Artificial General Intelligence and Existential Risk

The purpose of this post is to discuss existential risk, and why artificial intelligence is a relatively important aspect of existential risk to consider. There are other essays about the dangers of artificial intelligence that I will link to throughout and at the end of this post. This essay is a different approach that perhaps will appeal to someone who has not seriously considered artificial general intelligence as an issue requiring civilization’s attention. At the very least, I’d like to signal that it should be more socially acceptable to discuss this problem.

First is the section on how I approached thinking about existential risk. My train of thought is a follow up to Efficient Advocacy. Also worth reading: Electoral Reform Fantasies.

Background

Political fights, especially culture war battles that President Trump seems so fond of, are loud, obnoxious, and tend to overshadow more impactful policy debates. For example, abortion debates are pretty common, highly discussed political issues, but there have been almost no major policy changes since the Supreme Court’s decision 40 years ago.  The number of abortions in the US has declined since the 1980s, but it seems uncorrelated with any political movements or electoral victories. If there aren’t meaningful differences from different political outcomes, and if political effort, labor, and capital is limited, these debates seem to distract from other areas that could impact more people. Trump seems especially good at finding meaningless conflicts to divide people, like NFL players’ actions during the national anthem or tweeting about Lavar Ball’s son being arrested in China.

Theorizing about how to combat this problem, I started making a list of what might be impactful-but-popular (or at least not unpopular) policies that would make up an idealized congressional agenda: nominal GDP futures markets, ending federal prohibition of marijuana, upgrading Social Security Numbers to be more secure, reforming bail. However, there is a big difference between “not unpopular”, “popular”, and “prioritized”. I’m pretty sure nominal GDP futures markets would have a pretty positive effect on Federal Reserve policy, and I can’t think of any political opposition to it, but almost no one is talking about it. Marijuana legalization is pretty popular across most voters, but it’s not a priority, especially for this congress. So what do you focus on? Educating more people about nominal GDP futures markets so they know such a solution exists? Convincing more people to prioritize marijuana legalization?

The nagging problem is that effective altruist groups like GiveWell have taken a research based approach to identify at what the best ways are to use our money and time to improve the world. For example, the cost of distributing anti-mosquito bed nets is extremely low, resulting in an average life saved from malaria at a cost in the thousands of dollars. The result is that we now know our actions have a significant opportunity cost; if a few thousand dollars worth of work or donations doesn’t obviously have as good an impact as literally saving someone’s life, we need a really good argument as to why we should do that activity as opposed to contributing to GiveWell’s top charities.

One way to make a case as to why there are other things worth spending money on besides GiveWell’s top charities, is to take a long term outlook, trying to effect a large change that would impact a large amount of people in the future.  For example, improving institutions in various developing countries would help those populations become richer. Another approach would be to improve the global economy, which would both allow for more investment in technology as well as push investment into developing countries looking for returns. Certainly long term approaches are more risky compared to direct impact charities that improve outcomes as soon as possible, but long term approaches can’t be abandoned either.

Existential Risk

So what about the extreme long term? What about existential risk? This blog’s philosophy takes consequentialism as a founding principle, and if you’re interested in the preceding questions of what policies are the most helpful, and where we should focus our efforts, you’ve already accepted that we should be concerned about the effects of our actions. The worst possible event, from a utilitarian perspective would be the extinction of the human race, as it would not just kill all the humans alive today (making it worse than a catastrophe that only kills half the humans), but also ends the potential descendants of all of humanity, possibly trillions of beings. If we have any concern for the the outcomes of our civilization, we must investigate sources of existential risk. Another way to state this is: assume it’s the year 2300, and humans no longer exist in the universe. What is the most likely cause of our destruction?

Wikipedia actually has a very good article on Global Catastrophic Risk, which is a broad category encompassing things that could seriously harm humanity on a global scale. Existential risks are a strict subset of those events, which could end humanity’s existence permanently. Wikipedia splits them up into natural and anthropogenic. First, let’s review the non-anthropogenic risks (natural climate change, megatsunamis, asteroid impacts, cosmic events, volcanism, extraterrestrial invasion, global pandemic) and see whether they qualify as existential.

Natural climate change and megatsunamis do not appear to be existential in nature. A megatsunami would be terrible for everyone living around the affected ocean, but humans on the other side of the earth would appear to be fine. Humans can also live in a variety of climates, so natural climate change would likely be slow enough for some humans to adapt, even if such an event causes increased geopolitical tensions.

Previous asteroid impacts have had very devastating impacts on Earth, notably the Cretaceous-Paleocene extinction event some 66 million years ago. This is a clear existential risk, but you need a pretty large asteroid to hit Earth, which is unusual. Larger asteroids can also be more easily identified from further away, giving humanity more time to do something (push it off path, blow it up, etc). The chances here are thus pretty low.

Other cosmic events are also low probability. Gamma-ray bursts are pretty devastating, but they’d have to be close-by (with a few hundred light-years at least) as well as aimed directly at Earth. Neither of these is likely within the next million years.

Volcanism is also something that has the potential to be pretty bad, perhaps existential level (see Toba Catastrophe Theory), but it is also pretty rare.

An alien invasion could easily destroy all of humanity. Any species with the capability to travel across interstellar space with military ambitions would mean they are extremely technologically superior. However, we don’t see any evidence of a galactic alien civilization (see Fermi Paradox 1 & 2 and The Great Filter). Additionally, solving this problem seems somewhat intractable; on a cosmic timescale, an alien civilization that arose before our own would likely have preceded us by millennia, meaning the technology gap between us and them would be hopelessly and permanently large.

A global pandemic seems pretty bad, certainly much more likely than anything else we’ve covered in the short term. This is also exacerbated by human actions creating a more interconnected globe. However, it is counterbalanced by the fact that no previous pandemic has ever been 100% lethal, and that modern medicine is much better than it was during the Black Plague. This is a big risk, but it may not be existential. Definitely on our shortlist of things-to-worry-about though.

Let’s talk about anthropogenic risks next: nuclear war, conventional war, anthropogenic climate change, agricultural crises, mineral exhaustion, artificial intelligence, nanotechnology, biotechnology.

A common worry is nuclear war. A massive nuclear exchange seems somewhat unlikely today, even if a regional disagreement in the Korean peninsula goes poorly in the worst possible way. It’s not common knowledge, but the “nuclear winter” scenario is still somewhat controversial, and I remain unconvinced that it poses a serious existential threat, although clearly a nuclear exchange would kill millions. Conventional war is also out as it seems strictly less dangerous than a nuclear war.

For similar reasons to nuclear winter, I’m not quite worried about global warming on purely existential terms. Global warming may be very expensive, it may cause widespread weather, climate, and ecological problems, but I don’t believe humanity will be entirely wiped out. I am open to corrections on this.

Agricultural crises and mineral exhaustion seem pretty catastrophic-but-not-existential as well. These would result in economic crises, but by definition, economic crises need humans to exist; if there are fewer humans, it seems that an agricultural crisis would no longer be an issue.

The remaining issues are largely technological in nature: artificial intelligence, biotechnology, nanotechnology, or technical experiments going wrong (like if the first nuclear test set the atmosphere on fire). These all seem fairly concerning.

Technological Existential Risk

Concern arises because technological progress means the likelihood that we will have these technologies grows over time, and, once they exist, we would expect their cost to decrease. Additionally, unlike other topics listed here, these could wipe out humanity permanently. For example, a bioengineered virus could be far more deadly than what would naturally occur, possibly resulting in a zero survival rate. The cost of DNA technology has steadily dropped, and so over time, we might expect the number of organizations or people who have the knowledge and funding to engineer deadly pathogens to increase. The more people who have this ability, the more likely that someone makes a mistake and releases a deadly virus that kills everyone. An additional issue is that it is quite likely that military research teams are right now researching bioweapons like an engineered pathogen. Incentives leading to the research of dangerous weapons like this are unlikely to change, even as DNA engineering improves, meaning the risk of this threat should grow over time.

Nanotechnology also has the potential to end all life on the planet, especially under a so-called “grey goo” scenario, where nanobots transform all the matter on Earth. This has a lot of similarities to a engineered pathogen, except the odds of any human developing some immunity no longer matter, and additionally all non-human life, indeed, all matter on Earth is also forfeit, not just the humans. Like biotechnology threats, we don’t have this technology yet, but it is an active area of research. We would also expect this risk to grow over time.

Artificial General Intelligence

Finally, artificial general intelligence contains some similar issues to the others: as technology advances, we have a higher chance of creating it; the more people who can create it, the more dangerous it is; once it is created, it could be deadly.

This post isn’t a thesis on why AI is or isn’t going to kill all humans. We made an assumption that we were looking exclusively at existential risk in the near future of humanity. Given that assumption, our question is why will AI be more likely to end humanity than anything else? Nonetheless, there are lingering questions as to whether AI is an actual “real” threat to humanity, or just an unrealistic sci-fi trope. I will outline three basic objections to AI being dangerous with three basic counterarguments.

The first objection is that AI itself will not be dangerous because it will be too stupid. Related points are that AI is too hard to create, or we can just unplug it if it has differing values from us. Counterarguments are that experts disagree on exactly when we can create human-level AI, but most agree that it’s plausible in the next hundred or couple hundred years (AI Timelines). It’s also true that we’ve seen improvements in AI ability to solve more general and more complex problems over time; AlphaZero learned how to play both Go and Chess better than any human without changes in its base code, YouTube uses algorithms to determine what content to recommend and what content to remove ads from, scanning through thousands of hours of video content every minute, Google’s Pixel phone can create software based portrait photos via machine learning rather than needing multiple lenses. We should expect this trend to continue, just like with other technologies.

