## Saturday, November 26, 2016

### Finimize Resources On Finance

I found this gem of a website called finimize that tries to teach basic financial knowledge to people who may not already be well informed on finance.

Here's their explanation of Quantitative Easing:
Quantitative Easing (“QE”) is when a central bank directly buys its government bonds. The idea is that it’s enacted when a central bank’s target interest rate has already been cut close to zero. By directly buying government bonds, it decreases the yield those bonds pay (because when the price of a bond increases, its yield declines – see below). Many types of loans are based, directly or indirectly, on the yield of government bonds and so by depressing those yields, the central bank makes it cheaper for companies and people to borrow money than a low target interest rate would do on its own – and it’s hoped that the borrowed money will used to do things that boost the economy (like buying machinery or hiring workers).
Now compare that with Wikipedia's explanation, which kind of starts to make sense after you've read the Finimize explanation.

## Friday, November 25, 2016

### Damning Statistics on Democracy

Jason Breannan, in his book Against Democracy, provides the following selection on how ignorant the electorate in the US are:
I could write an entire book just documenting how little voters know. But since many others have already done that, I’ll only offer a few examples:
•During election years, most citizens cannot identify any congressional candidates in their district.
•Citizens generally don’t know which party controls Congress.
•Immediately before the 2004 presidential election, almost 70 percent of US citizens were unaware that Congress had added a prescription drug benefit to Medicare, though this was a giant increase to the federal budget and the largest new entitlement program since President Lyndon Johnson began the War on Poverty.
•In the 2010 midterm presidential election, only 34 percent of voters knew that the Troubled Asset Relief Program was enacted under George W. Bush rather than Barack Obama. Only 39 percent knew that defense was the largest category of discretionary spending in the federal budget.
•Americans vastly overestimate how much money is spent on foreign aid, and so many of them mistakenly believe we can significantly reduce the budget deficit by cutting foreign aid.
•In 1964, only a minority of citizens knew that the Soviet Union was not a member of the North Atlantic Treaty Organization. (Yes, that’s right: NATO, the alliance created to oppose the Soviet Union.) Keep in mind this is just a short time after the Cuban Missile Crisis, in which the United States almost went to (nuclear) war with the USSR.
•Seventy-three percent of Americans do not understand what the Cold War was about.
•Most Americans do not know even roughly how much is spent on social security or how much of the federal budget it takes up.
•Forty percent of Americans do not know whom the United States fought in World War II.
•During the 2000 US presidential election, while slightly more than half of all Americans knew Al Gore was more liberal than Bush, they did not seem to understand what the word liberal means. Fifty-seven percent of them knew Gore favored a higher level of spending than Bush did, but significantly less than half knew that Gore was more supportive of abortion rights, more supportive of welfare state programs, favored a higher degree of aid to blacks, or was more supportive of environmental regulation. Only 37 percent knew that federal spending on the poor had increased or that crime had decreased in the 1990s.18 On these questions, Americans did worse than a coin flip. Similar results hold for other election years.
(Check the book for references.) If we assume that there are independent standards for evaluating elected and appointed officials (i.e. standards besides from the fact they were elected), the fact that large numbers of the electorate are so ignorant should give us reason to suspect that the electorate are incapable of voting into office public officials who can govern well.

That several Trump-like presidents in the past has not been elected should be considered a serendipity.

And by and large, I think the statistics about the US electorate are the norm.

