Tuesday assorted links


2. Higher education is broken, and "buy more" is not a real fix, so sure. There must be bait and switch in any promise for free higher education for all.

Was Tyler at SXSWedu? (Yes, with a funny upside-down pic!) Any hope there? I did see a crazy kickstarter for $1000 competence based degrees. Something like that would certainly lower the cost far enough that "free" was within reach.

"Higher education is broken"

How? Why?

It's broken only in so far as it costs too little, and attracts too many people who shouldn't be in college at all.

And? Typical non sequitur answer. To say something is "broken" you have to demonstrate that

a) It was somehow better in the past

b) there is a better alternative (not sure how making it "free", ie removing barriers that keep out the lest prepared, is supposed to fix anything)

c) and you can't do that by asking people who hire them, if they would like to hire more prepared people :) Who would say no to that?

I link you to what is a conventional modern discussion, including the stat that only 43 percent of "academic industry leaders" said higher education prepares students with necessary workforce skills. There are many similar out there.

Do you know what "non sequitur" even means?

(Your a,b,c are not interesting.)

You can provide me with whatever you want, it's a non sequitor since it doesn't address the question of explaining how or why it is "broken"?

I.e., if 43% of whomever said something, does it tell us anything about whether this is

a) better


b) worst

than in the past? Understand now? For something to be "broken", it implies it was better in the past, but now is worst. Evidence?

My point b and c form above are the main points, regardless if they interest you or not. Its a basic concept in marketing learned in freshman year: you don't ask the customer "would you like more good stuff?" The answer is usually yes.

I think the modern definition of "non sequitur" is "not the answer I wanted."

AIG - This seems like to right way of thinking to me. I do not pretend to have concrete answers, but I don't think anyone has demonstrated a), there doesn't seem to be much actual discussion of b) and point c) is a patently critical viewpoint. Like, we shouldn't completely ignore the perspectives of people who it affects, since they may possess the most important perspectives, but they will also be biased by self interest, so we should take their perspectives with a heavy dose of salt.

If higher education in the USA is "broken" in a sense, I would consider the comparatively high share of students with very large student debt and poor job prospects. This gets in the way of family formation, entering real estate markets and may trap them in sub-optimal career paths because they MUST make money RIGHT NOW to pay off loans whereas with no debt they can more easily take some BS job in the short term and invest a lot of time to get on a better career path.

But that's a matter of financing. I think to evaluate the question of whether higher education is "broken", I would be more concerned about academic outputs. Is the academy producing a large number of flat earthers and religiuos cultists? Is it producing "too many" graduates in fields with little obvious value to career paths, intellectual inquiry or good citizenship?

Just because you're not satisfied with where things stand now does not mean that it's worse than it was before. Much like the "Make America Great Again" slogan, which presupposes that America used to be great but now has fallen, fixing a "broken" high education seems to presuppose that things are worse than they used to be, and it is far from obvious that this is the case.

I really hate to reiterate the obvious, but part of the "expanding benefit" of college degrees is that they have replaced high school diplomas as the minimum bar for many unskilled jobs. Retail sales assistants now have college degrees. This "shows" as a new and improved wage premium over high school, but only superficially.

But this is well known: College Graduates Starting Wages Drop In Last Decade

Go read something.

"2. The Bernie Sanders program isn’t actually free tuition."

That isn't really a correct statement. The Bernie Sanders program is "free tuition", it just requires states to adopt it and agree to pay for a significant amount of the cost.

It would still be "free" to the student, just not to the states. We all knew that somebody has to pay for it of course.

Yglesias' article does a nice job addressing why education costs are so high these days and how Sanders' plan, far from contributing to the problem, will be effective in reversing the trend.

I'm just kidding, he didn't touch the subject. I of course have no idea why that could be.

"Yglesias’ article does a nice job addressing why education costs are so high these days"

Average net tuition prices are ~$6k a year. An average of $30k in student debt overall.

Meanwhile, returns to college education have increased considerably over time.

If there's one thing about college tuition, it is that it is not high at all.

Curious where you're getting these numbers from. I did a quick google search and found the Net Tuition, Fees, Room, and Board at a 4 year Public University was about $14,120 and 26k+ for private four year institutions. Granted, most of that seems to be room and board, but still.


Sticker price is not net price. Nobody pays sticker price.


NY Fed 2014 analysis. Bachelor's net price ~$6,500. Associate degree net price, negative. Somehow, this is supposed to be "too expensive"?

$30k average student debt? That's less than the price of the average CAR bought by Americans.

Returns on bachelors degree, 15%/year average. 15%/year returns!! And people think that its "too expensive", and its "broken", and should be made free?

it should be made more expensive.

Room and board aren't the costs of education. Room and board are also costs one pays even if one lives with their mother, or rents off campus. One just accounts for them differently. People need to have a place to sleep and food to eat, whether they are going to college or not.

Room and board have to be paid regardless. It's meaningless to include these in the "cost of a college education".

The factor that's not included but should be is opportunity cost in the form of lost wages.

Imagine that there's a big construction boom in your area and starting wages for construction workers rise. Suddenly going to college becomes a much more expensive proposition regardless of what happens to tuition.

Indeed, we saw that during the early/mid 2000s there was a drop off in college attendance by young men in the Sun Belt as they rationally chose a high paying job in the housing sector. When the economy tanked, college attendance spiked.

>Room and board aren’t the costs of education.

Sounds like you should try to get your money back. It obviously didn't help.


Predictably, Yglesias is "selective" in his choice of facts. Check out https://nces.ed.gov/programs/digest/d15/tables/dt15_333.10.asp for some actual data. From 2007-8 to 2013-14 total public higher education revenue rose from $273 billion to $353 billion. Tuition rose from $48 billion to $70.5 billion. In other words, total revenue rose by $80 billion of which $22.5 was higher tuition. Non-tuition revenue rose by $57.6 billion. Part of that was higher hospital revenues and investment income. However, many other sources of revenue rose as well. For example, Federal non-operating grants rose by more than $10 billion.

NCES has a great wealth of education statistics. See https://nces.ed.gov/programs/digest/2015menu_tables.asp for a starting point. So does the Census by the way. Don't post them online. You will be banned.

We will be banned?

2. "Even though there's nothing wrong in principle with federally managed higher education, and even though there are plenty of countries around the world where it works fine, these institutions are resistant to such massive change."

Right on Matty, just like our sparkling public grammar schools. Nothing wrong in principal.

"Nothing wrong in principal."

Irony or clever humor?

I consider our public grammar schools to be the poster child for the manifestation of the principal of public education.

They're riding you because you used the wrong spelling of "principle" for this situation. It's a homophone.

And irony wins out.

"and even though there are plenty of countries around the world where it works fine"

What other country in the world has the institutions of American quality? How many other countries have over half of the top 100 universities in the world? And of the rest of the half on that list, how many are private universities?

What sort of idiocy is this?

What sort? Little Matt is very, very special.

That the Europeans have destroyed their university system goes without saying.

But American universities aren't doing so well, or at least it depends on what you mean by doing well. Yes, American universities produce a lot of people with Nobel Prizes, but not as many as you might think. What is worse, they tend not to teach them as undergraduates. The American college with the most Nobel-holding former students is Harvard. It is well ahead of everyone else with 73 alumni. But very few of them did their undergraduate degree at Harvard. Compare that with Cambridge which is in third place with 66. It is hard to find one of their Nobel winners who did not do their undergraduate degree at Cambridge.

So it depends on what you mean by "Top 100". If you mean Harvard has a lot of money and can attract the best academics from around the world, then yes, that is certainly true. It also means they can offer the best students good conditions for doing their doctorates. Most of Harvard's Nobel holders did their Ph.D.s there but nothing else from what I can see. This is just a follow on from having money to attract academics.