However, the difference between other technological global risks and AI is that the machine learning optimization algorithms could eventually be applied to machine learning itself. This is the concept of an “intelligence explosion”, where an AI uses its intelligence to design and create successively better versions of itself. Thus, it’s not just that an organization might make a dangerous technological breakthrough, like an engineered virus, but that once the breakthrough occurs, the AI would rapidly become uncontrollable and vastly more intelligent than us. The intelligence analogy being that a mouse isn’t just less smart than a human, it literally doesn’t comprehend that its environment can be so manipulated by humans that entire species depend on the actions of humans (i.e. conservation, rules about overhunting) for their own survival.

Another objection is that if an AI is actually as intelligent as we fear it could be, it wouldn’t make “stupid” mistakes like destroying all of humanity or consuming the planet’s resources, because that wouldn’t count as “intelligent”. The counterpoint is the Orthogonality Thesis. This simply states that an AI can have any goal. Intelligence and goals are orthogonal and independent. Moreover, an AI’s goal does not have to explicitly target humans as bad (e.g. “kill all the humans”) to cause us harm. For example, a goal to calculate all the digits of pi or solve the Riemann Hypothesis might require as much computing power as possible. As part of achieving this goal, a superintelligence would determine that it must manufacture computing equipment and maximize energy to its computation equipment. Humans use energy and are made of matter, so as a way to achieve its goal, it would likely exterminate humanity, and convert all matter it could into computation equipment. Due to its superintelligence, it would accomplish this.

A final objection is that despite experts believing human level AI will happen in the next 100 years, if not sooner, there is nothing to be done about it today or that it is a waste of time to work on this problem now. This is also known as the “worrying about overpopulation on Mars” objection, comparing the worry about AI to something that is several scientific advancements away.  Scott Alexander has an entire blog post on this subject, which I recommend checking out. The basic summary is that AI advancement and AI alignment research are somewhat independent. And we really need to learn how to properly align AI values before we get human level AI.

We have a lot of theoretical philosophy that we need to figure out how to impart to a computer. Things like how humans actually make decisions, or how to value different moral tradeoffs. This could be extraordinarily complicated, as an extremely smart optimization algorithm could misinterpret almost everything we say if it did not already share our values for human life, health, and general brain state. Computer scientists set out to teach computers how to understand natural human language some 60 years ago, and we still haven’t quite nailed it. If imparting philosophical truths is similarly difficult, there is plenty of work to be done today.

Artificial intelligence could advance rapidly from human level to greater than human very quickly; the best human Go player lost to an AI (AlphaGo) in 2016, and a year later, AlphaGo lost to a new version, AlphaGo Zero, 100 games to none. It would thus not be surprising if a general intelligence achieved superhuman status a year after achieving human-comparable status, or sooner. There’s no fire alarm for artificial general intelligence. We need to be working on these problems as soon as possible.

I’d argue then, that of all scenarios listed here, a misaligned AI is the most likely to actually destroy all of humanity as a result of the Orthogonality Thesis. I also think that unlike many of the other scenarios listed here, human level AI will exist sometime soon, compared to the timescale of asteroids and vulcanism (see AI Timelines, estimates are highly variable, anywhere from 10 to 200 years). There is also a wealth of work to be done surrounding AI value alignment. Correctly aligning future AI with goals compatible with human values is thus one of the most important challenges facing our civilization within the next hundred years or so, and probably the most important existential threat we face.

The good news is that there are some places doing this work, notably the Machine Intelligence Research Institute, OpenAI, and the Future of Humanity Institute. The bad news is that despite the importance of this issue, there is very little in the way of conversations, money, or advocacy. AI Safety research is hard to calculate in total, as some research is likely done by private software companies, but is optimistically on the order of tens of millions of dollars a year. By comparison, the U.S. Transportation Security Administration, which failed to find 95% of test weapons in a recent audit, costs $7.5 billion a year.

Further Reading

I have focused this essay on trying to convey the mindset of thinking about existential risk generally and why AI is specifically worrying in this context. I’ve also tried to keep it short. The following are further resources on the specifics of why Artificial General Intelligence is worth worrying about in a broader context, arranged by length. If you felt my piece did not go in depth enough on whether AI itself is worth being concerned about, I would urge you to read one of the more in depth essays here which focus on that question directly.

 


Leave a comment on the official reddit thread. 

Book Review: The Libertarian Mind

The full title of this book is The Libertarian Mind: A Manifesto for Freedom, written by David Boaz, Executive Vice President at the Cato Institute. This is actually the second edition of Libertarianism: A Primer, published in the late 90s by Boaz.

So why did I read an introductory book on libertarianism? Well, it had been a while since I’d really looked at a libertarian book, especially critically. As discussed in What is Postlibertarianism? v2.0, I’ve strayed a bit from a libertarian absolutist, and that post in an attempt to carve out a space independent from both the Right and Left, but also perhaps libertarianism itself. It seemed this might be a good time to revisit some of the basics to see if I had forgotten what had made libertarianism so appealing in the first place. David Boaz’s introduction to the political philosophy seems to be a good way to do that.

Intro and Libertarian History

The book is a solid introduction to libertarianism. Boaz discusses important libertarian talking points like the fact that the two-party political system in the US doesn’t necessarily hold all the answers. He also does a fair job tracing the history of liberalism in political philosophy, culminating in modern libertarian thinkers. That’s one of the better chapters of the book, and similarly, perhaps the most useful segment is Boaz’s recommended reading list on various libertarian topics, located in the final pages. There are literally hundreds of libertarian readings and authors mentioned, and I plan on adding a few to my future reading list.

I have never been as familiar with the pedigree of American conservatives and American progressives, and I would be curious to see what their similar reading lists or genealogy would look like.  Libertarianism included routes through Locke, Mill, Mises, Friedman, Nozick and many more. It was clearest here that while I may not agree entirely with the label of “libertarian” today, there is a broader liberal tradition, wide and powerful in scope, and it is squarely within that tradition that I find myself. 

Obviously then, I had broad strokes of agreement with this book in many areas, but I wanted to point out a few areas that I thought did a good job of applying libertarian critiques or approaches.

Positives

Boaz talks a lot about rights and rights-based approaches, which I’m not quite as excited about as I used to be (see Rules and Heuristics). Nonetheless, he makes a strong case for the consequentialist benefits of property rights: they reduce the amount of issues that must be political. Application of property rights settles disputes, allowing individuals to make choices about who they interact with and how. Alternatively, if the state is dictating policy, e.g. education policy, all education is determined by politics. Political losses then have greater effect on individual lives, since it’s often harder to opt-out of state policies you dislike.

Relatedly, the chapter on pluralism and tolerance was excellent. Also well stated was the chapter on the rule of law. This is a nebulous concept, and I think Boaz does a good job discussing the many aspects, including constitutional law, the importance of judicial activism (would have been surprising to me 8 years ago) to protect individuals from government, general warrants, regulatory loopholes for specific companies, and overcriminalization. Each of these are fairly disparate parts of law, but they are all important breaches of a uniform rule of law, and contribute to delegitimatize the state and democracy. 

The chapter on public choice theory resonated, and I especially liked the terminology of a “package deal” to refer to political candidates, and how that could be so limiting. And as you would expect from a libertarian, the discussion of free markets, price theory, opportunity costs, and free trade were pretty straightforward. One highlight included the importance of entrepreneurial profits and the value of entrepreneurs seeing value missing in the economy, taking risks, and profiting by fulfilling needs. Another was the argument that the “balance of trade” wasn’t a useful measure since it doesn’t acknowledge that by definition, goods are traded by individuals. Individuals benefit from trade because they wouldn’t take part in it otherwise. Trade balances don’t take into account international supply chains routed all over the world, simplifying imports to two countries, when value added can come from dozens.

Negatives

Now for things that didn’t quite work. The book acknowledges the fact that several of the founding fathers were slave owners. Nonetheless, since the book doesn’t spend much time on anything, it only lends a couple pages to the issue of slavery. That isn’t going to convince anyone from the social justice movement.  This is a recurring issue. Many times I did object to a point the book brought up, but there’s no time for any in-depth discussion, so most of the time I remained unconvinced.

For example, in the rule of law chapter, Boaz attacks the concept of unaccountable bureaucracy, demonstrating how bureaucratic rules can be authoritarian with no accountability. Nonetheless, elitist independent agencies could make more sense than democratic Congressional loudmouths; the alternative to bureaucracy isn’t necessarily that the government doesn’t perform that job, but that it is left to unrestrained democratic pressures. 

The book also spends some time arguing not just that welfare is expensive, but that it’s actively harmful. I’m not sure how much I agree, but welfare for the poor never seems like it should be the first priority of spending cuts; the top federal budget items are Medicare, Social Security, and Defense spending. I actually thought the discussion of mutual aid societies was intriguing although I’m not sure how well they’d work now. It was one of the better answers I’d heard of for the critique that bad things will happen if we get rid of the welfare state. Another related point: the book doesn’t state what a “good” tax level would be, just that we have high taxes now. It’s not wrong, but I found it a bit of a cop-out.

Finally, the book isn’t too concerned about inequality, like you’d expect. However, the claim was that innovative markets would constantly challenge and undermine those at the top, with new products and markets catapulting new successful entrepreneurs at the expense of the old. Again, this could be true, but there wasn’t enough time to really dig into it; certainly the Forbes top 400 richest people in the world would constantly change as markets shift over time, but would the richest 1% really be in much danger? Is it ok if they are not? Libertarians would probably also argue that market innovation and technological progress are more important than inequality (a poor person in 2018 has much more material wealth than a rich person in 1968), but are there political risks to allowing for large inequality? The book doesn’t have time to answer these critiques.

For my takeaways: the book did a bit better than I expected on pointing out that I still generally agree with the bulk of classical liberalism/libertarianism, and my critiques are more like policy tweaks than philosophical deal-breakers. However, it’s only an introductory book, and due to my knowledge in these areas, specific issues I have with libertarian orthodoxy weren’t well addressed, nor was they really meant to be.  I will definitely be looking at the extensive “For Further Reading” list for some libertarian writings on specific topics I’m concerned about. I would also state that this is a pretty good introductory book if you want 350+ pages from a representative libertarian. If you have already studied a lot of libertarian thought, I doubt you’ll find too much new here.