## Sunday, November 20, 2016

### Baker on Healthcare Cost Drivers In The US

I had previously posted about the drivers of healthcare costs in the US. The following from Dean Baker's book "Rigged" is illuminating:
A big part of the difference in costs is that our doctors are paid twice as much as doctors in other wealthy countries. Average pay for doctors in the United States is over $250,000 a year, and in some highly paid areas of specialization the average is over$500,000. Paying our doctors the same as Germany, Canada, and other wealthy countries pay theirs would reduce our health care bill by close to $100 billion a year. Doctors are able to maintain such high salaries in large part because of measures that protect them from competition. We have limits on the number of people who go to medical school and on the number of foreign medical school graduates who can enter U.S. residency programs, the completion of which is a requirement for practicing medicine in the United States.[84] State laws also limit the extent to which nurse practitioners and other health care professionals can perform tasks, such as prescribing medicine, that might limit the demand for doctors. Baker also expands with the following example: The average pre-tax earnings for orthopedic surgeons in the United States was$442,500, compared to an average of $215,500 in the reference countries. ## Saturday, November 19, 2016 ### Family Conglomerates and Failed States Damodaran's latest piece makes me ponder the following: Is there a correlation between the number/prominence of wealthy family conglomerates in a nation and the failure of that nation state? Here's an informative snippet (do read the whole thing though): In many countries, including populous ones like India, influence is wielded and decisions are made by a surprisingly small group of people who know each other not just through their business networks but also through their social and family connections. These “people of influence” include bankers, rule makers and regulators that determine which businesses get capital, what rules get written (and who gets the exceptions) and the regulations that govern them. Family group companies have historically used these connections as a competitive advantage against upstart competition (both from within and without the country), especially in an environment where you have to pass through a legal, bureaucratic and political thicket to start and run a business. I would say exactly the same connection-based cronyism applies in Turkey, Russia, Ukraine and Greece. In fact, such cronyism is so pervasive that if you meet a person from the 3rd world who happens to be from a wealthy family, it may be reasonable to suspect that the person's family benefitted from corruption. Note that family companies were also prominent in developing Japan and Korea. See this Wikipedia entry and here. It would be good to compare those Japanese and Korean companies with those in India. ## Friday, November 18, 2016 ### Patent and Profit As Promoting "Wasteful" Scientific Efforts While espousing why he thinks patents are bad overall, Dean Baker makes the following point in Rigged [Emphasis mine]: For example, Merck and AbbVie, along with several smaller drug manufacturers, are rushing to market alternatives to Sovaldi as a treatment for hepatitis C. In the context in which Gilead Sciences, the maker of Sovaldi, has a monopoly on effective treatments for hepatitis C, this sort of competition is highly desirable because it will lead to lower prices. However, if Sovaldi were being sold in a free market at$500 to $1,000 for a course of treatment, there would be little incentive to invest research dollars finding treatments for a condition for which an effective drug already exists. If drugs were sold without protection, research dollars would usually be better devoted to developing a drug for a condition where no effective treatment exists than developing duplicative drugs for a condition that can be well-treated by an existing drug. He is right that competition in biotech for markets of established drugs is essentially wasteful. You could make a similar argument for any kind of industrial research lab as well. Any time someone is doing research in secret, there may be other researchers who are duplicating those secret efforts. Science advances best when it is cooperative. The profit motive introduces wasteful competition. This is why the mathematicians' polymath project was such a success in some of their collaborations. It was open and directly accessible - no peer review, etc. Baker's idea is also reminiscent of how the chess engine development took off once developers started open-sourcing and sharing their code. But I have to object when Baker quotes$500 to $1000 as the "free market" price of Sovaldi. He's clearly not accounting for the development costs. Baker favors a model where government contracts researchers directly and then subsequently claims any rights on their inventions. He says that such a model is already being used by the defense industry. Will you get as many biotech inventions once the research is essentially socialized? I would like to think so but I have my doubts. Also consider the following. Because of Apple's dominance in the smartphone market, Samsung decides to enter the smartphone market by . Based on this, should we conclude "oh well, the smart phone is really a solved problem; Samsung's dollars would really be better spent elsewhere"? Should we have a consistent reaction to competition in drugs and smart phones? I think not. I think that drugs are different from smartphone's in that they can effectively cure a disease as to make any further research superfluous. Smart phones can be improved in a lot of different ways. These improvements would be spurred by competition; whereas in drug development, it's hard to develop on your competitors' ideas due to patent protections. Nonetheless, Baker's point about drugs makes one wonder... What if you could get everyone to cooperate as opposed to compete? ## Thursday, November 17, 2016 ### The Financial Industry: The Administrative Cost of Running an Economy? I have been reading more of Dean Baker's book "Rigged". Here he makes an apt analogy that eloquently argues the dangers of an overblown financial sector. In this regard the financial sector is like the trucking industry. Trucking, like finance, is essential to the economy. We need it for moving raw material to factories and finished products to stores. But an efficient trucking industry is a small trucking industry: we want to have as few resources as possible devoted to getting goods from point A to point B. This means that we don’t want to see a huge expansion in employment in the trucking industry or an explosion in the number of trucks and warehouses just to move the same quantity of goods. The same story applies to the financial industry. We should want to see as few resources as possible committed to it, or the minimum needed to enable it to support the productive economy. Instead, we have seen a massive expansion of the financial sector, from 4.5 percent of GDP in 1970 to 7. 4 percent in 2015. 20 The more narrow securities and commodities trading sector increased from 0.49 percent of GDP in 1970 to 2.03 percent in 2015,21 corresponding to$290 billion a year in additional spending in the 2016 economy.
Of course, if the transportation needs of the good produced got more complex (Amazon and internet retail?), we should expect the trucks and warehouses to expand both in size and in real dollar terms.

