Podcasts of February 2017

In February I listened to:

Jim Epstein on Bitcoin, the Blockchain, and Freedom in Latin America
Probably one of my favorite econtalk podcasts so far. Talks about how bitcoin is popular in Venezuela because of lack of access to food (strict government controls), so instead of waiting in line for hours for essentials Venezuelans have turned to bitcoin mining to get revenue. Mining bitcoin generally is cost prohibitive, since the electricity to do is very expensive, but electricity in Venezuela is largely subsidized by the government so it is one of the countries where it economically makes sense to mine bitcoin. The bitcoins can be transferred to USD or coupons at a lot of popular exchanges. This kind of activity is also seen where there are big tariffs (e.g. Mexico with a ~60% tariff on buying iPhones).

Also interesting is the idea of blockchains, the tech behind bitcoin. Essentially a shared database where people can access records. Has uses in housing (keeping records of buys & sales & ownership). Really interesting thing to think about even in the US where government plays a big role in ensuring sales are legitimate (e.g. can't accidentally sell the driveway I share with my neighbor). Highly recommend this podcast, I enjoyed & understood almost all of it (the former comes often, the latter comes rarely).

George Borjas on Immigration and We Wanted Workers
Also a really interesting one. Some main ideas include the comparison between immigration and trade. If we think of immigration as low skill workers threatening the jobs of some then trade will be largely the same, except immigrants are people & not widgets (paraphrased from Borjas) so there are some complicated differences. Brings up the great point that in immigration we often don't discuss the makeup of people, and what types of immigrants are coming into America and what types of jobs / workers they threaten.

Michael Munger on the Basic Income Guarantee
I would love for someone to listen to this and explain it to me better. It was a really interesting intense debate about BIG (Basic Income Guarantee), essentially a basic income that is guaranteed to everyone (say, 15k for everyone a year). There are lots of arguments presented, but one of them (at least as far as I understood) was that we are already providing some kind of BIG via lots of government programs, and instead we should scrap them (save the overhead, save the trouble) and consolidate them into a BIG. Stuff like "cliff effect" where people will avoid working to lose basic guarantees are considered, and Russ suggests something like a progressive BIG. 

Dr. Munger makes a really interesting claim at the end- that the BIG will not go up, but will rather go down as we become more efficient. As our technology improves things will fall in price, and while wages will fall, prices will fall even harder & faster- so the BIG will go down not up. Similar argument to what Elon Musk made (about automation making BIG a necessity).

Michael O'Hare on Art Museums
Er, I've said so far I liked all of them. I guess it's because the ones I didn't like (Grit with Angela Ducksworth, Ego is the Enemy) I just didn't finish. But this one was also good, and it's a subject personally interesting to me. Michael O'Hare talks about how art museums currently don't well serve their audience, and they generally don't act towards what he thinks their goal should be: "The purpose of an art museum is more, better engagement with art." To this end, he discusses admission prices, museum layout (including getting more chairs), and the surprising amount of assets museums have.

Museums, big museums like the ones we have in NY, often only display single digit percentages of their assets (he cites 4~5%), most of which sits unviewed in a basement. Michael argues that we can take these art pieces and sell them, either to other museums or to private collectors. This would a) easily, easily take care of admission prices and b) allow new museums to come into existence. By allowing these assets to flow more freely the museum can generate revenue to better serve the public, but this is currently not allowed because amongst directors there is a rule to never sell art (except to acquire more art). I'm not sure how factual all these numbers are but I found the idea very interesting.

This was also the first econtalk podcast I listened to!

Cathy O'Neil on Weapons of Math Destruction

The subject of this podcast is actually an answer I gave in a Palantir interview about problems in tech. I gave it really poorly and I didn't articulate it well, but Cathy brings up much better examples in a much more coherent way. She discusses Weapons of Math Destruction, algorithms that are secret, have big impact, and are destructive. Examples include recidivism rates, teacher ratings, college rankings, and advertising, and how these algorithms are destructive in very non transparent ways.

More than Terminator-esque sentient AI, I think this is the bigger concern with AI and ML and all these new algorithms. The developer isn't perfect and we should be careful of oracles.