 


Leave a comment on reddit.

2018 Predictions

Untestable knowledgeable cannot be scientific.  To avoid the problems of retroactively placing events into your narrative of the world, predictions must be laid out before events happen. If you try to use your model of the world to create testable predictions, those predictions can be proven right or wrong, and you can actually learn something. Incorrect predictions can help update our models.

This is, of course, the basis for the scientific method, and generally increasing our understanding of the world. Making predictions is also important for making us more humble; we don’t know everything and so putting our beliefs to the test requires us to reduce our certainty until we’ve researched a subject before making baseless claims.  Confidence levels are an important part of predictions, as they force us to think in the context of value and betting: a 90% confidence level means I would take a $100 bet that required me to put up anything less that $90. Moreover, it’s not just a good idea to make predictions to help increase your knowledge; people who have opinions but refuse to predict things with accompanying confidence levels, and therefore refuse to subject their theories to scrutiny and testability, must be classified as more fraudulent and intellectually dishonest.

Before I take a look at how I did this past year, and see if my calibration levels were correct, I should look at some hard fork predictions I made in July:

  1. There will be a Bitcoin Cash block mined before 12 AM August 2, US Eastern time: 80%
  2. The price of Bitcoin Cash at 12 AM August 2, US Eastern time will be <10% of Bitcoin’s price: 70%
  3. The price of Bitcoin Cash on August 5 will be < 10% of Bitcoin’s price: 90%
  4. The price of Bitcoin Cash on September 1 will be < 10% of Bitcoin’s price: 90%
  5. The value of all transactions of Bitcoin Cash around September 1 (maybe averaged over a week?) will be < 10% of the value of all transactions in Bitcoin: 95%

I did not predict that Bitcoin Cash would have long term staying power. In retrospect, I should have had more confidence that it would be similar to Ethereum Classic, which has remained for over a year now.

Now for predictions made at the beginning of the year:

World Events

  1. Trump Approval Rating end of June <50% (Reuters or Gallup): 60%
  2. Trump Approval Rating end of year <50% (Reuters or Gallup): 80%
  3. Trump Approval Rating end of year <45% (Reuters or Gallup): 60%
  4. Trump 2017 Average Approval Rating (Gallup) <50%: 70% (reference)
  5. ISIS to still exist as a fighting force in Palmyra, Mosul, or Al-Raqqah: 60%
  6. ISIS to kill < 100 Americans: 80%
  7. US will not get involved in any new major war with death toll of > 100 US soldiers: 60%
  8. No terrorist attack in the USA will kill > 100 people: 90% (reference)
  9. France will not vote to leave to the EU: 80%
  10. The UK will trigger Article 50 this year: 70% (reference)
  11. The UK will not fully leave the EU this year: 99%
  12. No country will leave the Euro (adopt another currency as their national currency): 80%
  13. S&P 500 2017 >10% growth: 60%
  14. S&P 500 will be between 2000 and 2850: 80% (80% confidence interval)
  15. Unemployment rate December 2017 < 6% : 70%
  16. WTI Crude Oil price > $60 : 70%
  17. Price of Bitcoin > $750: 60%
  18. Price of Bitcoin < $1000: 50%
  19. Price of Bitcoin < $2000: 80%
  20. There will not be another cryptocurrency with market cap above $1B: 80%
  21. There will not be another cryptocurrency with market cap above $500M: 50%
  22. Sentient General AI will not be created this year: 99%
  23. Self-driving cars will not be available this year for general purchase: 90%
  24. Self-driving cars will not be available this year to purchase / legally operate for < $100k: 99%
  25. I will not be able to buy trips on self-driving cars from Uber/Lyft in a location I am living: 80%
  26. I will not be able to buy a trip on a self-driving car from Uber/Lyft without a backup employee in the car anywhere in the US: 90%
  27. Humans will not land on moon by end of 2017: 95%
  28. SpaceX will bring humans to low earth orbit: 50%
  29. SpaceX successfully launches a reused rocket: 60%
  30. No SpaceX rockets explode without launching their payload to orbit: 60%
  31. Actual wall on Mexican border not built: 99%
  32. Some increased spending on immigration through expanding CBP, ICE, or the border fence: 80%
  33. Corporate Tax Rate will be cut to 20% or below: 50% (it was 21%)
  34. Obamacare (at least mandate, community pricing, pre-existing conditions) not reversed: 80%
  35. Budget deficit will increase: 90% (Not if you go by National Debt increase January to January)
  36. Increase in spending or action on Drug War (e.g. raiding marijuana dispensaries, increased spending on DEA, etc): 70% (hard to say: Rohrbacher AmendmentFY2018 DoJ changes)
  37. Some tariffs raised: 90% (reference)
  38. The US will not significantly change its relationship to NAFTA: 60%
  39. Federal government institutes some interference with state level legal marijuana: 60%
  40. At least one instance where the executive branch violates a citable civil liberties court case: 70% (I made this too broad as I can cite Berger v New York and the NSA violates it every day)
  41. Trump administration does not file a lawsuit against any news organization for defamation: 60%
  42. Trump not impeached (also no Trump resignation): 95%

Postlibertarian

  1. Postlibertarian.com to have >15 more blog posts by July 1, 2017: 80%
  2. Postlibertarian.com to have >30 blog posts by end of year: 70%
  3. Postlibertarian.com to have fewer hits than last year (no election): 60%
  4. Postlibertarian Twitter account to have <300 followers: 90%
  5. Postlibertarian Twitter account to have >270 followers: 60%
  6. Postlibertarian Subreddit to have <100 subscribers: 90%

I missed all the ones I marked as 50% confident, but I’ve realized this category conveys no mathematical information. I could have also listed the predictions as simultaneously saying that there was a 50% chance the exact opposite of the statement occurred, so actually, I got exactly half of them right, and I will always get exactly half of them right. This makes the category completely useless, and so I have decided to avoid posting any predictions of exactly 50% accuracy for next year.

In the other categories:

  • Of items I marked as 60% confident, 10 were correct out of 13.
  • Of items I marked as 70% confident, 5 were correct out of 7.
  • Of items I marked as 80% confident, 9 were correct out of 12.
  • Of items I marked as 90% confident, 7 were correct out of 9.
  • Of items I marked as 95% confident, 2 were correct out of 3.
  • Of items I marked as 99% confident, 4 were correct out of 4.

This may not look great, but is better than last year. Additionally, the big problem is the 95% predictions, which was severely hurt by my poor decision to make predictions about the Bitcoin hard fork, an event which hadn’t really happened before. Ignoring those predictions made in July would change my scores to:

  • Of items I marked as 60% confident, 10 were correct out of 13.
  • Of items I marked as 70% confident, 4 were correct out of 6.
  • Of items I marked as 80% confident, 8 were correct out of 11.
  • Of items I marked as 90% confident, 6 were correct out of 7.
  • Of items I marked as 95% confident, 2 were correct out of 2.
  • Of items I marked as 99% confident, 4 were correct out of 4.

That’s actually remarkably well, with perhaps some 60% predictions that needed more confidence. Moreover, it’s clear I had no business making predictions about Bitcoin with such high confidence, nor did anyone this year. I will definitely be dialing back my confidence levels in Bitcoin price predictions next year, and I’ve focused a bit more of whether Drivechain will be adopted.

Predictions for 2018:

World Events

  1. Trump Approval Rating end of year <50% (Gallup): 95%
  2. Trump Approval Rating end of year <45% (Gallup): 90%
  3. Trump Approval Rating end of year < 40% (Gallup): 80%
  4. US will not get involved in any new major war with death toll of > 100 US soldiers: 60%
  5. No single terrorist attack in the USA will kill > 100 people: 95%
  6. The UK will not fully leave the EU this year: 99%
  7. No country will leave the Euro (adopt another currency as their national currency): 80%
  8. North Korea will still be controlled by the Kim dynasty: 95%
  9. North Korea will conduct a nuclear test this year: 70%
  10. North Korea will conduct a missile test this year: 95%
  11. Yemeni civil war will still be happening: 70%
  12. S&P 500 2018 >10% growth: 60%
  13. S&P 500 will be between 2500 and 3200: 80% (80% confidence interval)
  14. Unemployment rate December 2018 < 6%: 80%
  15. Unemployment rate December 2018 < 5%: 60%
  16. WTI Crude Oil price up by 10%: 60%
  17. Price of Bitcoin > $10,000: 70%
  18. Price of Bitcoin < $30,000: 60%
  19. Price of Bitcoin < $100,000: 70%
  20. Lightning Network available (I can complete a transaction on LN): 80%
  21. Drivechain development “complete”: 70%
  22. Drivechain opcodes not soft-forked into Bitcoin: 70%
  23. No drivechains soft-forked into existence: 95%
  24. US government does not make Bitcoin ownership or exchange illegal: 90%
  25. Self-driving cars will not be available this year for general purchase: 95%
  26. Self-driving cars will not be available this year to purchase / legally operate for < $100k: 99%
  27. I will not be able to buy trips on self-driving cars from Uber/Lyft in a location I am living: 95%
  28. I will not be able to buy a trip on a self-driving car from Uber/Lyft without a backup employee in the car anywhere in the US: 90%
  29. Humans will not be in lunar orbit in 2018: 95%
  30. SpaceX Falcon Heavy rocket will attempt to launch this year (can fail on launch): 95%
  31. SpaceX will not bring humans to low earth orbit: 60%
  32. No SpaceX rockets explode without launching their payload to orbit: 60%
  33. Mexican government does not pay for wall: 99%
  34. Border wall construction not complete by end of 2018: 99%
  35. Some increased spending on immigration through expanding CBP, ICE, or the border fence: 80%
  36. No full year US government budget will be passed (only several months spending): 90%
  37. US National Debt to increase by more than 2017 increase (~$500B): 70%
  38. Increase in spending or action on Drug War (e.g. raiding marijuana dispensaries, increased spending on DEA, etc): 70%
  39. Some tariffs raised: 90%
  40. The US will not significantly change its relationship to NAFTA: 70%
  41. Federal government institutes some interference with state level legal marijuana: 70%
  42. Trump administration does not file a lawsuit against any news organization for defamation: 90%
  43. Mexican government does not pay for wall 99%
  44. Trump not removed from office (also no Trump resignation): 95%
  45. Democrats do not win control of Senate: 60%
  46. Democrats win control of House: 60%

Postlibertarian

  1. postlibertarian.com to have 10 new posts by July 1, 2018: 80%
  2. postlibertarian.com to have 20 new posts this year: 80%
  3. Postlibertarian to have more hits than last year: 70%

 

*I modified prediction #31 on January 24th from 70% positive to 60% negative. This feels early enough that I can still call it a prediction, and I’m not sure why I was so confident in December when I wrote these.