So perhaps the corresponding question for the financial sector should be: have the financing needs of the economy changed to require a bigger financial sector?

The answer is likely to be no but my point is that it's not just sufficient to point to a growth in the financial sector as immediately proving an increase in waste.

## Wednesday, November 16, 2016

### Healthcare and Employment in the US

I've been reading Dean Baker's free available book "Rigged: How Globalization and the Rules of the Modern Economy Were Structured to Make the Rich Richer".

In the US, it is common for employers to offer their employees healthcare as a benefit. But this disincentivizes employers from hiring more workers and causes them to squeeze as many hours as possible from each worker, since each worker has a fixed cost to the company. He writes:
Specifically, because health care insurance is mostly an employment-based benefit, rather than universally provided through the government, employers typically see it as a fixed cost per worker. So rather than hiring additional workers and paying for their health insurance, employers would generally prefer their existing workforce to work more hours. The preference for longer hours is even greater in cases where employers provide defined benefit pensions in addition to health insurance.
Of course, the pattern of providing health insurance and pensions as employer-provided benefits was the result of government tax and regulatory policy that favored employer-based benefits. The result was a pattern of employment in which better-paying jobs are more restricted in number than would have otherwise been the case. Consider this basic arithmetic: if we produced the same number of cars but the average work year in the auto industry had 20 percent fewer hours, we would have 25 percent more people working in the auto industry. The reality will always be more complicated, but the basic point is straightforward: for the same levels of output, shorter hours mean more workers.
It is a provocative argument, to say the least.

## Tuesday, November 15, 2016

### Frankfurt On Ineqaulity

The previous post about global and national inequalities reminded me of this old chestnut from Harry Frankfurt. Frankfurt argues that it is not morally desirable that everyone has equal resources. It's morally desirable that each has enough. In his words:
Mere differences in the amounts of money people have are not in themselves distressing. We tend to be quite unmoved, after all, by inequalities between those who are merely well-to-do and those who are extremely rich.
If we believe of some person that her life is fulfilling, that she is content with her economic situation and that she is not troubled by any resentments or sorrows that more money could assuage, we are not ordinarily much interested — at least, from a moral point of view — in a comparison of the amount of money she has with the amounts possessed by others. The fact that some people have much less than others is not at all morally disturbing when it is clear that the worse off have plenty.
In other words, if everyone had a Mercedes and one guy had a Ferrari, it would not make much sense to describe this as a terrible state of affairs.

I'm not too sure about this argument. For instance, based on an argument of this sort, could one find it acceptable that the economic growth in a country mostly benefits the top %1 on the basis that the bottom %99 percent "have enough"?

Of course, I'm also not sure that Frankfurt's argument has this implication or what Frankfurt thinks about economic inequalities in the world in a more practical sense.

Remember from the previous post that the Western Middle Class actually have more than the Asian middle class. So, can we say that the Western Middle Class should not be bothered about the rising inequality because they have "enough"?

Note also that, perhaps sadly or irrationally, people's happiness depend more on their relative wealth than their absolute wealth.