However if by "Top 100" you mean they can identify, recruit and educate undergraduate students, that is not so obvious. Cambridge can. Perhaps MIT can. That most American universities can is not entirely clear. There is a clear divide between mediocre teaching at high school and excellent conditions for academics. The cut off between those two does seem to be at the doctoral level.

Although the sad thing is the decline of German universities.

12 Georg August University of Göttingen
13 University of Heidelberg
14 Princeton University
15 Humboldt University
16 Ludwig Maximilians University of Munich
21 ETH Zurich
30 University of Zurich

I don't suppose they are winning many these days. Sanders' plan is clearly intended to copy Europe's "free college for everyone" plan. Which has been a disaster for places like Göttingen.

By top 100 I mean top 100. As in rankings. The rankings take into account many things. I don't necessarily have to agree with the rating methodology, but there it is.

Of the top 100, or even if you want to look at top 200 universities in the world...over half are American universities. And of course Harvard MIT, Stanford etc. But we're not even talking about those...exclusively.

Most state flagship universities in the US are on the top 100 list in the world. American universities by any objective or subjective measure are FAR FAR ahead of the rest of the world.

So how exactly does Matty think that "other places do it well"? What other place does it well? Europe has destroyed most of their universities and turned them into 4-year long kindergartens.

"The American college with the most Nobel-holding former students is Harvard. It is well ahead of everyone else with 73 alumni. But very few of them did their undergraduate degree at Harvard. Compare that with Cambridge which is in third place with 66. It is hard to find one of their Nobel winners who did not do their undergraduate degree at Cambridge."

Comparing Cambridge to Harvard is a bit of an apples to oranges comparison, no? Cambridge has about 12k undergraduates in 31 colleges in a country of 65 million people. Harvard has about 7k undergraduates in a country of 330 million people. If you're a UK kid interested in STEM, I would think that Cambridge is your clear top choice, while in the U.S., Harvard might not even be in your top 5.

Harvard University is a living monument to the memory of Oliver Cromwell and the Puritan mindset of the seventeenth century, which makes it the epicenter of evil in the western hemisphere.

Canada and Britain compare pretty well, once you account for population size. I think results might be skewed somewhat due to the sheer utility of studying in the most global of languages (history matters), but clearly there are a lot of things going right in these places.

4. This is really a PR job.

#6...I'd like to recommend a blog called ECONOMICS IN THE REAR-VIEW MIRROR ARCHIVAL ARTIFACTS FROM THE HISTORY OF ECONOMICS done by Irwin Collier. http://www.irwincollier.com Among other things, he gives the reading lists and, sometimes, notes from economic courses in the US beginning in the 1880s. I've discovered some real treasures dealing with my favorites Henry Simons, Frank Knight, Jacob Viner, and Joseph Schumpeter, but all the postings are interesting, including at least one on Ely.

Here's a gem...HARVARD. ECONOMICS OF SOCIALISM. MASON AND SWEEZY, 1938 James Tobin was a student in that class. Look at Sneezy's reading list...

Apr. 25-May 7
Sweezy Two weeks to be devoted to the following topics:
1. Marxian and Orthodox Economics
[Handwritten note:] Rev of Ec Studies June ‘35
2. The Allocation of Resources in Socialist Society
Lange, Marxian Economics and Modern Economic Theory [Handwritten note:] Rev of Ec Studies June ‘35
Hayek, Collectivist Economic Planning, Chs. I, III, V
Pigou, Socialism versus Capitalism

Reading Period:

Sidney and Beatrice Webb, Soviet Communism, Vol. II

3. Not sure why computers are relevant. Caterpillar could build the greatest offensive lineman the world has ever seen for about 40 grand, but the NFL still pays millions for left tackles.

Sure, and "Paul Bunyan" machines get more efficient every year. The #2MA question is whether we are all Paul Bunyan ...

Yes, we are all Paul Bunyan eventually, and at that point all bets are off. But that's not a useful way of looking at the social problem in the meantime. What happens is that lots of people are close to being Paul Bunyan, some change happens, they become Paul Bunyan, there is capital substitution / augmentation and then the displaced workers have to switch to other jobs that are a little further out on the Paul-Bunyan-Quotient dimension. So workers are constantly re-equilibrating to doing all non-Paul-Bunyan work, as they have been doing for centuries now.

But insofar as regards social stability and the smoothness or disruptiveness of this re-equilibration, the questions become ones of pace, scale, and human potential. How many workers are how close to being Paul-Bunyanized? What if they can't, for one reason or another, easily and quickly switch into the remaining non-Paul-Bunyan fields? Alexander Field argued that this was an important - and often overlooked - contributing factor to the Great Depression.

The labor market overall has a distribution for something like this that would measure how close any particular job is to capital substitution. It would be like an elasticity at any particular point on the optimal capital-to-labor-ratio output-productivity curve. You could ask "If we move the curve up 10% through technological innovation, what percent of labor needs to find something else to do? And what if we changed the ratio of wages to (interest+depreciation) on the marginal machine by 10%?"

What people are worrying about now is that it seems that these elasticities are suddenly higher than we thought, and will affect more people than we thought. Judging by the current canaries in the coal mines, the transition does not look like it will be a smooth one. We are richer now than in the depression and can bail people out with subsidies, but that's like keeping the them barely alive on life-support. Canaries need to sing to flourish, but they can't do so while intubated.

A nice, sell written, overview.

I don't think so. Making an offensive lineman is beyond our technology today. Atlas is getting close, but we're still a few years away. Putting a giant bulldozer on the field and saying it is an offensive lineman isn't going to fool anyone.

When we can fool people with robots playing football, and other sports, well then things may get interesting.

I have no doubt when the robots can do it they will be required to play football, as it's too dangerous for humans. This may be mandated even sooner than you'd expect.

As predicted back in 1991

"1. The culture that will be Toms River, New Jersey."

I agree that it seems like a fairly creepy cult if the complaints are accurate.

"a fast-growing Orthodox Jewish community that’s trying to turn a swath of suburban luxury 10 miles (16 kilometers) from Atlantic beaches into an insular enclave. The rub, a township inquiry found, is “highly annoying, suspicious and creepy” tactics used by some real-estate agents."

"They show up on doorsteps to tell owners that if they don’t sell, they’ll be the only non-Orthodox around. Strangers, sometimes several to a car, shoot photos and videos. When they started pulling over to ask children which house was theirs, parents put an end to street-hockey games."

"His account of dropping by a neighbor’s open house and being denied entry by its Orthodox listing agent is included in a 16-page report on real-estate canvassing issued by township officials Feb. 5."

"The Orthodox dominate Lakewood’s school board. Though most schoolchildren attend private religious school, the township provides free, gender-segregated busing, which helps account for about half of a $12 million budget deficit. Some Toms River residents fear a similar drain."

"Though most schoolchildren attend private religious school, the township provides free, gender-segregated busing"

That's just wasteful. You can't segregate people on the same friggin' bus? Boys in the back, girls in the front isn't good enough?

They say they wanted diversity.

Let's celebrate it; even if we aren't invited to the party! /s

"large families that ... limit interaction with secular society": what, even to the extent of refusing all government handouts?

Nope. Come to Brooklyn they love the handouts.

Steve Sailor incoming in 5 ...

1. In the 1960's we used to call the tactics "block-busting". Material there for an essay on the different American dreams.

#3 It tickles the fancy of well-to-do economics professors who see themselves as dilettantes in all that life has to offer, who like to ooh and aah at all this human diversity, omg.

But the Go schools are child abuse, pure and simple.

I've visited a number of go schools in Japan and China. Young students were happily engaged and energized by their learning and playing. I've also met dozens of former go school students, from those who dropped out to those who became professionals. The one common trait is a love of the game and an enthusiastic engagement. But, it's not a productive activity in the sense of building careers and contributing to GNP. It's more accurately viewed as an artistic and cultural activity, which has its own, different value.