Leave a comment on reddit.

Bitcoin Value Questions

Does Bitcoin offer something of value today?  Does it have the potential to be more valuable in the future? Here are some thoughts how you might be able to answer yes or no to these questions.

I.

The first point is a question of how currencies have value. How does the US dollar have value? In a very concrete and practical sense, the dollar is valuable due to legal tender laws, where any legitimate transaction that occurs in the US must accept US dollars as a form of payment. Moreover, US taxes must be paid in dollars. However, that’s not a majority of the dollar’s value.

The US dollar has value because people believe it will be accepted in the future. That’s why the dollar is valuable in countries outside America where users are presumably not under US legal tender laws. Why do people believe it has value? Well partially its derived from the practical points made above combined with the size and scope of the US economy; if dollars are used in the United States, often by legal mandate, and if the US economy is large and vibrant, it will need lots of dollars. The US economy, even if it struggles, won’t be gone overnight, so you can bet in five or ten years, there will be plenty of transactions that need to occur in dollars. There’s also the point that trade with people in the United States mean dollars cross borders pretty easily. This creates a self-fulfilling prophecy; since people know there are Americans and traders who will accept dollars, other people accept dollars too, knowing they will be accepted in the future.

That accounts for the demand side of dollars. On the supply side, there is at least implied trust in the US central bank, the Federal Reserve. This may rub Ron Paul fans the wrong way, but I think it’s somewhat undeniable. People in the US and outside see the inflation track record of the American dollar and agree that it’s unlikely to be really poorly managed. Perhaps that’s just because alternative central banks are even less trustworthy, perhaps it’s because the Fed has a reputation of being stingy about inflation. It’s hard to say. What is undeniable is that the US dollar is widely used and held throughout the world.

II.

Does Bitcoin have a role to fulfill in the market when the US dollar serves as an excellent international medium of exchange and store of value? Yes. Bitcoin is inherently digital, meaning you just need some information, on a computer, in your head, or written on paper, in order to use it. Dollars require a bank, and if international, they require a bank that reports to a local government which may or may not allow foreign currency holdings.

This means today Bitcoin offers some advantages over American dollars in certain situations without any scaling updates to the Bitcoin network that we’ll discuss later. Such areas include international transfers, domestic currency mismanagement, and anonymous transactions.  International transactions because all you need is an internet connection, not a bank or Western Union office. Bitcoin transactions have fees, but they can be lower than international wire fees. Domestic currency mismanagement is Bitcoin’s clearest use case. Venezuela has experienced hyperinflation as its currency is worth less than World of Warcraft gold. Bitcoin has become highly useful as it does not lose its value over time like Bolivars. Bitcoin also saw a spike in India when they unanimously outlawed large denomination cash bills. In another interesting case Zimbabwe actually uses the US dollar (after hyperinflation destroyed the currency last decade), but because they cannot print it, liquid cash is scarce in the country, so Bitcoin is highly valuable since it is more easily imported than dollars.

Finally, Bitcoin is of course useful for illicit activities, such as the fabled Silk Road dark net trading site.  Not much to add here, except to point out that another cryptocurrency, Monero, may actually fill this niche better if you’re just looking for confidential transactions. More on other cryptocurrencies in the final section.

III.

However, if you are in a developed country, it’s unlikely Bitcoin is better than your national currency in terms of ease of use, acceptance by merchants, quickness of transactions, cost of transactions, etc. Certainly people who believe in Bitcoin politically can pay these increased costs and use it anyway, but that’s essentially paying for a political statement.

Bitcoin may be a better long term store of value than a state currency, e.g. the US dollar. It is governed by an algorithm as opposed to a committee. Algorithm changes are difficult and slow, and there is currently a cap on the total number of Bitcoins that will ever be created. If the US hits the Fed’s estimated inflation target of 2%, then the value of any currency owned by residents will halve in about 34 years.  However, Bitcoin is volatile, and buying it as a store of value uses it as an investment. Some Bitcoin investment today is certainly speculation. And if a decent chunk of the Bitcoin price is caused by investment/speculation instead of current usefulness, then a better store of value/investment could rapidly pull the money out of Bitcoin. Perhaps some investment is acceptable, but doing more radical actions, like putting your life savings in something that can lose its value relatively quickly isn’t a good idea.

We should keep in mind that there are people even in developed countries that have limited access to banking and credit. Large commercial banks are notorious for charging fees to customers who specifically don’t have the cash to spare on those fees. Bitcoin may be a way for those with poor access to banks to “be their own bank” and hold their savings securely without needing a national bank. Perhaps transfer fees are too high to make this practical, but at the very least, this is a potential market for Bitcoin, if scaling issues can be solved.

There is one other use case where Bitcoin is clearly superior to even a developed world currency. That would be a tax-free asset and currency. It’s not particularly difficult to purchase Bitcoin and then launder it through another cryptocurrency or through CoinJoin (an anonymization protocol) and make the money untraceable. Assuming Bitcoin’s basic use cases of international transactions and troubled currency refuge continue to grow, Bitcoin offers a big tax haven. I should note, of course, that this is plainly illegal, and I suspect the more tax evasion an individual undertakes, the more likely they are to be scrutinized by authorities.

IV.

We’ve established Bitcoin has explicit use cases and therefore offers value today. We’ve also established that some of these uses cases may grow in the future. What about threats to Bitcoin’s value?

If a significant use case of Bitcoin is illicit transactions and tax avoidance, then I would claim Bitcoin is a direct threat to the state, even in developed countries. As stated in “What is Postlibertarianism? v2.0“, widespread adoption of cryptocurrencies could mean the end of taxable transactions, and possibly the end of the modern state. I’m not interested in making a judgment about whether this is good or bad, but I think the threat to states is undeniable (if still very far away).

The obvious next question: if states have an incentive to stop Bitcoin, can they do it In cases where Bitcoin has solid use cases, as in Venezuela and Zimbabwe, it seems highly unlikely. Bitcoin was built to be censorship resistant; deleting a node does almost nothing to the network, as all nodes are peer-to-peer and you can quickly switch to talking to another node or two or fifty. To shut down a Bitcoin payment network in a country, you’d likely have to shut down access to the outside internet. However, with new developments in the Bitcoin space, even partitioning a country’s internet from the outside won’t work anymore; Blockstream is currently broadcasting Bitcoin blocks from geostationary satellites (yes, really) to most of the world. Their goal is total global coverage. However, you can only receive the blockchain, not send transactions with this technology. So recently, Nick Szabo and Elaine Ou introduced a protocol for sending and receiving Bitcoin transactions (and block headers) over HF radio.

In reality, Venezuela hasn’t made Bitcoin illegal anyway. It seems unlikely that Nicholas Maduro’s ineffective government could substantially threaten the internet. China, while having the Great Firewall and having shut down Bitcoin exchanges, has not made the possession or use of Bitcoin illegal. These technologies are really only a just-in-case scenario. However, if you do live in a country with no internet or interaction with the outside world (North Korea), you still might not be able to use Bitcoin; no internet, no distributed systems, no censorship resistance (although the North Korean government itself uses Bitcoin to avoid international sanctions).  While I have to concede this point, it’s also important to acknowledge that technological advancement has enabled South Korean soap operas to be smuggled across the border; in the future Bitcoin may find a way into the Hermit Kingdom as well.

However, North Korea is one of the worst-case situations. In almost any other country, cheap computing technology and simple internet infrastructure has taken hold in an irreversible trend. And that’s all that’s really needed to use Bitcoin.

…Probably. What if a high trust societies made Bitcoin illegal? What if the United States and Europe made it illegal to own or transact in Bitcoin? I don’t think this is likely, as democracies tend be very slow when it comes to legislation, especially regulation where financial markets can make a lot of money. Moreover, institutional investors have already created legitimate companies in the US and Europe and so there would be lobbying, deliberating, compromising, etc. Japan has already recognized Bitcoin as an official form of payment, and if nothing else, the US making Bitcoin illegal would create an odd situation for American citizens living in Japan and vice versa.

But let’s say it happens.

It’s undeniable that Bitcoin’s value would drop. If you were already using Bitcoin for illicit activity, you might keep using it, but it might expose you to additional legal risk where it didn’t before. However, if you were using Bitcoin as an investment/speculative vehicle or as a way to send international transfers, an illegal Bitcoin is significantly less appealing because it would expose you to legal risk that you wouldn’t otherwise have to deal with at all. Bitcoin’s growth proposition wouldn’t be zero, but it might be pretty grim, and perhaps relegated to country’s with weak state legitimacy (and where widespread mistrust of the state means ordinary activities are criminalized anyway).

However, like I said previously, I find this scenario unlikely. Moreover, the Bitcoin network isn’t just waiting for governments to act, it is constantly under development with a large technical community.

V.

Can Bitcoin scale to take on more roles and use cases? Can it upgrade to become more censorship resistant? Definitely.