There is something fishy going on; I do think that Frankfurt must be mistaken somewhere but I'm not too sure I can flesh it out.

For Frankfurt's original paper, see here.

## Monday, November 14, 2016

### Reducing Global Inequality At The Cost of Raising National Inequality?

Branko Milanovic has a provocative article hinting at the possibility that the reduction  in global inequality may come at the cost of raising national inequality. Here's the key chart:

While the Asian middle class and the top %1 have seen an increase in real income, the US and Western lower middle class have not seen as much gains.

Brankovic's larger point is that people care more about national inequality than global inequality. Thus they are unlikely to accept a reduction in global inequality if it comes at the cost of increasing national inequality.

I'm reminded of some studies that show it is not the absolute wealth people care about but the relative wealth. (I can't find any references at the moment.)

## Saturday, November 12, 2016

### Don Trump

What's happening is so bizarre on so many levels that Leiter, who had sworn off political blogging many years ago, is back at it.
And like any Mafia don, Trump, of course, prefers family: his "transition team" includes his three adult children and his son-in-law (a fifth of the team, and probably the most important fifth). I suspect one or more of them will end up with positions in his Cabinet or staff.
See the full post here.

## Tuesday, November 8, 2016

### Brennan's Archetypes of Political Actors

In "Against Democracy", Jason Brennan identifies 3 archetypes of people in relation to how they approach politics. These are hobbits, hooligans and vulcans.

"Hobbits" are uncaring about and indifferent to politics. If they do, they rarely vote.

"Hooligans" are so tightly perched to their views that political views have become an identity for them, akin to religions for a religious person.

And finally "vulcans" are dispassionate, informed and able to change their views. They are the only group who can articulate their opponents' views without being charged as being unfair to the view being articulated.

The classification is interesting because it doesn't identify people as conservative or liberal. It's supposed to be view neutral.

•Hobbits are mostly apathetic and ignorant about politics. They lack strong, fixed opinions about most political issues. Often they have no opinions at all. They have little, if any, social scientific knowledge; they are ignorant not just of current events but also of the social scientific theories and data needed to evaluate as well as understand these events. Hobbits have only a cursory knowledge of relevant world or national history. They prefer to go on with their daily lives without giving politics much thought. In the United States, the typical nonvoter is a hobbit.
•Hooligans are the rabid sports fans of politics. They have strong and largely fixed worldviews. They can present arguments for their beliefs, but they cannot explain alternative points of view in a way that people with other views would find satisfactory. Hooligans consume political information, although in a biased way. They tend to seek out information that confirms their preexisting political opinions, but ignore, evade, and reject out of hand evidence that contradicts or disconfirms their preexisting opinions. They may have some trust in the social sciences, but cherry-pick data and tend only to learn about research that supports their own views. They are overconfident in themselves and what they know. Their political opinions form part of their identity, and they are proud to be a member of their political team. For them, belonging to the Democrats or Republicans, Labor or Tories, or Social Democrats or Christian Democrats matters to their self-image in the same way being a Christian or Muslim matters to religious people’s self-image. They tend to despise people who disagree with them, holding that people with alternative worldviews are stupid, evil, selfish, or at best, deeply misguided. Most regular voters, active political participants, activists, registered party members, and politicians are hooligans.
•Vulcans think scientifically and rationally about politics. Their opinions are strongly grounded in social science and philosophy. They are self-aware, and only as confident as the evidence allows. Vulcans can explain contrary points of view in a way that people holding those views would find satisfactory. They are interested in politics, but at the same time, dispassionate, in part because they actively try to avoid being biased and irrational. They do not think everyone who disagrees with them is stupid, evil, or selfish.
I'm also fascinated by the cynicism that entwines Brennan's thinking.

Is the problem with democracy that we don't have enough vulcans, especially in poorer nations?