Jonathan Skinner on Health Care Costs, Technology, and Rising Mortality
Health care is one of my favorite subjects so far, mostly because I have so little knowledge about it. Jonathan Skinner addresses how innovation is not always good in health care, and how there are many inefficiencies produced even with better technology. For example, a technology may be found to be effective for a small controlled group, but then is spread and used across the country in hospitals with much less certain effectiveness and much greater cost. An alternative is mentioned in British health care system where certain services are not provided by the government due to cost, but that sometimes results in lives lost and sicker people. Primary takeaway for me is that US health care is definitely innovative, but perhaps the innovation is not necessarily worth its cost.

In the last few minutes, he brings up a really interesting idea: a central institution that processes all insurance claims, reducing administrative costs, and providing central data for regional variations and where spending is going (and how effective it will be).

Richard Epstein on Cruises, First Class Travel, and Inequality
This was a confusing one. I didn't follow all the arguments or agree with all of them, but I wasn't sure if the latter was mostly due to the former. The podcast opens with a discussion of consumption inequality (e.g. first class seats) and how that benefits us all (the extra money is used in a way that helps us all, e.g. bigger planes, tvs, ambient lighting, all the new 777 stuff). They then move onto a general discussion about inequality, and this is where I kind of disagree. Russ & Richard make the point that a lot of billionaires are also philanthropists, and even though many people dislike the idea of billionaires having so much control, the money and aid they contribute is better than the government stepping in. Basically voluntary redistribution is much better than (i.e. more effective, efficient) a forced government transfer program.

They also discuss tax and regulation, and Richard makes the argument that they both produce incentive effects, such as curbing innovation. We are much better off by punishing those who cross the line than by intensely regulating them and affecting their growth.

I guess my problem with that argument is I mistrust some of these people in power and (maybe foolishly) think the government would better serve the people's needs.

Doug Lemov on Reading
This one was quite good, if not that insightful (for me personally). Doug discusses the importance of reading, and the importance of close reading. It's important for our education to emphasize reading difficult texts closely, and have the attention & grit & capability to get through difficult but rewarding texts. In particular, reading any book is not the same as reading difficult books, and one way we can incentivize reading is by teaching it in such a way that we combine difficulty with rewards.

I think one of the more valuable skills I got out of 4 years here at Columbia is how to skim texts, how to closely read texts, and how to read texts to distill the core arguments. 

Doug also makes a great case for reading books out loud. Made me think about how wealthy families tend to be more able to read out loud and provide resources for their kids.

John Cochrane on Economic Growth and Changing the Policy Debate
John was very interesting to listen to and had a good conversation with Russ. The takeaways I had from this podcast were: America is stagnating in economic growth, and that measure is one of the most important. Other issues we debate are often subsidiaries of that one problem. Innovation & growth is stagnated by taxes & regulation, and by removing these barriers we will have a better standard of living. 

There's currently a ton of inefficiency in how we operate (complicated tax codes, complicated loopholes, complicated laws) which in large part is due to a regulatory nightmare. This gets into an interesting discussion about policy debate, and how we can change the conversation to actually reform the system. We need, essentially, a leader/party/politician that can form this coalition and put together a "grand bargain."

I also found the point on corporate taxes interesting. Corporate taxes, according to John, never punish the company and instead punish the consumer through higher costs and thus higher prices. It would be better to remove corporate taxes and allow more freedom to invest. This also has the added benefit of removing the incentive companies have to hire smart people to find tax loopholes.

Josh Luber on Sneakers, Sneakerheads, and the Second-hand Market
This one was half about the economics of sneakers and half about the data science. I think both were interesting, but I preferred the data science part of it. I've never been a sneakerhead but I've had friends who were really into sneakers. The economics were pretty clear; low supply (rarity) and high resale prices drive people to wait in line and generates a huge resale market (like concert tickets, some clothing brands, Supreme metro cards etc.). The data analysis part was really interesting. Cleaning up data from ebay auctions & pricing the shoes correctly is a really really difficult problem, and it was cool to listen to the various problems. I especially found interesting the problem of misjudging prices anecdotally. Since undervalued shoes are quickly bought on ebay, if you spend a lot of time looking at the data, you would tend to have a higher price point for the shoe, just because of sampling bias. 