2. It's worth reading Sanders's actual plan...

to see that his "plan" for funding this using a 0.5% tax on stock trades is absolutely ridiculous. Even the leftiest of left-wing economists thinks that while a good idea, the Sanders FTT is too large by an order of magnitude. That, at the rate he proposes, it would almost certainly lower government revenues.

But in my view, that is besides the point. Market-makers don't owe you anything and never got bailed out. They pay income taxes already and we don't have other federal taxes on gross transaction volumes.

Plus, this is a matching funds program. The Federal government only covers 67% of the cost. The state needs to provide the other 33%, which means *state taxes* also need to be raised somehow. As Yglesias also points out, to get matching funds, universities need to reduce reliance on lower-cost adjuncts. (Strange requirement in a program intended to make college *more* affordable. Sounds analogous to an affordable health care program that required greater use of specialists and less use of nurse practitioners.) So, if tuition currently totals $70B, taxes actually need to increase by *more* than $70B to cover the increased costs.

Separately, why would we expect incidence of FTT to fall on market makers? Wouldn't we expect a significant fraction to fall on investors in the form of bid-ask spread?

Your question is a good one and ultimately it's a dynamic problem.

Currently we have many securities and derivatives markets that have very tight spreads. Bid-ask spreads of roughly 1-5 bps (.01%-.05%) are basically the norm. Market-makers are able to capture some fraction of this. A rough figure is something like 0.25 bps profit per traded contract or share. So obviously market-makers cannot quote this tightly under a 50 bps tax. If they quoted a 100 bps spread they would be able to break even on a round trip. If they had some directional alpha they could perhaps make a small profit (perhaps 0.25 bps again) at this spread.

(Keep in mind that market-making algos are not just passively quoting. They are generally quoting larger on one side, or closer to the mid, or holding inventory at some times and dumping it at other times, etc. In order to make money market-making in today's competitive market environment one needs to have some directional alpha. Just passively quoting at the top of the book is certain to lose money due to adverse selection effects.)

So imagine that market-makers are collectively quoting a 100 bps spread. Suppose the mid price for XYZ is $100. That means the bid is $100.5 and the offer is $99.50. Next suppose that I want to buy XYZ and hold long-term. I can buy for $100.5 or place an order for less than $100.5 and hope it gets filled. If I quote somewhere in (99.5,100.5) then I will now represent the best bid and will have priority over all the lower priced bids. A similar thing happens with long-term investors looking to sell on the other side. So the observed bid-ask spread will actually be consistently tighter than the round trip tax in active markets. This is readily observable in e.g. Hong Kong's stock exchange where there is a sizable tax. What you won't see is consistently tight spreads. I would clearly not want to post an order to buy at e.g. $100.49. If I am a long-term investor the penny saved is not worth the risk of the market moving away from me. I would just take the offer price and get immediate execution. Some equilibrium is ultimately reached, but market-makers are basically shut out of the market. This is empirical. I know a lot of market-markets and none can make money in Hong Kong. And the spreads tend to be around half the roundtrip tax.

Tight spreads ONLY ever occur where market makers are able to price very efficiently and make money by doing high volume at slim margins. With a large tax on transactions, the bid ask spread becomes wide and the fair price is always inside the bid ask spread. That means no market makers. They lose their livelihoods because they dared to enter a politically unpopular profession.

The cost of capital goes up throughout the whole economy in a direct and obvious way. It's an economic disaster, but I guess I'm more hung up on the whole injustice of it all.

Perhaps you would be well positioned to make the argument in favour of how these market makers benefit the economy. Like, I don't care if they're personally making money or not, what value does this contribute to the economy? I understand the importance of price signals, but aren't most of them basically gaming the system in a zero sum manner? If the buyer wants to buy at 100.49 but because of sliiiightly less liquidity buys at 100.50 to ensure the buy, is this really relevant at the aggregate level of economic efficiency?

It seems to me that a lot of very very smart people are caught up in zero sum games rather than doing more useful stuff. But you seem well positioned to counter this perspective, if it is possible to do so.

Before getting into a deeper argument, I think there are some pretty straightforward things you are ignoring. First, market-makers purchase goods and services, pay taxes, raise families, give charity, etc. so the burden is on you opponents to show that they are taking something out of the economy. It seems quite clear that their presence is additive. There is also the basic matter of individual liberty. I'm amazed by the hubris people display when it comes to this area. Why should anyone have to justify their occupation to self-righteous leftists?

But getting to the more narrow issue. You want to know how market-making specifically benefits the other participants in the markets. But you seem to answer your own question. They lower the cost of trading by competing with each other to offer more attractive pricing to investors and hedgers. That one penny saved on one share adds up to quite a bit of money when you consider the sheer size of the stock market (or bond market, or commodities markets, or...). Market-makers aren't on welfare. They don't need your blessing or subsidy to operate. They justify their own existence 100%. They offer the best pricing and that is why people prefer to trade with them. You think that there are too many market-makers (how many would you say there are?) and that they are driving transaction costs too low. That is a very silly thing to assert. The market-makers are happy to quote wider spreads! They would prefer that the cost of liquidity was higher and that they could capture a bigger spread when they make markets. Unfortunately for them, market participants demand extremely tight pricing. And they are willing to shop investment products, exchanges, etc. to get the lowest costs.

The aggregate level of economic efficiency is obviously higher when the cost of capital is lower. You may not think that a 0.05% change in the cost of capital is significant. But somehow you think the cost of having market-makers is somehow greater than whatever is gained there. Why?

Thank you for the response. I have a few counterarguments.

Rregarding the claim of tax payments implying that there is a contribution to the economy. The zero sum nature of the game is critical: consider that the loser of that zero sum game makes less money, and therefore the expected value of taxation should be nil (assuming that everyone involved in these games is in a high income tax bracket - your gain is their loss and all... ). Presumably similar reasoning applies to resources for raising family, charitable givings and the rest.

On the matter of price signals, I'm not at all convinced. The role of price signals in the economy would still be accurate on average even the price signal were issued only once a day. However, it seems a lot more practical to have trades occurring very regularly throughout the day. What might setting the price in 0.00001 seconds in a zero sum high frequency transaction game do for price signals that cannot be done in, say, 1 second, or even a few minutes for that matter? Can spreads not be narrowed in an enviroment where all players are required to operate in a 1 second timeframe, disallowing any from having a nanosecond speed advantage?

You make the argument that it's lowering costs. I don't buy the argument, but perhaps it's worth fleshing out a bit? How is it lowering costs for whom? I'm seeing this as zero sum, in which case lower prices for one party implies higher prices for the other party, implying zero contribution to the economy.

You argue that 1 cent times large volumes is a large amount of money. This is clear. Hence, the interest in huge investments to get 0.00001 seconds ahead and beat the competition to the game. I understand the point about $1.505 and $1.495 actually being a large spread for a large volume of transactions. But I'm really narrowing in on the high frequency high speed transactions, where billions are being invested for faster networks. For consideration of price signals and liquidity, is there an argument that doing this all in 0.00001 seconds instead of just 1 second as actually improving the price signal in some substantive way?

This is not a left/right thing for me, rather, with a background in economics, I am inclined to look at an area of economic activity and wonder whether it contributes to current happiness or future production potential. I readily acknowledge the apparent logical contradiction, but I am not asserting a negative moral judgment, rather, a positive moral judgment of a possible alternative allocation of resources. This may sound foreign to you, but I assert that, indeed, an area of economic activity actually DOES have to justify itself to society, especially once the subject of "unfairness" comes up, in this case where huge investments are made into ultrafast trading in zero sum games which produce no additional value.

My goal is not to lecture or attack people involved in this zero sum game, which is perfectly legal. There is nothing bad or evil about it. I'm really and truly looking for "pro" arguments, because I'm not aware of any, and as you can see, I'm not convinced by the ones you have presented.