### Brennan Against Democracy

From the New Yorker article I linked earlier, this passage does a good job of challenging the universal regard shown for elections:
In a new book, “Against Democracy” (Princeton), Jason Brennan, a political philosopher at Georgetown, has turned Estlund’s hedging inside out to create an uninhibited argument for epistocracy. Against Estlund’s claim that universal suffrage is the default, Brennan argues that it’s entirely justifiable to limit the political power that the irrational, the ignorant, and the incompetent have over others. To counter Estlund’s concern for fairness, Brennan asserts that the public’s welfare is more important than anyone’s hurt feelings; after all, he writes, few would consider it unfair to disqualify jurors who are morally or cognitively incompetent. As for Estlund’s worry about demographic bias, Brennan waves it off. Empirical research shows that people rarely vote for their narrow self-interest; seniors favor Social Security no more strongly than the young do. Brennan suggests that since voters in an epistocracy would be more enlightened about crime and policing, “excluding the bottom 80 percent of white voters from voting might be just what poor blacks need.”
I have immediately obtained the book "Against Democracy" - I intend to read and blog about it.

However, here's an immediate thought regarding the quote. Brennan claims that "[e]mpirical research shows that people rarely vote for their narrow self-interest". Would that still be the case if you gave voting rights to a competent minority?

I am reminded of the self-serving votes about MPs' salaries in various parliaments all over the world.

I applaud with enthusiasm to come up with alternatives to democracy but I'm skeptical as to any system of governance can succeed with an uninformed electorate. As Kant said, "out of the crooked timber of humanity, no straight thing was ever made."

## Monday, November 7, 2016

### Genetic Lottery For Body Types And The Prior Probability For "Banging a Model"

A recent discussion I had reminded me of the following passage from David Epstein's excellent The Sports Gene:
Big Bang data in hand, Norton and Olds devised a measure they called the bivariate overlap zone (BOZ). It gives the probability that a person randomly selected from the general public has a physique that could possibly fit into a given sport at the elite level. Not surprisingly, as winner-take-all markets have driven the Big Bang of body types, the genes required for any given athletic niche have become more rare, and the BOZ for most sports has decreased profoundly. About 28 percent of men now have the height and weight combination that could fit in with professional soccer players; 23 percent with elite sprinters; 15 percent with professional hockey players; and 9.5 percent with Rugby Union forwards.
...Particular professions that require unique physiques have an even more concentrated winner-take-all structure and outdo even professional sports. The BOZ for regional catwalk models is less than 8 percent, dropping to 5 percent for international models, and to just 0.5 percent for supermodels.
As always, nowadays when one of your friends brags about "banging a model", you have a prior with which to evaluate that claim, namely at most %8. (I'm happy to help whenever and wherever you need me.)

### Economic Growth of Turkey under AKP

Erik Meyersson has a new paper (draft) about the economic development of Turkey under AKP. Here's a relevant quote:
I find that Turkey under AKP grew no faster in terms of GDP per capita when compared with a synthetic counterpart from a wide pool of other countries. Restricting the pool of control units to only include Muslim countries shows Turkey growing slower than its Muslim counterpart. Moreover, analysis of post-crisis recovery periods shows Turkey growing no faster than comparable post-crisis cases across time. However, expanding the outcome set to health and education reveals large positive differences in both infant and maternal mortality as well as university enrollment, consistent with stated AKP policies to improve access to health and education sectors for the relatively poorer segments of the population.
I have yet to finish all of it but the paper seems to confirm one of my judgements about the elections with an uninformed electorate. Namely, the electorate really aren't qualified to discern good governance from bad governance.

This is primarily abused in the following way. While feeding proverbial peanuts to "the poorer segments of the population", the parties in power can massively usurp wealth and other resources and commit countless abuses such as wholesale crippling of any dissident media. The poor and uninformed meanwhile think they are having a good time through their diet of proverbial peanuts. Thus, while keeping the electorate poor and uneducated (thus dependent on peanut handouts), the parties in power are able to fend off any electoral challenges. A cesspool of democracy indeed.

Note that Dani Rodrik wrote along similar lines about the Turkish economic myths before.

## Sunday, November 6, 2016

### Doctors Per 1 000 population

In my previous post on the healthcare composition in the US, I had cited a CDC report as claiming that "[p]hysician and clinical services accounted for 23.5% of total personal health care expenditures".