I particularly liked the comparison between shoes and drugs, e.g. central supply chain, various distributors, lots of resellers...

Chris Blattman on Sweatshops
Not really sure what I learned from this one, although I still thought it was interesting. Chris explained his experiment in Ethiopia where he split a couple hundred people into 3 groups, factory, entrepreneur, and control, and found that after a few years, a high number of the factory group changed jobs or left factory jobs, and entrepreneurs tended to have pretty decent return on the initial capital and be happier. This suggests that factory jobs and industrialization may not be the only or optimal solution for poverty.

One particular point I found interesting was about lacking middle management. Middle level management was often most lacking in these sweatshop, people who can organize and run the factory themselves. Chris points to this as an attractive part of Ethiopia/ Kenya, with a fairly developed middle class and a stable economy. 

If nothing else, I appreciate the lack of a strong call to action. The experiment was localized and certainly had interesting results but may not apply to all of Ethiopia, let alone all of the countries where there are still sweatshops.

Abby Smith Rums on Remembering, Forgetting, and When We Are No More
Not that impressed with this one... agree that we should preserve the past and agree that it's important to engage stuff in its context, not really sure what I learned. To be fair though I also skipped a big chunk of it.

Jason Zweig on Finance and the Devil's Financial Dictionary
This one was good, and then ok, and then REALLY good. There were four things in particular I found interesting:

1) Jason describes how in 30 years of financial advising, most of what he does is just repeating himself several times a year, since good financial advice doesn't change very often. Instead, the newer stuff that comes out often turns out to be bad for the investor. 
2) The instantaneous access to information anytime we want is a very bad thing, and often prompts us to make hasty decisions out of fear or greed. I quote, "The greatest asset an investor can own is self-control."
3) Teaching finance to kids through competition is a horrible idea, because it teaches a) gambling b) trading on margin c) financial markets as a contest. Instead we should focus on teaching basic arithmetic, humility, and self-control (admittedly hard).
4) Sort of unrelated, but "No individual can assist or save the age. He can only express that it is lost." - Kierkegaard, and his father's response: "Well, he's right. But that's why you have to try to assist or save the age." I thought that was lovely. 

Tom Wainwright on Narconomics
This podcast was about narconomics, i.e. the economics behind the drug industry. Tom explains how typical efforts in the War on Drugs involves destroying the supply of the drug through ways such as spraying pesticide from planes on poppy fields or burning supply that they find. This is an ineffective way to combat the use of drugs, because, and I found this super interesting, the cost of the raw material is not even close to the cost of the final product. To produce 1kg of cocaine requires 1 ton of poppy seed, which costs 500 dollars, but is marked up progressively until it reaches the United States, rising up to 20,000 and finally to 100,000+ before entering the hands of the actual consumer. Tom draws the analogy between trying to increase prices by hurting supply and trying to increase prices of art by increasing the cost of paint. The marginal increase in material cost is so much smaller that the effect will be miniscule.

Russ and Tom then discuss how hurting demand is a better strategy, and here is where I started to find the podcast less interesting. I didn't really hear any concrete ways to hurt demand, although they mentioned stuff like helping addicted people, educating younger people to not do hard drugs, etc. One thing I found interesting was Tom's discussion on marijuana and its legalization. By legalizing marijuana, we make it safer, since it becomes a legitimate business that doesn't have to be largely done in cash, and more importantly, we can rely on the FDA for federal regulation on marijuana instead of state by state. I am curious if his perspective on all drugs is to legalize, allowing for better regulation and completely killing the market of the drug cartels. Kind of interesting stuff. I find my favorite podcasts are the economics meets unexpected field, such as the sneakerheads one.

Michael Munger on Slavery and Racism
Big fan of Mike Munger... I liked the BIG one as well, will probably watch some more of his. This one was about a paper he co-authored about the institutions of slavery. Munger defines racism in his paper as combination of bigotry and an institutionally privileged position (so it is difficult for minorities to be racist, although everyone can be a bigot). Munger argues that in order to justify slavery, slave owners of the South had to create an ideology to persuade themselves that slavery was good, leading to the creation of these institutes of racism to match these moral beliefs and logical conclusions.