My sentiment of the matter is that any prospective trading tax should be high enough to deter wasteful investment in such zero sum games, but not so high as to negatively affect the price making ability of the market. Is 0.01% enough? 0.001%? Not any higher than it takes to reduce the "misallocation" into network technologies designed to win zero sum games.

Why not apply the same reasoning to the obvious advantage of investing lots of money into getting good information, which makes it "unfair" for mom and pop or smaller players? Easy. This has an obvious and strong benefit in working towards a correct price signal, ensuring good allocations of capital across the economy.

Here's my "pro" arugment. This investment could be driving the leading edge of network technologies. But I don't think the industry tries to make that argument because it will just draw more attention to the zero sum nature of the game. However, I don't actually advocate for a transaction tax, although I lean heavily to favouring it, as a result of the network technologies innovations this might drive. Having countered each of your arguments (I think, but perhaps you do not find it convincing), I'm honestly curious to expand the "pro" side of the debate.

Short: it is natural for an economists of various stripes to be concerned about the optimality of deployment of capital in the economy. Zero sum transactional games do not contribute to the economy, which is much different from having winners and losers due to higher quality information and more accurate price signals.

FINALLY, to answer YOUR question. OK, maybe 0.05% is too high. 0.001%? I repeat myself somewhat. The value of price signals to the economy is based on quality of information. It's not actually important for economic efficiency if there's a 0.05% spread at noon today, or lasting until next week. What is important is that the price generally settles at the "right price" given a rationally large amount of information dedicated to determining this price. Also repeating myself, traditional advantages arose from amassing higher quality informational inputs, whereas investments in network speed provide no informational value whatsoever. You may as well be jumping in front of people at the store, buying up what they were just about to reach for, and then demanding a premium if they still want to buy it. The resulting income may be taxable, but that does not imply an actual contribution to the economy.

Yes, actually, economic actors have to justify themselves, especially in sectors which dominate the economy. Not you personally, I would never in a million years hold you to that personally - it is legal and there is nothing remotely evil about it. But I'm trying to understand whether recent changes in that entire sector are positve from the societal point of view, which I clearly doubt. If you're contributing nothing to anything, then the activity should be discouraged so that you will produce things that make people happy. Playing zero sum games is as useful to the economy as being homeless, with the unfortunate situation that all of these bums are otherwise truly brilliant people with huge potential contributions to society.

What value does you or any of us posting on the Internet has to the economy? Perhaps Bernie should impose a $0.50 tax per comment/message/email? Why not, apparently the only thing that matters is GDP utilitarianism.

Nathan, your response concedes all of Mikes points but you simply ignore them become you aesthetically disagree with market making - probably because of your anti-market bias. Let me paraphrase your response to Mike, "yeah market markets reduce costs for all buyers and increase price signaling, but not by much and I don't like them anyway". Typical emotional authoritarianism.

Your suggestion that market-making losses that offset income cancel the tax revenues from market-making gains is just plain absurd. The size of losses among market-makers that lose over a given period is dwarfed by the gains of the winners. The billions of dollars of write-downs from market-making losses exist only in your imagination.

OK so you've conceded that 0.05% of the entire securities and derivatives universe is an enormous amount of dough. Now you're down to .001%. Moving the goal posts I see.

As for your efficient allocation argument. That's just really freaking authoritarian. I'm in favor of taxing clear negative externalities. You are just off your rocker. Let's tax Hollywood and the NFL too. How about a tax per pass attempt because we like rushing. No one benefits from it in your little mind so let's tax it. Jeez. There should be a very high bar for taking away someones economic freedom, no?

You also seem confused about how all this could not be zero sum. Keep in mind that transaction volumes are not independent of available liquidity. In markets with higher trading costs, people hold bad investments longer and are slower to get into good investments. When market-making isn't competitive, "specialists" with advantaged positions at exchanges capture the spread. This is pretty basic markets 101 stuff. The fact that speculators / professional traders provide liquidity and price discovery has been well known for ages. You seem to think that once the cost savings are small enough they become their own cost. Get it through your head that in very, very large markets (hundreds of trillions in value) cost savings of a fraction of a percent are EXTREMELY meaningful. And the amount of resources that go to producing these savings are in proportion to the savings, not way beyond as you imagine.

And yes, you are right that there are obvious positive externalities in the research and tech that these people produce. In machine learning, software, networks, databases, etc.

You should meditate on the distinction between frequency and latency.

High "frequency" is what you absolutely expect to see in a low-margin business. That is markets working. You also expect to see automation / algorithms reducing costs that are passed on to market participants.

It is possible that there is some deadweight loss on the latency (not frequency) side. We don't want less trading and higher costs. We don't NEED extremely low latency. Many agree on this. There are venues that are trying to create markets that don't force market-makers to compete with each other on latency. Most market-makers would prefer not to have to spend on microwave data, etc. Taxes are the absolute worst way to achieve this.

Thomas - I don't claim that commenting on the internet contributes to the economy in the strict sense of the word. However, this is not a zero sum interaction. You learn (if I'm not obstinate), I learn (if I'm not obstinate), ideally leading to better information and citizenship (this does not seem to reflect the actual way things are working these days for the most part, however).

Yes, you may well charge "what use is that to the economy?", to which I may reply "well, I'd be happy to spend more time contributing to the economy, but it's hard when you're extremely outspoken about many things on both sides of the political spectrum and are mainly specialized in areas relevant to policy analysis". However, I have a small client base of some academics who appreciate this in editing and translation work, and supplement my income with teaching.

Moreover, I do not concede his points. I understand the importance of having an accurate price signal. I disagree with the premise that it matters if it takes 0.0001 seconds or a few hours to transmit this signal, or whether a 0.05% spread really matters for long-term price signals, since it really doesn't matter for efficient capital allocations if you have to wait a few hours or day or two to get the price your information tells you is right. However, it is good for information analysts to be able to profit from this information fairly quickly, because it encourages them to get as good information as possible before others do.

There is no anti-market bias in what I say. Rather, I communicate doubt that 0.00001 second trades are required to achieve accurate pricing in the market. In fact, I have a rather pro-market bias - I want more equal footing in transaction speed so that information, not network speed, defines who wins the "game".

Mike, if you gain $1000 in the trade, someone else is making $1000 less. This is the definition of zero sum. The idea that you're creating money in a zero sum game just defies any sort of logic I'm aware of. HOWEVER, if this reflects better use of information, then it contributes to more efficient capital allocation, and being able to capitalize on this information quickly offers a strong incentive to get the best information possible. I'm not saying it's evil to win a zero sum game where all entrants know the rules of the game, I'm questioning the economic roles served by the associated activities.

"Now you’re down to .001%. Moving the goal posts I see."

The point isn't the specific amount. That is an empirical question. The point is that it is not good for the market if everyone is allocating resources to get their signal transmitted 0.00001 seconds faster than investing resources into better information production and information analysis. I would be equally contented with a rule which required all traders to have precisely the same latency. I'm not anti-speculation, although this can have its own problems, I'm pro-fair, in this case additionally motivated by the view that traders should be investing in information quality and analysis, not ultra fast latency.

"As for your efficient allocation argument. Thats just really freaking authoritarian. "

That's what finance is for. You're supposed to be making money by producing high quality information and information analysis, which results in zero sum gains and losses between traders, but is very useful for efficient capital allocation. Moreover, painting it as authoritarian is absurd. All decision makers remain 100% free, in the free market, to allocate capital as they see fit, on the basis of the best available information about the most profitable/efficient allocation of capital. You could hardly be more anti-authoritarian.

Like, most states tolerate gambling. But, what if you're playing dice and only one player has the technology to take a quick peek a microsecond before the dice finally settle, and update their bets accordingly? That's a stacked deck, and there is hardly anything more anti-market than a stacked deck. Because, markets require fair play, in addition to freedom.