One big item we’ve talked about before is the Lightning Network. The idea behind the LN is pretty simple: you can create payment channels by putting some Bitcoin in escrow through a time-locked transaction that is signed but not posted to the blockchain. This channel can be continually updated with new transactions representing different payments back and forth across the channel until the channel closes by posting the final “net” transaction to the Bitcoin blockchain (read more about it here). This uses the blockchain as a settlement layer, and saves on transaction fees since only two transactions are ever posted to the blockchain (to open and close the channel) even if lots of payments occur.

There is another interesting aspect of this technology, which is that you can use a LN channel as an initial hook into a larger network. So if you (Person A) already have a channel open with Person B, you could pay Person C without opening up a new channel as long as both B and C have a channel between them already open. A pays B, then B pays C, and everyone updates their current balances on two payment channels, but no one needs to post anything to the blockchain, so no transaction fees are needed.

This is pretty good for scaling. However, it is somewhat negative for privacy. The most efficient way any Lightning Network will exist is through large central hubs. This is because end users will want to open a single payment channel (since it’s cheaper and ties up fewer funds), so they will want to connect to a hub everyone else is connected to. A hub that doesn’t stay available all the time would be unhelpful if you want to make instantaneous payments at any time, so the trend will be towards large, continuously available hubs. These hubs will also need access to lots of liquid cash as they will have lots of funds tied up in open channels, while also needing to have liquidity available to open new channels at any time.

This will lead to hubs with lots of cash and thus corporate backing. These large hubs will best be able to scale lots of LN instant payments while keeping LN node fees low. However, a central payment hub would have lots of information about its users, users who are using a single Bitcoin address for all of their transactions. Thus each address would have much more information leaked to the LN hub nodes, which you could track across time.

Of course, if you wanted more anonymity, you could just use a regular Bitcoin transaction; any service or individual who has a Lightning address must by definition have a Bitcoin address. This seems a reasonable tradeoff: instant transactions that can be tracked over time vs anonymous transactions that you pay a higher fee per use.

VI.

Another impressive project is Drivechain.  This project would allow for sidechains in the Bitcoin ecosystem. These would be soft-forked in (that means no network split), and these sidechains would not need to impact the mainchain. The sidechain could run its own nodes independent of the Bitcoin chain, although in practice we would expect Bitcoin nodes to watch the sidechains since we would imagine sidechains would only exist if there was significant value added there. The way these work is that Bitcoin would be sent to an escrow account watched by the sidechain. That would allow those coins to appear on the sidechain and be governed by any rules the sidechain wants.

Interesting sidechain ideas include Hivemind (decentralized Bitcoin prediction markets) and MimbleWimble (homomorphically encrypted confidential transactions). Needless to say, there is an enormous amount of potential here. Drivechains would allow limitless innovation, allowing new blockchain rules to flourish while maintaining the network effects and avoiding the coordination failure of multiple currencies or blockchains.

However, there are risks with this approach. One risk is that money stored in the sidechain is sitting in an escrow account on the mainchain. Mainchain nodes don’t have to watch the sidechain, and so if incorrect transactions are posted trying to withdraw money from the sidechain, it’s up to the miners to enforce the correct rules. As long as miners believe sidechains enhance the value of Bitcoin, there shouldn’t be a problem.  But if we don’t get to that point quickly, drivechains could be a short-lived experiment ending in grand theft. I’m hopeful this is not the case though, and sidechains would offer such a massive increase in the value of Bitcoin that several will survive and grow.

VII.

Let’s take a moment to elaborate on the implications here.  The creation of a MimbleWimble sidechain or the addition of the related idea of Confidential Transactions to Bitcoin would be game changers for Bitcoin privacy. Tax avoidance with Bitcoin would become simple, easy, and possibly unstoppable. Combined with improved scaling or the essentially limitless use cases for Bitcoin sidechains, there will be a combination of high demand and availability of Bitcoin with widespread privacy.  Even if governments can continue to collect tax revenues, their ability to combat Bitcoin would be completely diminished.

The interesting corollary is that governments aren’t really getting in the way of Bitcoin. Maybe they’ll crack down on it in the future, but for now there isn’t a lot of indication for heavy regulation. In the US, electoral politics means there will be a deregulatory environment for the next year, maybe three.

Finally, the Bitcoin and cryptocurrency space is not done developing. Sidechains offer the potential to incorporate all sorts of new rulesets and innovation into Bitcoin. The potential here is literally unknowable. For these reasons, I believe Bitcoin has the potential for significant value.

I would also of course like to point out that this is just some blog on the internet so take my advice as policy speculation and not investment speculation. There are plenty of other financial risks to Bitcoin I don’t have time to cover. This includes that if you lose your private keys, your money is gone forever. It includes that there could be an unknown flaw in the Bitcoin code that could be exploited, losing money and crashing the price of Bitcoin. It includes that government agencies could compromise developers and pay them off to put in code that helps to destroy the network. Bitcoin is risky and speculative. The fact that it has a lot of potential does not guarantee that it will have value in five years.

VIII.

A final note on other cryptocurrencies. There are many other cryptocurrencies, and I’m doubtful on all of them for two reasons. (1) If Drivechain is successful, most use cases for other coins will be gone. (2) As it is, even if other chains have cool features, they don’t have the network effects of Bitcoin. Collective action failures mean that better features may be passed over if it involves transaction costs distributed over many individuals; in other words, it will be nearly impossible to get users, vendors, developers, and miners to switch over to a different cryptocurrency. In the long run, we’d probably expect one or two cryptocurrencies to dominate. This may be Bitcoin or it may be something else, but today, Bitcoin is the clear market leader. To bet on another cryptocurrency is to bet against the market and to bet against the large ecosystem that Bitcoin has built. This seems very risky.

Thanks for reading, and if you enjoyed this, feel free to donate to the Bitcoin address on the sidebar!

 


Leave a comment on the official Reddit thread.

Electoral Reform Fantasies

It’s been a particularly divisive…month? year? presidency?  Maybe you could even argue this last decade or so has been increasingly polarizing. Last election cycle specifically was unlike anything we’ve seen in the modern political era in terms of highly unpopular candidates running against each other, just look at the numbers:

Trump won with the lowest popular vote percentage of any president since Bill Clinton in 1992, when Ross Perot ran as a third party candidate getting 19% of the vote. In fact, Trump won the lowest percentage of any president in US history when no third party got more than 5% of the vote. Actually, we can go further; every case in which a US president was elected with less than Trump’s 46.1% had a third party getting over 8% of the vote that year. Except 2016.

Thus, we should first acknowledge that political frustration with political parties is nothing new in American politics. The only difference is that this time, there are no other parties to turn to.

This is a problem. Organizations acquire rules and absorb ideas over time. Sometimes those ideas are toxic to the organization, and it is out-competed. I’m mostly imagining the creative destruction of the market, but the same logic can apply to religions, non-profits, and political parties. However, the Republican and Democratic parties have constructed excellent barriers to entry, helped along by American electoral rules. Perhaps these barriers to entry have always existed, but they seem particularly effective at present.

I believe this lack of competition has resulted in two parties that are having difficulty providing a platform for new political ideas or approaches. Without competitive pressures, there is a lack of popular outlet and political advocacy, resulting in frustration. With only two political parties to work with, the idea of a political dichotomy seems inescapable, with every single culture battle melding together to become one gargantuan struggle between two fiercely divided tribes.

This is by no means the only problem we face: sluggish postindustrial economic growth, cost disease, shrinking populations, etc, are all issues. However, it’s quite possible our outdated political system may be stifling any solutions. Thus, I’d like to provide some ideas to fix the way we run our democracy.

Primaries

Presidential primaries seem to be the toughest to fix, but primaries themselves would become much less important with other reforms. Primaries today tend to favor more extremist candidates, while general elections (and, by definition, most people) favor more centrist ones.

One way to solve this is with an open primary, which some states have. California even has an “open blanket” primary, where the top two vote-getters in the primary are on the ballot in the general election, regardless of party. Of course, California does not use such a system for president (Donald Trump would have likely not been on the ballot if they had). There are drawbacks here, as theoretically several centrist candidates could split the “centrist” vote and leave two extremists running in the general election.

One possible way to help improve the presidential primaries might be to rotate the order in which states are the “first” primary. Iowa has often been the first state, but New Hampshire actually has a law that it must be the first presidential primary by a week (Iowa has caucuses, so New Hampshire has decided those don’t count). New Hampshire isn’t a great bellwether: going back to 1980, in election years where a candidate won a competitive primary and then won the presidency (i.e. not 2012, 2004, 1996, 1984 when a sitting president was re-elected), New Hampshire got Donald Trump in 2016, George H. W. Bush in 1988, and Ronald Reagan in 1980. It wrongly selected Hillary Clinton over Obama in 2008, John McCain over George W. Bush in 2000, and Paul Tsongas over Bill Clinton in 1992.

Iowa isn’t any better. It selected Obama in 2008 and George W. Bush in 2000. And it wrongly selected Ted Cruz over Donald Trump in 2016, Tom Harkin over Bill Clinton in 1992 (Harkin was from Iowa, but Paul Tsongas came in second, not Clinton), Bob Dole over George H. W. Bush in 1988, and George H. W. Bush over Reagan in 1980.

So in our first two primary states over the last 30+ years are 3/6 and 2/6 respectively when picking a president from a competitive field. Not great.

There’s some merit to simply holding a national primary all at once. The argument against it is that this may bias the primary system against discovering good lesser known candidates who can campaign in small states more easily than a national stage. However, there’s no evidence indicating such a system of candidate discovery functions with the small states at present. Maybe we need other states that better represent a microcosm of the country. Maybe such states don’t exist.

Ballot Access

Did you wonder why there wasn’t a well-known centrist Republican candidate running as a third party in the race last year? It seemed to be the perfect storm. A significant minority of Republicans were not a fan of the party’s nominee; the party’s previous nominee had called out Trump in an aggressive speech earlier in the year, and the Democrats had nominated a fairly progressive, well known candidate that most conservatives disliked.