So in order to understand what drives these physican costs, I asked myself the following: Is there a shortage of doctors in the US? Could that be why physician costs are so high?

So I looked up the doctors per 1 000 population datum over at this OECD site.

This is 2009 data, so I don't know how things may have changed in the last 7 years or so.

I don't know how to interpret this data. In one sense, US does appear towards the bottom of the list. On the other hand, countries like Japan that have even less doctors (per 1000 population) have less healthcare costs.

## Saturday, November 5, 2016

### Short Review of Jordan Ellenberg's "How Not To Be Wrong"

I've finally finished reading "How not to Be Wrong".

Here is Bill Gates' review of the book.

Here are some of my thoughts on the book.

Let me begin by saying the book offers a lot of interesting tidbits. Here's one where Ellenberg talks about how mathematicians try to prove a statement by day and look for counterexamples by night. The section where he talks about it is so good that I feel justified in quoting at length:
In fact, it’s a common piece of folk advice—I know I heard it from my Ph.D. advisor, and presumably he from his, etc.—that when you’re working hard on a theorem you should try to prove it by day and disprove it by night. (The precise frequency of the toggle isn’t critical; it’s said of the topologist R. H. Bing that his habit was to split each month between two weeks trying to prove the Poincaré Conjecture and two weeks trying to find a counterexample.)
Why work at such cross-purposes? There are two good reasons. The first is that you might, after all, be wrong; if the statement you think is true is really false, all your effort to prove it is doomed to be useless. Disproving by night is a kind of hedge against that gigantic waste.
But there’s a deeper reason. If something is true and you try to disprove it, you will fail. We are trained to think of failure as bad, but it’s not all bad. You can learn from failure. You try to disprove the statement one way, and you hit a wall. You try another way, and you hit another wall. Each night you try, each night you fail, each night a new wall, and if you are lucky, those walls start to come together into a structure, and that structure is the structure of the proof of the theorem. For if you have really understood what’s keeping you from disproving the theorem, you very likely understand, in a way inaccessible to you before, why the theorem is true. This is what happened to Bolyai, who bucked his father’s well-meaning advice and tried, like so many before him, to prove that the parallel postulate followed from Euclid’s other axioms. Like all the others, he failed. But unlike the others, he was able to understand the shape of his failure. What was blocking all his attempts to prove that there was no geometry without the parallel postulate was the existence of just such a geometry! And with each failed attempt he learned more about the features of the thing he didn’t think existed, getting to know it more and more intimately, until the moment when he realized it was really there.
This passage speaks to me for the following reason. Lately, I've been learning about investing in individual stocks (don't worry though, most of my monies is nicely stashed in index funds). After being in the game for 4 months or so, I see that most stock analysts (and investors) are overly optimistic. (There are some structural reasons for this.) Thus it might be a great idea to defend the bullish case for a particular stock (i.e. defending that the underlying business will do well and price will rise) by day and do the opposite, defend the bearish case by night.

In fact, as does Ellenberg, I invite the reader to take one of his/her cherished beliefs and play the devil's advocate by defending the said belief's negation.

Besides shining tidbits such as the quoted passage (see also my earlier post on p-values; there are frankly many), I think that the book doesn't do a good job of what it's purporting to do, that of explaining non-trivial but easy-to-understand mathematics to an intelligent but not necessarily numerate reader. I had prior exposure to most of the mathematical content in the book, so I could just ho-hum my way through bits where the author gets quasi-technical. Had I not known the subjects in greater detail, I would have been worried about receiving such summary treatment - and the suspected dilution of cognitive content. For instance, in one place, the author talks about projective geometry. I am only a little familiar with projective geometry, so I didn't really understand the points of the author. I didn't worry about missing author's points because if I really wanted to understand projective geometry, I would have to start with a more technical text.

I would recommend this book to readers who already know quite a bit of mathematics. For them, it might be an entertaining review of many mathematical ideas' distilled essence. For readers who would like to become more sophisticated mathematically, I would recommend first-order mathematics books instead.