"In markets with higher trading costs, people hold bad investments longer and are slower to get into good investments. "

I understand this perfectly well. I would not support a tax that had any meaingfully negative influence on this. I'm pinpoint focused on the games which try to get 0.00001 seconds ahead. Holding a bad investment for 2 seconds instead of 1 is not meaingful from the perspective of capital allocations in the long run. In fact, holding a bad investment for days or weeks longer is not even particularly meaningful at the level of the aggregate economy. The real delay is in amassing sufficiently good information to get out of a bad investment or enter a good investment, not the rapidity of the transaction once the information is anlayzed and it becomes clear that it is time to move.

Also, I perfectly well understand that 1 penny times 1 trillion transactions sums to a large amount of money. But you have not made the case that this is not zero sum. 1 penny times 1 trillion transactions is a large sum of money for both the winners and losers, and you have not made the case that additional value is provided.

Look, I'm perfectly open to the argument along the lines of "it's legal, and there's nothing remotely evil about it". But the notion that it is "authoritarian" for regulators to provide minor judges better resource allocation in society is not reasonable.

While in this discussion I'm very much in the role of putting "you guys" on the defensive, my actual interest is to develop understanding of "pro" arguments. I think it is reasonable to suggest that it is natural to be "bribed by self interest" as Francis Hutcheson would say, and you lend more credibility to your own arguments then is perhaps warranted. Also, the fact that the two people who took the other side of the argument basically opened with "what, do you hate freedom that much? Dictator", in response to a minor judge, suggests that deep down you MIGHT not actually believe the pro arguments as much as you feel.

I'm not sure that we're going to get much further with this now, but 1) I'm interested to understand the perspective of how it is not zero sum for high speed transactions to take profits out of the hands of market makers, with specific consideration that it doesn't really matter for the quality of price signals if it takes 0.00001 seconds to "get the price right" or if it instead takes a few hours or day or two. 2) I'm also curious if you might further defend the perspective that it is "authoritarian" to provide incentives which prioritize information quality over transaction speed.

Admittedly, I'm exposing some ignorance in speaking of "high frequency" as opposed to latency. What I really mean to be talking about is the latency question. The advantage of latency means that many firms must reallocate resources away from information analysis and into reduced latency, which means that the actual quality of the price signal can be expected to be lower. I'm highly skeptical that all the fancy algorithms actually improve the quality of price signals or the ability of markets to reach the "right price", but I'm quite willing to give the benefit of the doubt here. Not so for re-allocations away from information quality and information analysis (including algorithms) and towards latency reduction. The "high frequency" bit enters the equation because that is part and parcel with the low-latency winning strategy. High frequency is not in and of itself a sign of a problem - it could reflect a very large number of different actors trying to incorporate new information and someone with better information can profit from high frequency trades which ideally are profiting due to the fact of others having lower quality information or analysis.

Sanders has no idea what he is doing. All he knows is he wants to have power over others resources, badly. He knows if he gets into the presidency he will have at his disposal the full power of the military to enact his wishes and he can't wait.

Whether it is a good idea, a long run good idea, or whatever matters not. What matters is that it is his will that is being enacted.

LOL....yeah when Congress doesn't pass his laws he will just use the military to stage a coup. Very likely to happen.

I think a 0.05% trade tax could be justified, in order to combat unfair trading practices by people with super fast connections in zero sum games which add essentially nothing to liquidity or pricing accurarcy.

But 0.5%? I think that would definitely stuff up markets and signifincalty reduce the ability of the market to achieve "correct" pricing, especially in the short run. I don't really care if the investors are losing 0.5%, what matters is price signals. 0.05% will not affect price signals, but will end unfair trading practices. That it raises revenues should be seen as a trivial afterthought. Too many billions are being thrown at zero sum high frequency trading, and this should be disincentivized (taxed)

What you don't know about this topic is a lot.

I'm no expert, as are 99.999% of people. But the point is that billions are being spent on infrastructure, computing and human resources to get the trades 0.0001 seconds quicker, so they can observe a move that is soon to come and game the system to play the profitable side of a transaction, like, the last trade was at $1.0005, they observe a bid for $1.0007, then scurry in to snap up shares at buy at $1.00055 then get in some safe sells at $1.0006 or $1.0007. That's basically what's happening, no? What I don't understand is the precise nature of how they interpret/receive the signals and sneak in for the microsecond buy-sell to get their zero sum win in the bag.

They are not making the market. They are observing a price signal and squeaking in ahead to rob gains from the legitimate price signallers. That is not beneficial, and is a net loss to society because resources could be allocated towards more useful things instead of these zero sum games.

I'm pretty sure that most financial consultants or even traders would struggle to properly explain this stuff, so if you need to refine the description to more accurately describe the practice, then this would be entirely welcome. There are a couple finance people on this forum sometimes, so maybe one wlil deign to fill in some gaps to us non-experts.

No one has to explain shit to you. You have absolutely no concept of individual liberty. It's appalling. You are ignoring your buried premise that because this practice is scapegoated by politicians that it needs to be on the defensive.

Your fantasies about market-making are absurd.

"They are not making the market."

Yes they are. It is necessary to have low latency and transact high volumes to make a living as a market-maker. Full stop. You can read up about IEX and EBS and other markets that have looked to mitigate latency advantages. It's a legitimate endeavor. It's something a lot of market participants, even many market-makers, want. Government regulations and tax have had nothing to do with these advances. In fact, government actions have cemented latency advantages. Your silly tax ideas just destroy the market. They make everyone far worse off. Terrible, terrible ideas. Get your head out of Bernie's ass.

And in your little anecdote. If you take out the market-maker. Do others get to trade at better or worse prices? Do they filled sooner or later?

The beauty of time-price priority markets is that when a new person starts quoting, everyone else gets to trade at better prices. It's one of the most brilliant inventions of modern capitalism and has enabled the most liquid markets on the planet. I know it's hard for you to wrap your head around why an Indian programmer with an FPGA is good for Americans. But just look at the spreads. Your only suggestion is to push the spreads 5x wider and hurt everyone.

Look, my background is economics, not finance. It doesn't matter for the economy if you get the price right in 0.0001 seconds or a week. All you're doing is beating someone else to the trade and capturing their benefit. Zero sum. But, in order to incentivize investments into high quality information, which ensures good allocation of capital across an economy, it is desirable for financiers to be able to capitalize on new high quality information fairly quickly. Is an hour fast enough? Probably, since that information probably takes months and years to put together. How about a second? Man, getting down to nanoseconds, it is beyond plausible that this is nothing more than zero sum shenanigans with no value to the economy, and many billions are being thrown at that nanosecond advantage.

Yes, I'm asking you to justify yourself, or, rather, changes in the field in recent years (of course, you are personally not liable for doing things which are perfectly legal and not at all evil). That you get all angry and resort to "liberty" as your first argument suggests that you do not actually truly believe that this is contributing anything valuable to the economy.

I believe in liberty. How about a 0.05% tax on your liberty? Just high enough to prevent the misallocations to zero sum games through higher network speeds, an entirely different sort of investment than investments to ensure that capital markets are well-informed. We're hardly discussing a command and control economy, and you make it sound like I want to ban the entire profession and rule your personal life or something crazy like that.

I repeat. It really doesn't matter to the efficiency of capital allocations if it takes 0.00001 seconds, an hour or a week to get the right price signal through. But to incentivize financiers to invest in high quality information, we want them to be able to captalize on it as soon as they get it. There is no information process in the world that relies on a nanosecond advantage, after spending months and years to get that information right.

Focus on quality of information, not transaction speed. That is what is good for the economy, and it is 100% absolutely the job of regulators to look out for the public interest in such regards.

Look, I don't see this as a stream of taxes, I see this ONLY in terms of influencing poor economic allocations.

Mike, a 0.05% tax wouldn't matter if you're investing based on fundamentals. It would, however, put quite a lot of day traders out of business. They will find something more useful to do which actually benefits the economy. Perhaps they will revert to their traditional roles in hoovering up information that helps to improve price signals and capital allocations.