Well, it turns out there was one, Evan McMullin, but he was only on the ballot in 11 states, accounting for 84 possible electoral votes.  Why? Because it costs hundreds of thousands of dollars to get onto the ballot in most states. The Libertarian Party candidate Gary Johnson was the only third party candidate on the ballot in all 50 states. In fact, he was the first third party nominee to be on all 50 ballots since 1996. Johnson did better than previous Libertarian Party candidates, and so the LP will not have to spend as much money for ballot access in the coming cycle…yet they are still looking to raise $130,000 this year just for ballot access costs.

This needs to change. There can be no serious competition to the current parties without fixing the ballot access problem.

Take a look at the Wikipedia article on the topic for a good overview. One problem is that major political parties are often exempted from ballot access requirements entirely. Other times, parties that get over a certain percentage of the vote are not required to gather signatures. The signatures are often rejected, so in reality the signature requirements are really 20-30% higher than actually stated. Ohio is an interesting example, as it requires a candidate to file in March, before they are actually nominated at their party’s convention. To get around this, the LP of Ohio filed a placeholder candidate in 2016, and then changed it to Gary Johnson later in the year. Of course, he had to file as an independent candidate since Ohio’s independent requirements are much less burdensome than trying to get the Libertarian Party be recognized as a state political party.

A possible solution would be to at least even the playing field by having a federal law forcing all qualification rules to apply to all parties running for federal office, including the Republican and Democratic parties. This would require them to waste resources on gathering signatures as well. Of course, the major parties could handle large numbers of signatures more easily since they have more resources available, but it still might be difficult enough to push them to reduce the total number of signatures to more practical levels.

More direct reductions in the ballot access requirements would be great as well, but perhaps not as directives from the federal government for the sake of federalism. Of course, none of this will happen, as there are no third party members in office at the national level, and thus no interest in reforming third party access at the state level.

House of Representatives – Single Transferable Vote

This one is totally crazy I know. It would definitely require a change in law, as it’s currently against the rules to have more than one representative from a district. However, I don’t suspect it would be unconstitutional, as each state creates their own districts and runs their own elections.

An STV system is unambiguously better than our current system. Single Transferable Vote is a voting system where you rank several candidates in a multi-member district. The candidates that reach a threshold of support (something like 33% for a three seat district, 25% for a four seat district, etc) are elected. If not enough candidates reach the threshold, unpopular candidates are eliminated with voters’ next choices receiving their votes instead, until all seats are filled. This helps achieve a proportional representation while maintaining local legislators. Currently all Representatives are elected in single member plurality elections, also known as First Past The Post (FPTP). For an easily digestible explanation of STV, watch CGP Grey’s video on the system.

STV systems do well when there are many seats available in a single district. Ireland has used as many as six seats in a single district, Tasmania has used as many as seven. Given the US population of 320 million, the average congressman represents over 700,000 people, with the median being even higher. However, many Americans live in cities much larger than 700,000, and so there are many cities that could support single citywide districts with five or ten congressional seats filled by STV. These could much better reflect the diverse viewpoints of those living in cities. Of course, cities wouldn’t be the only ones who benefit from this, as gerrymandering can also be done to disenfranchise rural voters depending on who’s drawing the boundaries.

Gerrymandering is itself much harder with STV multi-member districts. This itself is an indication that an STV system is better than what we have now. Even if STV is poorly implemented with districts that only have three or four seats, it would be a vast improvement in representation and political competition than what we have today.

This reform is certainly the most important reform for third parties. I don’t think third parties will solve all our problems; other countries have plenty of third parties with little to show. But it’s certainly a necessary step in providing alternatives to the duopoly people are obviously very sick of. Moreover, even if third parties aren’t super successful, the threat of competition will force the two major parties to react. We need a diversity of opinions and new ideas, and without third parties, everything has to be filtered through a party system with vested interests and previous baggage.

President – Approval Voting

The electoral college system is supposed to select a candidate from a wide range of possible candidates, with the college of electors itself imagined as acting as a bulwark against the excesses of democracy. This didn’t really pan out the way the founders of the United States might have hoped. Instead, several elections have resulted in presidents being elected despite other candidates actually receiving a plurality of the popular vote.

Those were:

  • 1824, when Andrew Jackson won 41% of the vote in a split election that was thrown to the House of Representatives since no one had an electoral college majority. The House picked John Quincy Adams, who lost in 1828 to Andrew Jackson. This one is less concerning because there was no clear majority, so while Jackson didn’t like it, the system “worked”.
  • 1876, when Samuel Tilden handily won an outright majority of the popular vote, and probably won the electoral college, but a “bipartisan” commission gave 15 “disputed” electoral votes to Rutherford Hayes instead. I’m still bitter.
  • 1888, when sitting President Grover Cleveland won a close popular vote victory, but lost in  the electoral college to Benjamin Harrison. Cleveland would win the rematch (both popular vote and electoral college) in 1892.
  • 2000, when Al Gore won a plurality of the vote, but lost Florida by a few hundred votes, and so George W. Bush became president.
  • 2016, when Hillary Clinton won a plurality of the popular vote, but Donald Trump won the electoral college.

If we set aside 1824, which I think is reasonable, we have 4 elections out of 58 total in American history in which the electoral college has selected against the popular vote winner, despite only two major candidates in those elections. This is an error rate of 6.9%.

But how to fix this? There have been several times when the electoral college was helpful in sorting out a multi-candidate election.  In 1860, Lincoln won a plurality with only 39.8% of the vote, but the electoral college gave him a majority. From a voting system perspective, this may not be seen as a victory, as Lincoln’s election was so divisive, it precipitated southern secession. However, in 1912, Woodrow Wilson won the electoral college with only 41.8% of the vote in a three way race. 1968 and 1992 may also be considered elections where the electoral college helped establish a winner when the plurality winner only had vote totals in the low 40s.

Moreover, any debate about the electoral college, especially after this most recent election must necessarily have political implications. Nonetheless, I believe I have a system that is strictly better than our current system, preserving any usefulness it has. The proposal is as follows.

Ballots for president will ask two questions, one asking the voter to select all candidates which they will be ok with being president (approval voting) and one asking voters to select their single favorite candidate (first past the post/ our current voting system).

The president will be chosen based on who receives the highest percentage in the approval voting ballots, as long as the percent total is above a threshold. Here I’m recommending 55%. In the case of no candidate receiving above 55% of the vote, the system simply defaults back to the electoral college system using the second, first past the post / favorite candidate vote.

I suspect this would encourage much more positive campaigns, as candidates try to attract as many voters as possible rather than scare voters away from voting from their opponents. It would also make third party campaigns much more useful, as there is less strategic voting with approval voting. If a popular centrist party had a candidate with broad appeal across the spectrum, they could get votes without causing right or left wing voters to fear their votes are “wasted”. Moreover any candidate that wins the approval vote would have a strong mandate with a super-majority of voters supporting them. This is what the electoral college was supposed to bring us, a wide base of support for the president, but this system will guarantee it outright.

In the worst case scenario, if I am wrong about these predictions, the system is simply what we have right now, today. There is no way for it to do worse than our current system since it’s fall back is our current system. In this way, it is also conservative and gradual in its reform, in ways other voting systems are not.

Conclusion

These reforms are likely long shots, but I think it’s undeniable that our current system of government is deeply flawed. These are just my current best ideas, so if you read this and have some voting systems that you think would be more politically palatable or mathematically accurate, be sure to let me know on Twitter, Reddit, or email.

 


Leave a comment on the official reddit thread.

Picture Credit: Vote here, vote aqui. Erik Hersman. Licensed under CC BY 2.0.

Twitter Distractions

The U.S. military launched drone strikes on Libya on Friday, the first in Libya since January.  Trump has yet to mention these airstrikes as he’s been too busy fighting with professional athletes about how they protest.  If I’m counting correctly, there have been six Middle Eastern countries Trump has authorized military strikes in despite no authorization from Congress (and seven if you include Somalia): Libya, Syria, Yemen, Iraq, Afghanistan, Pakistan. Not to mention Trump has praised Saudi Arabia, a state that directly funds Wahhabism and an oppressive war in Yemen that does nothing to reduce radicalization.

Important criticisms of Hillary Clinton last year included her foundation receiving millions of dollars of support from the human rights disaster Kingdom of Saudi Arabia. But I’m not sure which is worse: taking bad people’s money or actively praising them. In fact, in what meaningful way is Trump’s Middle Eastern policy different from Clinton’s? Clinton was for a two state solution, while Trump didn’t seem to know what that meant–is that it?

Trump’s foreign policy has been pretty incompetent in other areas outside the Middle East. He’s failed to provide appointments for many ambassador positions, including South Korea. Speaking of which, Trump said he would control North Korea, but the DPRK has conducted more missile tests during his presidency (that’s 7 months) than any presidency in history. Even by using his own stated (terrible) goals of renegotiating NAFTA, tearing up the Iran nuclear deal, and reducing sanctions on Russia, he has failed to do what he said he would. In the case of Russia sanctions, this came at the hands of his own party overruling him in Congress.

Trump is a loud, robust failure in foreign policy. And rather than spend any energy actually trying to end military involvements like he said he would, or even do routine things like appoint ambassadors, he is igniting culture wars on Twitter. I think he prefers these to actual policy because there are no metrics to success when engaging in a cultural flame war online. It’s just “our tribe” vs “their tribe”, and no one can win because we’re not actually discussing anything. I think there are nuances to be had in this week’s particular flare up with the NFL and the national anthem, but they’re not worth teasing out because it’s so easy to get bogged down in an emotional fight.

So rather than engage with Trump’s culture war cage match this week, I think it’s more productive to point out that there are real issues he’s supposed to be dealing with, and he’s failing miserably. We’ve been at war for 16 years now. Soon, recruits will be traveling to battlefields that Americans have been fighting in since before these soldiers were born. But Trump would rather tweet about football players protesting.

