetric cylinder of space extending from the tip of your seat up to the ceiling, and along the bisecting midpoints of the armrests on either side of you. Not all things are ‘rights’ and not all of those are spelled out. When it comes to reclining, the simple fact is that reclining reduces the space available to the person behind you, touches them, involves their personal space and body. Whether that is ok involves at minimum you and that person, it’s not something you can just assert is ‘my right!’ and ignore the effect on others. “But if the airplanes don’t want us reclining why do the seats have that function.” Well maybe they shouldn’t but that’s beside the point. Some seats still have old ashtrays from the days before smoking was banned. Again, beside the point, if you want to ‘recline’ maybe that’s ok but you still need to confront & justify the fact that you’re affecting the person behind you, and how, and have good solid reasons for being ok with that effect. Do you? By the way I use the word ‘recline’ advisedly here because it is barely anything resembling ‘reclining’ that we are even talking about and continues to strike me as utterly bizarre that anyone from any walk of life would ever value that physically-imperceptible 4-degree difference so much they will defend their ‘right’ to it to the death. If no one were able to ‘recline’ starting tomorrow what would be lost, utility-wise? Nothing measurable. The supposed gigantic comfort gain you get from such a tiny ‘recline’ is all in your head, if you thought about this rationally you’d acknowledge I’m right, and you should let yourself be convinced to stop wanting it. Meanwhile the bruises on the knees of the person behind you is not in his head, I promise. Oh, but bruising some other person is ‘your right’ because you ‘paid for’ that right, right? Yeah no.

Student loans
August 26, 2014, 11:10 pm
Filed under: Uncategorized

Barry Ritholtz got some charts from a DB analyst to say that concern about student loans is all hype. Which, according to my Law Of Ritholtz™, of which no violations have ever been observed in captivity, must therefore mean that this is, likewise, the consensus conventional wisdom of Smart Wall Street. Indeed I have no doubt whatsoever that it is. But I still have some questions.

Now, I’m prepared, in fact happy to believe student loans are much ado about nothing but I just don’t understand how the data or charts he brings to bear show that. One is the idea that the average loan balance is small and (at <$25k for the 70th percentile) 'manageable'. This seems to me a relative term. Manageable for whom? In what situation? Compared to what? And what if we all agree they're manageable for most, like 90 percent, but there's still some sizable 10 percent tail that struggles with them? What is the effect of their struggle on the overall economy? Wouldn't that require some oh what's the word an-al-y-sis? (I mean I'm sure $25k is a drop in the bucket to Barry Ritholtz, but…)

And oh yeah if they’re so manageable, what’s with the delinquency rates? You at least gotta acknowledge them, if only to explain them away with some just-so story, in a piece like this…no?

Second he argues with/debunks the Zero Hedgeland notion that student loans are the ‘next subprime’ and will ‘tip the economy into crisis’ or something, which seems to be a case of picking the most extreme possible extrapolation of what a rational person might legitimately worry about re: student loans so it’s easier to argue against. But are we not allowed to worry about anything short of the ‘next subprime’ from now on? What if they’re just a major drag on/headwind against the economy – on ‘demand’, if that helps all the Keynesians among my gigantic readership – isn’t that bad enough? Bad, at least? Worth noting? A little?

Now, there is the observation that larger loan balances should correlate with graduate school, and that’s a good point, except: couldn’t it just be indicative that increasingly too many people are going to graduate school and (therefore – because more people chasing not-more jobs) won’t exactly reap the returns historical experience might lead one to associate with the extra degree? Of course it could. In which case, that’s bad not good. But I guess we don’t really know, here.

He closes with some ‘context’, that student loans outstanding were some $1.3 trillion while credit cards were some $900mm. Huh, so the ‘context’ is that student loan balances are one-third…higher than card balances? And this is supposed to illustrate what? Moreover, a peek at the data he linked to shows that student loans have exploded some 50% in 5 years during a time of, otherwise, national deleveraging (e.g. card balances decreased over that same time). Notice those 5 years coincide, coincidentally, with federal direct involvement in student loans (as opposed to merely wrapping them). So if nothing else we have an unprecedented program to try to grapple with and integrate into our forward-looking views; wouldn’t a hype-debunking article try to do so instead of just citing static-snapshot numbers?

Finally, the methodology. It’s based on comparing, er, balances. Raw principal amounts, unadjusted for anything. Get one simple big number, compare it to another big number, throw them in a chart maybe, that’s all you gotta think about. (Smart Wall Street!) But what does it mean to say there are $X student loans out there ‘and’ $Y credit card balances anyway? Not all that much. It’s kind of apples/oranges. Credit cards are a different animal. Big fractions of those balances turn over every month. Credit cards are dischargeable in bankruptcy. They don’t sit on the taxpayers’ balance sheet as an asset with big fat future carry gains priced in. There is not the everpresent poltical risk (perverse incentive?) posed by the Elizabeth Warrens of the world constantly whining about them and musing about ‘cleverly’ restructuring them, and writing them down somehow without recognizing any losses thus hiding the problem elsewhere. Etc. etc.

There’s also no concept of leverage or risk-adjustment anywhere in this discussion. Is the economy levered to student loans in a way that makes the effect of that ~$25k/per not exactly 1-to-1, dollar-for-dollar? I don’t know, but couldn’t it be? Hard to find clean data, but I believe at their max, subprime mortgages outstanding were maybe $2-3 trillion. Hey, which is only twice as big as the student loan market right? So it was only twice as big a problem? Equivalently: student loans are ‘half a subprime’ already – and rising? Oops, this debunking just imploded on itself.

But of course that’s silly. The raw numbers don’t tell the whole story, you need to bring in some appreciation of the risk and the leverage. You know, with the subprime loans being made to bad credits (and fraudsters), then those loans being securitized, and unplaced tranches of those deals being re-securitized into CDOs, and those CDOs wrapped by bond insurers and AIG, and people selling CDS on AIG, and CDS on AIG going into synthetic CDOs, and people stuffing levered super-senior synthetic CDOs into ABCP vehicles, and pensions buying that ABCP, and….etc. etc. etc. As everyone knows, it wasn’t really only a “$2 trillion” problem.

The point being that you just can’t really learn or understand very much about the world just by looking at the unadjusted raw amounts outstanding of things, you also have to grapple with the overall market structure and how whatever you’re looking at actually feeds into some bigger picture, maybe even drill into the micro details. Well, same goes for student loans: who takes student loans? who leaves them outstanding? what’s their likely life arc? how will these balances change their behavior/outcomes? Obviously there is nothing directly analogous to CDOs in the student loan world playing anywhere near the same role as was in subprime but mightn’t there still perhaps be some mechanisms by which the effect of student loans is magnified or otherwise not-well-encapsulated by merely looking at things like aggregate balances and individual balance histograms?

Again: I don’t know. But I think there could very well be. However, you won’t identify such mechanisms if you don’t even look for them because it’s much easier just to dismissively ‘debunk hype’. With some charts. So just saying, if you want to convince me that you’ve ‘debunked’ the ‘hype’ around student loans, you now know exactly what sort of piece not to write.


Get every new post delivered to your Inbox.

Join 437 other followers