If a 0.05% tax is fatal to your business, obviously it's not of much value. Given how smart people in the field are, this is a tragic waste of potential.

You seem to be concerned that people spend money to reduce latency. That it is deadweight loss beyond a certain point. That isn't a crazy suggestion, but why do you assume that the government has to tax everyone by an order of magnitude more than the amount spent on this tech to reduce the spend? The spend on latency-saving tech is already constrained by e.g. bid-ask spreads. There are MANY exchanges (e.g. IEX, EBS, do your research) that already have certain features that save participants from telco investments that reduce latency. There is a huge difference between, "I speculate that a benevolent dictator could better allocate resources" and "this should be illegal". I see no reason why markets themselves can't address the allocation issue here. Again, we are already seeing that at many venues.

You are also obsessed with the idea that these investments somehow rob average Joe investor. They don't. The exact opposite is true. The market-makers need to spend on latency so that when a massive hedge fund goes to trade, they don't get run over. They can't just quote wider because some other market-maker that has lower latency will always quote tighter and capture all the flow. The low-latency allows the tight quotes that benefit average Joe. Low latency has no negative effects on average Joe's retail order that gets executed off-exchange almost all the time anyways, at better pricing than is available on the exchange. (Again, do your research before insisting on destroying markets and jobs).

"If a 0.05% tax is fatal to your business, obviously it’s not of much value. Given how smart people in the field are, this is a tragic waste of potential."

Again, another inane comment. The point here is that it isn't a 0.05% tax on gross revenue, or profit, the way a normal tax works. It is a tax on gross transaction volume. An irrelevant quantity. Transaction volume is something we want more of, not less of, to achieve liquid markets and low cost of capital for the economy. And low transaction costs for investors.

You are also imagining that the resources that go into market-making exceed the cost savings. This is a very illogical thing to suppose. Where does all the extra profit come from. You throw around words like "zero sum", but frankly, you are extremely ignorant about all this and you are completely reasoning backwards from your conclusions: HFT BAD, GOVT GOOD

Mike - I think there is a major flaw in your reasoning. The "benevolent dictator" would not be allocating resources. Rather, this is regulators providing a small incentive away from one activity (low latency) and towards another (high quality information and analysis). All decision makers would remain 100% free in their decisions, but the rules of the game would be slightly tweaked.

Re: Joe trader. I agree that this is essentially irrelevant. Joe trader suffers from such woefully lower quality information and analytical capacity that the difference in transaction times is irrelevant. Moreover, Joe trader might buy a few thousand dollars in shares at a time, and I assume that such trades are essentially inconsequential in any of the related discussions. Rather, I'm concerned about an even playing field between institutional investors, the ones with the deep pockets who must contintually allocated resources into lower latency rather than information and analysis. Joe trader is already so utterly outgunned that this is such another straw on the camel's back.

"The point here is that it isnt a 0.05% tax on gross revenue"

I perfectly well understand that. If the contribution to price signalling quality and liquidity is completely undermined by a 0.05% transaction tax, then the purported benefit cannot have been anything other than marginal. Observing market volatility in any stock, commodity or index demonstrates that price signals are with very high certainty not even remotely this accurate.

I also perfectly well understand the benefits of liquidity. But it seems that the marginal utility of such utiltiy is approaching zero, and there are external costs because so many firms must invest in latency reduction instead of information quality and anlaysis.

Normally, one would expect the companies on the losing side of the latency game to make much noise to fight for an equal playing field. But, I assume that this doesn't happen because it's equivlaent to transmitting that the trading company is presently systematically disadvantaged in its trading operations, and this could lead to a major outflow of capital. Strangely, a situation where behind closed doors is almost certainly the best way for them to advocate for their concerns.

On zero sum and "where the profits come from". Indeed, where are the profits coming from, if not out of someone else's pocket without creating any additional product? Honestly, I wasn't aware that traders would argue that this is not zero sum, which it patently is, and imagined the focus would instead be on the quality of the price signal required to win in this zero sum game.

This is such an inane comment it's unbelievable. Everyone should be strictly worse off so that no one can have an advantage? That's amazing.

Generally, pigouvian arguments about taxation are aimed at curbing negative externalities. People wasting there money (as you see it) is not a negative externality.

It's not just a waste of money. It's like a fast lane for high fliers and a slow lane for the bums. It speaks out strongly to principles of fairness.

Suggestion: financiers shall be held to a speed limit of 80% of what everyone else has to drive at until they resolve the problem. Fair?

But that's not even a remotely close analogy. Because the speed advantage gets you to the finish line first every time. The better analogy would be that financiers shall be held to a speed limit of 0% of what everyone else has to drive until they resolve the problem. Fair?

There is nothing remotely strange about taxing/regulating things which are suboptimal allocations or subsidizing/regulating things which improve optimality. Why should finance be the exception? I mean, please go ahead and advocate for yourself, but lets not pretend that there is a benefit where there is none. If you hate the tax so much, how about a regulation which requires that all orders take precisely 1.00000000 seconds to get from the computer to the exchange? Then, traders can focus on what they were supposed to be up to the whole time - applying new information to ensure that price signals are applied without too much delay.

Bizarre. If your concern is value investors there is no problem at all. If your concern is that you get angry (look at your language lol) that Joe schmo is being 'robbed' by market markets who reduce spreads and make it cheaper for Joe to buy or sell, well there is so much irrationality and emotion to unpack that it would require a full non-Marxist education in economics.

I can't build as good smart phones as Apple. They shouldn't be able to make a phone that is 80% as good as the one I can make.

Not everyone has to be equal in their ability to play in the smartphone manufacturing and design market. The suggestion is absurd on it's face. Similarly, not everyone has to be equal in their ability to compete in electronic market-making. If the public was being hurt, then there might be reason to step in. But they aren't. They're benefiting.

Is finance in the business of providing low latency transactions or amassing high quality information and analysis?

If it is the first, the firms should be trumpeting their low latency to the world to attract investors. But, they will not do that because it will attract blowback. Rather, they will make investments internally to achieve internal gains.

Imagine that Apple were in fact selling phones based on the premise that they offer uncrackable privacy? How about if they were doing so without really saying so, as opposed to trumpeting "the FBI and NSA will never in a million years crack your phone - so buy it"? How long until regulators stepped in to interfere with the market? It is normal for regulators to step in when major segments of the economy are doing things that are deemed socially undesirable.

2. Under Sanders' plan, the gravy train (euphemistically called "higher education"), with high salaries to professors and administrators funded by the federal government subsidies for tuition to colleges and universities, public and private, would end. Would that be a good thing or a bad thing? Are college professors and administrators overpaid or underpaid? If you believe they are overpaid, Sanders is your man; if you believe they are underpaid or paid just about right, then the status quo is your man.

Exactly. Sanders plan would destroy higher ed in general. Not destroy it in the way dumb conservatives or dumb leftists think it will.

It will destroy it by eliminating the incentives for smart people to devote their time to doing research, advancing knowledge etc. I.e., Sander's plan would make America's universities like Europe's universities: 4-year nannies for overgrown babies.

The value of American higher ed is in the research and knowledge it...CREATES...not so much in what it teaches. Teaching is easy. Go to community college if that's all you want.

Oh I forgot: Leftists are "pro-science", even though they don't realize that that science comes by paying people hefty sums to engage in it.

You can't think of any better way to subsidize scholarship than by overcharging undergrads?

As was linked to above: The Net Tuition and Fees at Public 4 year colleges in the US was around $4K last year. The gross is around $9K per year so the average undergrad already gets over half of tuition & fees covered.


$16K for a college degree is a great deal. It's about half the cost of the average new car sold last year.

Yep. OVERCHARGING!!! $4k net price per year, to get a degree which on average provides over $1 million in additional lifetime earnings.