## Thursday, November 3, 2016

### Democratic Cesspool

Trump in the US (as of now, unelected but latest polls indicating his resurgence), Putin in Russia, Duterte in the Philippines... There are lots that make one doubt democracy.

One other expression of the popular will that probably wasn't in the best interests of the popular will was the UK referendum on EU membership.

Now the high court in London has ruled that parliament must approve the Brexit before it can be officially triggered, adding further acrobatics to the democratic circus.

An Oxford philosopher, Les Green, has interesting thoughts here.

I've been so frustrated with elections in the past 5 years or so. I want to study more in depth what sort of awful results election-based power allocation schemes find themselves in. This New Yorker essay is a good place to start.

## Wednesday, November 2, 2016

### On the discovery of Sovaldi, Gilead's miracle drug for Hepatitis C

Continuing with our theme of healthcare costs, I thought this Sci-Am article was interesting. The author does a good job of refuting some of the clumsy claims regarding drug prices, e.g. focusing on the manufacturing costs and dismissing discovery/research costs. The author himself is a drug scientist. This quote is extremely interesting:
Drug discovery is one of the most wasteful research activities on the planet and it's all because most of the times we have no idea of how to go about discovering a new drug or what's going to happen when we put it inside the human body. The complexities of human biology thwart us at every stage and luck plays an inordinately large role in our success. Even basic issues in drug discovery - understanding how drugs get past cell membranes for instance - are generally unsolved problems, and the profligate inefficiency of the process would truly shame us if we knew how to do it better. The path from a new idea in pharmaceutical research to an actual drug is akin to a path trodden by a blind man along the edge of a cliff at night; any survival, let alone success, seems like a miracle in retrospect. No drug scientist will admit it, but every drug scientist crosses his or her fingers when a new drug makes it to the market because we are just not smart enough to figure out exactly what it will do in every single patient. That is why most drugs fail in advanced clinical trials, when hundreds of millions of dollars have already been spent on them.
This is kind of reminiscent of what my high school physics teacher used to tell us. (Apparently, he was a research chemist working on pesticides before he took up high-school science. He told us they'd randomly test chemicals on insects and pick the combo that yielded the least bug survival rate.)

But note that the author doesn't discuss the following.

1. US consumers pay significantly more for the same drugs than their counterparts in the rest of the developed world do. As I mentioned before, this probably stems from the absence of an "all payer rate setting" in the US drug market.
2. Gilead acquired a company called Pharmasset that had begun developing Sovaldi. The research that inspired the creation of this company was originally federally funded.
3. The US taxpayers also fund the FDA, whose effective functioning is critical in bringing these drugs safely to the market.
So all in all, the US taxpayers seem to be funding a lot of useful things like university research and FDA, only to be punished later with higher drug costs.

Nonetheless, I'd rather have big pharma get filthy rich than, say, Wall Street, e.g. Wells Fargo executives. They do cure the sick after all.

## Tuesday, November 1, 2016

### Joseph Berkson on p-values

I'm about to finish Jordan Ellenberg's "How Not To Be Wrong: The Power of Mathematical Thinking". Perhaps, I'll write a slightly longer review later but for now, let me point out that my favorite section of the book has been its discussion of p-values. To demonstrate the problems of over-reliance on p-values, Ellenberg recounts the following:
But here’s the bad news: the reductio ad unlikely, unlike its Aristotelian ancestor, is not logically sound in general. It leads us into its own absurdities. Joseph Berkson, the longtime head of the medical statistics division at the Mayo Clinic, who cultivated (and loudly broadcast) a vigorous skepticism about methodology he thought shaky, offered a famous example demonstrating the pitfalls of the method. Suppose you have a group of fifty experimental subjects, who you hypothesize (H) are human beings. You observe (O) that one of them is an albino. Now, albinism is extremely rare, affecting no more than one in twenty thousand people. So given that H is correct, the chance you’d find an albino among your fifty subjects is quite small, less than 1 in 400,* or 0.0025. So the p-value, the probability of observing O given H, is much lower than .05.
We are inexorably led to conclude, with a high degree of statistical confidence, that H is incorrect: the subjects in the sample are not human beings.