They had economic and social incentives to do so, since there were many more slaves and a greater economic need for slaves. Munger discusses this from a Humean perspective, arguing that reasons are the slaves of passion, and passion ultimately comes from self interest. Because of this self interest & these incentives, slave owners self deceived themselves into thinking that slavery was good through racism. It is not a concerted planned central creation; it is more an emergent phenomenon from this shared set of ideal ideals. 

I found this point really interesting- Munger applies the veil of ignorance to racism. Suppose we all go into a room, and when we come out, some are going to be slaves, some are not, independent of who we are when we go into the room. If you still believe in slavery, then that could be your impartial spectator standard, but if you don't, then there's not much of a moral argument there.

Robert Frank on Dinner Table Economics
This one is slightly different from most podcasts in that they discuss a variety of different cases of "dinner table economics." According to studies, a lot of students after taking their economics courses cannot answer/ apply basic economic principles to questions any better than students without any experience. To that end, Russ and Robert talk about some real world problems with interesting economic solutions.

Q1) Why old wine is more expensive than new wine?
A) Old wine is more expensive because of 1) demand shrinks as more and more people drink the limited old wine and 2) there is an additional cost to old wine (which is the return on investment if it was sold immediately and that money was put into a bank account). Similarly, and I thought this was more interesting, what happens to the price of pork when the price of corn goes up (assuming pigs eat corn)? The price of pork will go up in the long run to accommodate for the greater cost, but the price of pork will go down in the short run as pigs become more expensive to keep, so more pigs will be slaughtered and brought to market.

Q2) Why do men rent tuxedos, and women buy wedding dresses?
A) If we assume (and I think it's fair to do so) that women generally want unique wedding dresses, and the sizing of wedding dresses varies much more greatly than sizing of tuxedos, then there is a much greater cost to maintaining and keeping enough dresses for rent (something Robert calls money on the table assumption). As a result, renting dresses might end up being more expensive than just buying them.

Q3) Why is it more expensive to rent bikes than cars?
A) Similar to before, because of scale, cars are much easier to rent than bikes, and require less maintenance, and they are often cheaper to buy (because they have a special deal with the car company & they purchase in bulk). On the other hand, bikes require some custom tinkering, and the size just isn't there to support more efficient ways of renting bikes.

Pretty interesting, but nothing too difficult or profound (although those are maybe just my thoughts after listening to the answers).

Robert Aronowitz on Risky Medicine
This podcast was about how the healthcare industry is increasingly focused on risk reduction instead of treatment, and highlights the high opportunity cost. I found this one specifically enlightening especially since I was born and raised in Taiwan, and in many Asian countries the norm is to take medicine, see a doctor, or get a checkup as a default response.

Because of over diagnosis and risk centered medicine, a lot of money is wasted on unnecessary checkups and false positives. These results and checkups are often invasive procedures that are physically taxing on the patient and also cause a lot of psychological stress and anxiety. One example Robert mentions is how mammograms are now a routine part of a woman's annual checkup when data suggests that these mammograms often do more harm than good (although, of course, data is complicated and not always black and white). Another closely related example is PSA tests for prostate cancer. A high PSA level often means recurring 6 month PSA exams to monitor changes, and Robert finds that many men get a prostatectomy to not only reduce the risk of cancer but also to get rid of the anxiety that comes with it. I am reminded of the Scrubs episode where Dr. Kelso introduces full body screenings at Sacred Heart, and Dr. Cox tries to persuade patients to not get one to prevent the test results from driving them crazy. 

I think the key to his argument is that a lot of data from clinical trials is incomplete or inconclusive, and it is dangerous to rely on partial data since we still know very little. A procedure with a slim and dubious statistically better chance at detecting an illness accompanied with a much greater opportunity cost just doesn't really make much sense.

I am a little torn though, because even if it's irrational, I doubt I will ever tell my parents to not get regular checkups...