What incredible nonsense.

No way is it $4k 'net' per year. Not any credible university. Did you even go to one?

"No way is it $4k ‘net’ per year. Not any credible university. Did you even go to one?"

Did you bother to actually look at the link I posted? That data is from collegeboard.org from a wide data set.

And secondly, yes I have two different degrees and for the first degree my net costs for tuition and fees per year was approximately $0 when you factor in Pell Grants, scholarships, etc. The second degree was mostly out of pocket, so my average was roughly $6K per year.

Both were major American universities.

It's definitely not 4k per year. I mean, I've read countless articles from the white, female children, of very wealthy families who had to pay full freight at undergraduate and graduate institutions in pursuit of gender studies degrees. Seriously, I've read countless BuzzFeed articles and a buzzfeed article is just as reputable as words out of President Obama's mouth.

What percentage of all scholarship is produced by public 4 year colleges?

"What percentage of all scholarship is produced by public 4 year colleges?"

I'm assuming you meant the standard dictionary meaning of scholarship.

Per the dictionary - scholarship: academic study or achievement; learning of a high level.

Assuming were referring to the US & that faculty percentages mirror student percentages: Public colleges account for roughly 75% of college students and 4 year institutions account for roughly 60% of students. So, roughly 45%.

It's clear you know nothing about the topic. Try sticking to whatever it is you actually know about.

Well I was certainly schooled by the internet expert. With my links and facts completely overtaken by your off hand cryptic comments.

I don't feel any need to educate you. You and I are both aware that you are not familiar with the scholarly landscape in the United States and yet somehow still feel qualified to opine on it.

"You and I are both aware that you are not familiar with the scholarly landscape in the United States and yet somehow still feel qualified to opine on it."

LOL, the topic of thread was specifically about the Sanders plan for free tuition and if it would have unexpected negative consequences. You've spent the whole thread making cryptic Socratic questions.

If you've got a point you should state it clearly. It might be good or it might be bad, but at least it will come off as a genuine attempt to have a discussion.

Fine. Other than state flagships and a handful of second tier in e.g. California, very little scholarship is produced in public colleges. Many non-flagship four year public colleges have 4/4 teaching assignments and don't have any research expectations for hiring or tenure. Virtually no scholarship is produced by two year colleges and literally none at for profit colleges. Your use of undergraduate enrollment as a proxy for scholarship is completely untenable.

brad - each and every one of us starts at zero. 99.9% of us will never even know what the cutting edge is in more than just a handful of sub-fields, let alone contribute to pushing it further except in our roles as consumers and taxpayers which contribute investment towards pushing this cutting edge further.

Much of scholarship, almost the entirety of it, involves learning about things which are already known. You don't have to produce Watson's and Cricks to have a high quality academic program which instills knowledge which enables people to be, for example, good nurses, doctors, etc.

Did you read AIG's post? He was talking about scholarship, not teaching. As for teaching, the vast majority of the value added in a college degree is in the admissions department, not the classroom. We waste an enormous amount of money on zero sum signaling games.

5. Call me skeptical about the relative import/meaning of this finding. If you dick around with a large set of numbers long enough, you're bound to find some weird things about them. It's a probabilistic certainty, and the last digit of consecutive primes appears to me to be hyper-specific. If there are two few consecutive primes ending in "1" in prime numbers in the next hundred million primes, maybe I'll believe them.

But who knows. I've been reading a lot lately about how psychologists think a lot of their "statistically significant" findings are rubbish, so maybe I've just been primed to be skeptical.

I'm such a sceptic that I've always assumed that the claim that the primes cropped up randomly was just schoolboy hyperbole. As so often, I assumed, people say "random" when they mean "haphazard" or even just "I can't predict".

Yeah, I consider "random" to often be "I haven't got a clue why these numbers are all over the place". For many things, it's a fudge factor roughly equivalent to the "God did it" arguments of yore, or much worse, a theoretical abstraction used to justify the accuracy of many statistical methods in social sciences, never mind that the surveying methods are not in fact random (no people on mobile phones, no people with no phones, no people who don't answer unidentified callers, no people who chase surveyors away from the doorstep, no homeless, too many unemployed, lots of people with nothing to do and so don't mind to spends lots of time answering surveys ...).

Even in natural science, like, the Heisenbeg principle - OK, things appear "random" in a large dataset of observation, but honestly, do serious physicists actually believe that basic principles of the universe are essentially "random" in nature, and not in fact determined by some underlying principles?

Yes, they do. Not all of them, and many serious physicists disagree (most notably Einstein), but the majority of them, yes. The standard point of view on Quantum Mechanics is that it inherently contains *randomness*, not reducible to our imperfect knowledge or means of computation or understanding. This point of view has resisted several strong challenges, e.g. the Einstein-Podlansky-Rosen question, which was decades later answered by Bell and Aspect (among others).

Thanks. I thought that was just one of those convenient fictions they teach in lower level courses because reality is too complex. Like starting with the Bohr model as though it's real, and then later teaching you that it's wrong and things are more complex.

There is no need to be skeptical about the original research, which is important and beautiful in its field (full disclosure: I work in this field, number theory, though in a different area, and am a friend and occasional collaborator of one of the author). But skepticism is indeed warranted regarding the summaries given of this work in the popular press, and the paper linked to by Tyler.

Number theorists employ the word 'randomness 'in an informal, non-mathematical, sense: to say that the distribution of prime numbers is not constraint by structures, except trivial ones. For example an elementary-school student could easily prove
that there are only two primes ending (in basis 10, for simplicity) by 2 or 5, and none by 4, 6, 8, 0, but it is a hard theorem (Dirichlet's theorem, around 1830), to prove that besides their obvious constraint (coming from the fact that 2,5,4,6,8,0 ar not relatively prime to 10),mthere are effectively infinitely many primes ending in 1,3,5,7.
Most of the development of the theory of prime numbers has followed this pattern.
Finding structures is trivial, proving that the prime don't obey other structures take 99.9999% of the work. What is interesting in this paper is that it exhibit a new structure, which is certainly true because supported by both strong empirical evidence and some heuristics, but that seems new and for once, non trivial to prove. Paradoxically, part of what makes this paper surprising is that the author are not able to prove that this structure actually holds (and believe me, if it was easy, those guys would have done it).
While it is not exactly the first time that a similar situation happen (cf. Chebychev bias discussed by one of the article), it breaks the usual pattern (structures among primes are trivial, proving that they do not exist is hard), and this is sufficiently rare to be exciting news.

Hey, I was searching for an answer to these questions and couldn't find any. If you have a quick and easy answer that would be awesome.

1) If someone cracks prime numbers, is there an obvious replacement for crytography purposes?

2) Are prime numbers useful for anything else if their cryotography applications become nul?

Sure, prime numbers are useful for polling intervals to reduce communication spikes. I keep a print out of the The University of Utah's "The 1,000 smallest prime numbers" on my wall for just that use.

Also, Prime numbers are useful for cicada reproduction.

#2. Higher education at public schools is purely is a state issue and can easily be made tuition free without spending any more tax payer money just by cutting the per student spending and or by eliminating slots for some of the less promising majors or just eliminating some of the less promising majors.

VTW: http://un-thought.blogspot.com/2016/02/a-problem-with-federal-government.html

No, it can't.

Nor should it. Why do you want "tuition free" college? Why should people not pay for the returns they get? Cause politicians figured out that they can get the votes of the dumbest demographic of all: kids, by promising them more free stuff.

I really disagree with your insinuation that young people of this generation are coddled, and that tuition should be raised. Just because it generates a high return doesn't mean that it's not more expensive than it needs to be. Tuition has been increasing well ahead of inflation. Why were universities able provide university education at much lower prices a generation ago? Also, you may have noticed the post from Professors Tabarrok and Cowen about whether recent marginal increases in education investment provide skills or signalling. If part of the improved outcome is due to signalling, that part will cause an increase in the apparent ROR of a degree without helping average outcomes.