 


Leave a comment on the official reddit thread.

Bitcoin Hard Fork Predictions

Tomorrow there is scheduled to be a hard fork of the Bitcoin blockchain and network. There’s a fair amount of uncertainty over what will happen. The hashrate is unknowable until the fork occurs. The price seems to be around 10% of the price of Bitcoin. However, there aren’t too many exchanges that will be accepting this currency, and there are even fewer places you can actually spend it.

I’m going to make some predictions about it to put on record what I think is going to occur and to see how correct or incorrect I end up being.

  1. There will be a Bitcoin Cash block mined before 12 AM August 2, US Eastern time: 80%
  2. The price of Bitcoin Cash at 12 AM August 2, US Eastern time will be <10% of Bitcoin’s price: 70%
  3. The price of Bitcoin Cash on August 5 will be < 10% of Bitcoin’s price: 90%
  4. The price of Bitcoin Cash on September 1 will be < 10% of Bitcoin’s price: 90%
  5. The value of all transactions of Bitcoin Cash around September 1 (maybe averaged over a week?) will be < 10% of the value of all transactions in Bitcoin: 95%

I have mixed hopes for the success of Bitcoin Cash. On the one hand, I wrote previously that if the two factions in Bitcoin split, we could have a competitive market showing which rules were better. However, due to network effects, I still don’t think it could happen and be very successful. Supposing it did succeed though (had a pretty high market price), what would that mean? I suppose it would mean forks would become more common. That might be better for competition, but not for stability of the currency.

Ultimately, the idea that it would be fairly easy to make a successful hard fork of Bitcoin would be pretty devastating to Bitcoin’s health. It would mean consensus doesn’t mean much, it would mean the Bitcoin community could splinter pretty easily, which would therefore mean Bitcoin’s usefulness as a currency decreases as each part of the community would be using their own forked blockchain and coin. Something like sidechains seems like a much better implementation of this idea.

I should probably also disclose that I do not have much faith in the current governance model of Bitcoin Cash, and that does concern me a bit as well. I hope that hasn’t clouded my judgment of the actual technological and economic implications, but only time will tell if my predictions are true.

Free Market Assumptions in Healthcare

I’ve encountered an unexpected concept when debating and discussing healthcare solutions in the United States.

Healthcare in the United States contains both public and private actors, but it’s most important characteristic for a libertarian critique is its lack of price signals. Healthcare is not purchased in an open market. Patients select healthcare providers based on reputation and what their insurance covers, but most patients do not choose their insurance provider. That is usually done by their employer or the government (in the case of Medicare and Medicaid). Conversely, healthcare providers do not charge patients, they charge insurers where prices can differ by provider and by procedure. EconTalk recently had Christy Ford Chapin on to discuss the history of American healthcare and I would highly recommend the episode.

The libertarian position (and mine) is that healthcare could be improved with prices. There are many ways to do that, you do not necessarily need patients to pay those prices, but you need them involved in the decision making process. Otherwise, there is no downsloping demand curve, and therefore there is no incentive to improve efficiency in the market. Thus, we see higher prices over time instead of the usual results of technological innovation: higher diversity of choices, higher quality goods, and lower prices.

This post is not a defense of whatever Republican healthcare bill is now being floated to replace or repeal Obamacare. This is only an argument that having known prices and price transparency would allow for demand and supply curves in the health care market. Such a characteristic could be part of a host of possible healthcare policy landscapes, and I’m only saying that a landscape that has prices is likely better than one that does not. Today, and for most of the history of healthcare in America, the healthcare industry has not been governed by an openly priced market.

Such a libertarian critique is separate from the argument that a “free-market” system with prices would hurt the poor. This is a valid critique that could be addressed with direct cash grants or other form of government subsidy that avoids having healthcare prices set by the government (refundable tax credits, health savings accounts, etc).

The remarkable argument I’ve heard is that if we allowed healthcare to be purchased in a market with prices, it would fail because you “can’t have” a free market in healthcare. It’s hard to nail down exactly what these people imagine would happen, but it seems that they believe prices themselves would not obey the laws of demand and supply. I will now list some arguments I have heard, some of them several times, and why they are incorrect. Certainly these arguments are poor and perhaps I am wasting time with them, but apparently they are common enough that I have run into them several times and therefore must be addressed.

“Healthcare Demand is Inelastic”

This is by far the most common point I’ve heard. It’s not usually stated in economic terms, but more like “if you are in need of emergency medical care, you’ll pay any amount, and this breaks normal market assumptions”.  However, I’ve also heard it stated that evidence of competitive markets working in elective procedures (Lasik or plastic surgery) does not apply to regular medicine because of demand inelasticity.

Firstly, the assumption that demand for medicine as a good is totally inelastic (i.e. quantity would not respond at all to price) is obviously wrong. That would imply there are no unnecessary procedures done ever.  Yet we all are aware that because doctors are often paid per procedure, they are often incentivized to conduct tests because there is very little downside (i.e. it costs neither the patient nor the doctor anything to run the extra test). If there is no elasticity, then there is no room for reducing the amount of procedures done by doctors. I doubt that.

Nonetheless, let’s grant the assumption, or at least let’s say that demand elasticity is very low.  That means at higher prices, you’re likely to consume a similar amount of medicine. That sounds more reasonable; if you’re sick, it’s not your choice.

Ok well…so what? We can have competitive markets with marginal revenue very close to marginal cost even if demand elasticity is low. Gasoline is a classic example of an inelastic good, yet the gasoline market is highly competitive. Prices work without issue here. Again, we’re not saying that poor people would be really happy with prices, we’re just saying that prices would exist if patients could purchase healthcare in a market.

Another related point is that if you have a medical emergency, you’re not really in a position to negotiate prices. This, however, is not just due to demand inelasticity, but also monopoly pricing.  If you’re injured, you can’t just go to a different hospital, so the ER you arrive at is pretty much the only place you can go. This is a fairly good argument for government intervention in the ER. However, insurance is also a pretty good solution; if you won’t be able to make a choice in the moment, you buy insurance so that when the moment comes, you are already prepared. There is no economic reason that medical emergency insurance could not be purchased in a free market. Additionally, medical emergencies are a small part of the medical industry. The vast majority of medical procedures are not emergencies, and so for most situations, monopoly pricing is not an issue.

“Knowledge is imperfectly distributed in medicine”

Again, the fact that market actors have imperfect knowledge does not mean a market cannot exist. It may mean there are market failures, but government interventions are subject to government failures which may or may not outweigh the benefits of trying to fix the market failure. Moreover, this proves way too much, as it implies that you can’t have any market with imperfect knowledge, yet all markets suffer from this, and plenty are functioning just fine. In fact, I’d argue that prices are the single best way to spread knowledge.

Imperfect knowledge is usually fixed through regulation, like accreditation or inspections. You don’t have the knowledge to know that your airplane doesn’t have mechanical problems. Nonetheless, you are quite capable of comparing the prices of different airplane tickets, and you’ll likely respond to market forces when purchasing a ticket. Certainly imperfect knowledge is an argument for regulation, and I’m sure I’d disagree with plenty of people on how much regulation is necessary, but there is no world in which it then makes sense to argue that imperfect knowledge precludes a functioning price system.

“People are irrational”

First, people don’t have to be economically rational, nor do markets have to be free from regulation in order to create accurate economic modeling. To make the claim that economic analysis can’t be done with healthcare because the market is not perfectly competitive, or actors are not perfectly rational, again proves too much; economic analysis would be “fatally flawed” in all markets. The only question that needs to be asked is whether it’s possible patients might call two different places for a quote on a chest CT or an MRI. If some of them would do this, there would be competitive pricing, even if most don’t know what an MRI actually does.

“Healthcare is too expensive for a market to function”

This point sort of ignores the thesis that we are arguing, as all I’m trying to say is that prices can exist in the healthcare market. However, this is related and while it’s a bad argument, I want to address it briefly.  Healthcare is pretty expensive, although I suspect that it would be cheaper if market prices were used. The obvious answer to me would be to imagine if the government gave a large amount of money to an individual to pay for their healthcare for a year. That would fix the endowment issue where the poor are excluded from the market. In this hypothetical, my thesis suggests that there would be a variety of options for healthcare spending, such as paying out of pocket, buying a high deductible insurance plan, subscribing to a doctor network, etc. All of these would be examples of functioning markets in healthcare. Additionally, if recipients were allowed to roll over funding into the next year, they’d be incentivized to find good deals this year.

My thesis is not that the government should stay out of healthcare, but that interventions that keep prices in place are preferred.

“Morally, patients should not have to pay for healthcare”

Again, this isn’t really an argument against my thesis, but I have heard it. It’s a bad argument, so I’ll address it briefly.

If we take a consequentialist utilitarian moral standpoint, there is no a priori humanitarian reason why patients should not pay for part of their healthcare. In other words, if patients paying for part of their healthcare creates benefits for all of society, including almost all patients and future patients, then the moral thing to do (from a utilitarian perspective) is to have patients pay for some of their healthcare.

So would there be benefits if patients paid for healthcare? Well, first you have to establish that prices can exists. We’ve done that for the theoretical, but how about the empirical?

Empirical Data

The first point is that in the area closest to healthcare where there are transparent prices, elective procedures, we see functioning markets with costs going down over time.  Highlights include:

1. For the top ten most popular cosmetic procedures last year, none of them has increased in price since 1998 more than the 45.4% increase in consumer price inflation (the price for the hyaluronic acid procedure wasn’t available for 1998), meaning the real price of all of those procedures have fallen over the last 18 years.

2. For three of the top five favorite non-surgical procedures in 2015 (botox, laser hair removal and chemical peel), the nominal prices have actually fallen since 1998 by large double-digit percentage declines of -25.2%, -43.8% and -23.5%. That is, those prices have fallen in price since 1998, even before making any adjustments for inflation.