The question of whether students should pay the full expected return of their education is an interesting one.

There are two really strong counterarguments to this line of thinking.

1) Due to liquidity constraints, a "bias" for the present and risk aversion (you never really know that you will achieve the expected return), we can expect youth to underinvest relative to the "economic optimum", and

2) There are positive externalities to education - smart/educated people who network with smart/educated people create spillovers in their work on teams and knowledge exchange in the workplace and over drinks, etc., most famously observed in the Silicon Valley. Even if youth were perfectly rational with respect to their individualized expected returns in relation to point 1), this implies that it is nevertheless optimal for society to further subsidize their education in order to achieve broader social gains which the individual cannot capture.

In which case, we may be going in the wrong direction by saying "engineers make lots of money, so they should be expected to pay more for education because they'll make it back, but other fields should not have to pay as much". This is totally ass backwards. If returns are higher in engineering than art history, then we should subsidize engineering MORE and art history LESS.

Forget about individualized returns. Investment in higher education is all about maximizing society-wide returns (including soft stuff like an informed citizenry which can spot bullshit a mile away, which is not easily captured in the economic analysis).

I have no problem with making students pay the full costs, in fact I would support that, but the states already spend enough to eliminate tuition. Quality of the education would be effected but the differences would be very small or they could do it by reducing the number of students in the state systems.

From here: http://www.gao.gov/assets/670/667557.pdf it looks like tuition is only about 20% of spending. You could easily give the students 95%+ of the utility for 80% of the spending.

5. There are strange lines in Ulam's spiral (spiral made out of Erato's sieve) and there is this mind-blowing paper:
Matiyasevich, Yuri Vladimirovich. "Primes are nonnegative values of a polynomial in 10 variables." Zapiski Nauchnykh Seminarov POMI 68 (1977): 62-82."
by https://en.wikipedia.org/wiki/Yuri_Matiyasevich
(the paper concludes with "Let's point out that the degree of this polynomial is 15905", if you wonder "why nobody uses it" =) )

My bad, I didn't check up proper version of the paper (it's still impossible to comprehend, but as it is in English, it can be followed what exactly the result is), here it is:

Interesting pair of comments , Cassian... it was polite of you not to say "prime numbers are not less random than I (italicized I) think, Tyler". Julia Robinson (Adlai Stevenson's first cousin by marriage) did some work in this area, apparently. While I have known for years that primes are not random in the accessible reaches with which we are familiar - if they were, we would have achieved high quality random number generators a lot quicker - I had not pondered, until now, the question of "how much of a percentage of the mathematics that is known can be related to the question of whether primes are random" and the conscientiously more interesting question "how much of a percentage of the mathematics that can be known is related to the question of whether primes are random"? On the other hand, I still have problems with "long division", so for me this is all just stardust.

#2) Given that people can already qualify for free tuition by enlisting in the military, why is "free college" now such an imperative for liberals? I understand why someone might *prefer* not to enlist, but this path does give most people a workable way to earn college tuition. Furthermore, obviously, many people do go to college --- 60-65% of people have at least some college education --- which means that they can afford it. What is the case for creating a *universal* entitlement for something that most of the population can already afford or have a way to earn? Why not focus effort on the much smaller fraction of the population that is both poor and ineligible for military service?

"why is “free college” now such an imperative for liberals?"

Votes. Votes. And votes.

The Democratic party has only one playbook: appeal to people on the bases of race, gender, age and promising all of the above free stuff.

18-25 year olds just don't vote as often as the Dems would like them to. So something is needed to energize them. "Free stuff!" always works as a marketing ploy.

Clearly votes are at stake, after all this is politics. But you're ignoring that there is a very legitimate debate about the economic and social benefits associated with broadened access to education.

Yes, "free stuff" can win votes. But often there are very good economic and social arguments in favour of "free stuff" (not to disregard the existence of counterarguments, however).

I'm not aware of any debate on the matter of whether or not the group of people who do not qualify for scholarships sufficient to offset any remaining cost to attending a public university, after financial aid, have any hope of their education producing positive externalities, are you?

Any amount of education has positive externalities regardless of who gets it. However, the size of these positive externalities is lower for some people than others (say, introverts who do not discuss/share their knowledge, people who are just not very smart, etc.), in which case the size of the public investment may not be fully recouped.

I think the vast majority of the public is stuck in either "government=bad" or "help=good" mindset to have much rational discussion of this stuff outside academia.

Clearly, there are some non-retarded individuals who have virtually zero prospects of repaying (whether explictly by taxes or more broadly via positive externalities) the full cost of subsidies to higher education. Whether these are few or many in number is debatable, but most of the debate is ideological chest beating and foot stomping, it seems. I've translated perhaps a couple dozen papers which deal with related issues, and you might be surprised to know that even the cutting edge of "knowledge" on such questions is not really all that advanced.

There is much progress to be made on such questions.

hey, just saying, c n as my comments r definitely open to blog owner's discretion; but some of the slag i m reading, makes me wonder . . .

Pro Go players, like chess players, make much of their living teaching, tutoring, analysis, playing exhibition games and games for hire against strong amateurs. It is an empirical question as to whether the strong chess programs have cut into this income stream, and someone more into chess than I could answer that. I quit about the time Kasparov lost and bandwidth increased, bringing computers into online games.

The effort required to become a master is unlikely to pay off monetarily anymore. I'd quit if I were a kid.

"(Does Magnus Carlsen have a present?)"

There was a "60 Minutes" piece that showed him playing something like 10 games at once. So what's the big deal about that? Well, his *back* was to all the boards. They were just calling out their moves, and he would call out his. Pretty amazing.

From (5):
"If Alice tosses a coin until she sees a head followed by a tail, and Bob tosses a coin until he sees two heads in a row, then on average, Alice will require four tosses while Bob will require six tosses (try this at home!),"

Do NOT try this at home as an empirical exercise, unless you're going to code it up. You'll going to be doing a lot of coin flipping before you start to converge on these averages.

For one thing, I don't even know anybody named Bob.

2) I think it's important that students who demonstrate a strong commitment to academic learning or a clear vision of a career path are not restricted by lack of financial resources. But I think zero cost tuition will lead to too many students studying things out of mere interest, which doesn't bother me in principle, but in many cases this is not a clear economic benefit and not a clear benefit to good citizenship. I wouldn't want to get in the game of saying "this is worth studying, that's not worth studying", but when students cover at least SOME of the cost, they will be a little more careful.

For a poor student, even tuition of $2000 a year can be a very big deal for filtering out students who basically just don't know what else to do and so go to uni to study just any old thing that catches their fancy.

To improve access to higher education, I would find proposal to massively expand means-tested scholarships far more interesting and useful than "free tuition".

There's also another type of problem. With $0 tuition, some people would simply never leave uni! They'd hold part time BS jobs and soak in knowledge at free universities for the rest of their life. There may be some social benefit to having such knowledgable types, but there's a big difference between someone who amasses 5 BAs and someone who proceeds on to PhD studies, for example.

This is a strong comment.

As was pointed out above. We have about 2/3rd of high school grads already going to college. Of the remaining 1/3rd, a lot of them just don't want to go to college. There are at most 10-20% of high school grads that don't pursue college because of the cost. We don't need a massive spend on the other 80-90% just to account for these 10-20%. We can just subsidize the poor.

And we already do. The cost of community college for a student from a low-income family, after putting together state and federal grant money is very nearly zero. Subsidized loans are available to cover living expenses, etc. If that student can't manage to learn anything in school then they can move on to TANF, SNAP, EITC, Housing Assistance, etc. and never pay back their student loans.

#1 - How about the culture that will (or might) be New Hampshire?

"How many libertarians would it take to commandeer a state government?"

Comments for this post are closed