3. Most importantly, none of the ten cosmetic procedures in the table above have increased in price by anywhere close to the 93% increase in medical care services since 1998. The 23.2% average price increase since 1998 for last year’s top five most popular surgical procedures, isn’t even close to half of the 93% increase in the cost of medical care services over the last 18 years.

However, there are some doctors who just take cash for normal, non-elective procedures. These would be procedures where there is “inelastic demand”. What happens to these doctors? Do they go bankrupt immediately? Is everyone confused and bewildered? Not really, it just works like any other market. They post their prices online, and people come and pay for their procedures directly, without insurance. The Oklahoma Surgery Center is one of the more well known health centers with this approach:

The Surgery Center would charge $19,000 for his whole-knee replacement, a discount of nearly 50% on what Villa expected to be charged at his local hospital. And that price would include everything from airfare to the organization’s only facility, in Oklahoma City, to medications and physical therapy.

And once that happened, lots of groups were incentivized to send their patients there, making other Oklahoma hospitals compete.

While no organization keeps track of how many cash-based medical centers have cropped up nationwide in recent years, Smith and Lantier say they’ve witnessed an explosion. In Oklahoma City alone there are roughly three dozen centers that are all or partly cash based, specializing in everything from radiology to oncology.

The RAND institute ran an RCT in the late 70s that found patients who cost shared saw a reduction in unnecessary procedures. Obviously it’s pretty old, but I’m doubtful human nature has changed that much from the late 70s; if people have an opportunity to save money, they will do so. Healthcare policy should utilize that.

More recently, in 2008, Oregon had a Medicaid experiment, where several people were given access to Medicaid based on a lottery. Thus, a study was conducted to determine what the affects were of having access to Medicaid. As you would expect, patients with Medicaid coverage were much more likely to utilize healthcare generally, and more likely to go to the ER. The price of medical care went down when this group was enrolled in Medicaid, and consumption of medical care went up. This supports the notion that healthcare has a downsloping demand curve…just like every other market that has ever existed.

Finally, there was a study done in 2015 looking at the healthcare system and it’s lack of prices. It found that transaction prices, that is prices negotiated between hospitals and insurers, still accounts for much of the differences in private inpatient healthcare spending. It also found that even after controlling for several different variables, hospital monopoly power was responsible for higher prices. This seems to indicate to me that if we had significantly more price transparency in a functioning market, hospitals and patients would respond to those incentives, creating incentives for lower prices and better, more efficient care.

Conclusion

This isn’t revolutionary by any means, but there’s seems to be plenty of empirical and theoretical reasons that if we had transparent pricing systems in the healthcare industry, it would function similarly to prices everywhere else in the economy. Certainly the use of insurance complicates things, but the way we use medical insurance is a result of the unique way we created the medical payments system as detailed in the EconTalk episode mentioned at the top of the post. There is no technical reason we need to retain that system, and I think transitioning towards more procedures having known prices would be beneficial, whatever that system would be.

 


Leave a comment on the official reddit thread.

Links 2017-7-16

The Libertarian Web Directory has been updated with some more interesting people, notably Ben Shapiro and Kmele Foster.

At the end of June, the libertarian academic world was alight in condemnation of Duke Professor Nancy MacLean’s book, Democracy in Chains. The book tries to tie James Buchanan, the Nobel laureate economist who essentially founded Public Choice Theory (an important part of how I’ve defined postlibertarianism) to racist segregationists, the Koch brothers, and Donald Trump. It’s poorly researched, takes quotes out of context, and implies guilt by several degrees of separation. As it turns out, Duke University’s Political Science Department is perhaps the world’s best place to learn about Public Choice Theory with several presidents of the Public Choice Society teaching there, including James Buchanon’s frequent co-author Geoffrey Brennan. Professor MacLean decided against talking to any of them, favoring instead to try and fabricate a grand right-wing conspiracy. The reason I know this is because Duke Political Science Professor Michael Munger wrote a devastating response to MacLean’s book that I highly recommend reading. It’s not just great for inter-academic drama, but the piece does excellently summarizes and discusses the important parts of Public Choice Theory, and how James Buchanan created and utilized this model of government.

Speaking of defining postlibertarianism, I’ve amended my explanatory article to be version 2.0, since technically former blogger Joshua Hedlund defined postlibertarianism pretty well 6 years ago. We’ll call that 1.0.

The Economist has a nice short piece on the benefits and deficiencies of Medicaid, which apparently covers some 80 million Americans, and about 100 million people some time during each year. Republican efforts to target Medicaid back towards actual poor people now makes much more sense to me.

Towards the end of the month, developments will happen in Bitcoin. Either BIP 91 will be activated through the SegWit2x agreement, or SegWit2x will fail and BIP 148 will start to orphan blocks on August 1. Either way, it seems Segregated Witness has a high chance to be activated. If you have no idea what I’m talking about, this is a useful guide to the upcoming months.  If you have no idea why Bitcoin would need activate anything, read my last post on Bitcoin here.

Google has collected a group of “classic papers”, which are papers that were highly cited in the subsequent 10 years. These papers are from a vast array of subjects and areas. Whatever area you are interested in is there.

I’m sure I’ve been one of the libertarians disappointed by Rand Paul’s support of many of President Trump’s initiatives and ideas. Matt Welch writes for Reason both acknowledging that politicians will always disappoint us, but also arguing that we should not give up on libertarian politicians when they drop a small bit of ideological purity.

This post at the blog Otium defending individualist culture was really long, but very good.

Scott Alexander discusses whether elections are decided by turning out your own base or by winning over voters from the other side. There isn’t a ton of good data on it, but he still makes a pretty compelling case that a small but important segment of the population are really “swing” voters that switch sides. One of the more interesting studies simulates RCTs by looking only at elections where a politician wins a primary by a very small percentage (like less than 0.1%). This means the chances of the more extreme candidate having won the primary are essentially random. In those cases, there is a significant difference between general election popular vote percentage depending on whether the moderate or more extreme candidate won the primary. Very cool.

The Fifth Column podcast from the week Donald Trump Jr.’s Russia emails came out is excellent. I recommend listening every week, but this episode is especially good. Besides my own blog about it, Politico has a nice summary by Jack Schafer, The LA Times had an interesting point about how Robert Mueller’s investigation pushed this information out in the open, and the Washington Post points out that the President approved of Trump Jr.’s Sunday statement which blatantly lied about the meeting last June.

Julia Galef has a good rundown of the most important disagreements between NIMBYs and YIMBYs, specifically on whether San Francisco should allow more housing to be built.

Trump Junior and Russia

This weeks’ Fifth Column Podcast had an excellent discussion of Donald Trump Jr. and his attempt to get opposition research from the Russian government.

I have been pretty skeptical that the Trump administration and campaign actually did anything wrong with regard to Russia. I thought Trump was a poor candidate because he seemed very comfortable with authoritarians like Putin, which together with many positions he’s taken made me concerned about authoritarian policies he’d implement. Some of that concern has been well founded (e.g. appointing Jeff Sessions, removing an FBI Director he didn’t like), but in other places it has not. Trump seems to be bothered by criticism and a free press, but it’s clear the press is as robust and independent as ever.

I also thought Trump surrounded himself with really bad people, some of them with obvious terrible connections to the Russian government, like Paul Manafort, Roger Stone, and Michael Flynn (all of whom were fired). Again, this points to his friendliness with authoritarians and perhaps his incompetence with whom he surrounds himself with, not that he’s had contact with the Russian government itself.

I thought the intense media scrutiny over Russia wouldn’t provide anything, and it was dragging on for far too long. I thought the worst thing we would find would be that Trump fired FBI Director Comey for investigating his ties to Russia, not for any connections to Russia itself. But I was wrong. The New York Times reporting on this was so impressive that Donald Trump Jr., the target of the Times‘ article actually felt it would be better if he published these emails himself!

Thus, I’m literally typing these word for word from Trump Jr.’s twitter account. He got an email saying “…and in their meeting offered to provide the Trump campaign with some official documents and information that would incriminate Hillary and her dealings with Russia and would be very useful to your father. This is obviously very high level and sensitive information but is part of Russia and its government’s support for Mr. Trump”

Trump Jr’s response: “…if it’s what you say I love it especially later in the summer”.

I don’t know if this is illegal, as Trump Jr. has some amorphous relationship with the Trump campaign and allegedly isn’t involved in the administration at all, despite regularly going on TV to defend the administration. However, it’s at the very least unpatriotic and seems pretty unethical. If agents of a foreign government are reaching out to you to attack political opponents, your response should be to forward it to the FBI, not to set up a meeting as soon as possible. Trump was elected to “drain the swamp”.  I thought that meant that many legislators are too close to special interests or are lobbied by large companies to pass favorable legislation. This is certainly problematic, but actively trying to get help from foreign governments to get into office seems at least as bad. Donald Trump Jr. is a swamp monster if ever there was such a thing.

Additionally, just like Hillary Clinton lied constantly about her email scandal (she didn’t have a private server; ok she did but there was nothing classified; ok it was classified but it was secure…), it turns out Trump Jr. and many other campaign members just straight up lied about having pursued Russian opposition research. Trump Jr. stated in a March interview that he had met people that were Russian but not meetings “…that were set up. None that I can think of at the moment. And certainly none that I was representing the campaign in any way, shape or form.” Here’s a full timeline, it’s pretty damning. This is the undermining of democracy that I thought Trump supporters claimed he was there to fight.

I wish all presidents were subjected to this sort of scrutiny. We had wall to wall media coverage for months with very few real stories until this week. The Obama administration crushed leakers and transparency, blocking FISA requests in an unprecedented manner. They brought government secrecy to a new level, while waging wars in several middle eastern countries without Congressional approval. Trump has only been in office for 6 months, and so it’s not hard to argue that the Obama administration was much worse in terms of doing illegal things, murdering civilians, spying on Americans and journalists, etc. Yet, media coverage has been much more aggressive on Trump (at least from the left, while the right-leaning media has often been at least as sycophantic as anything we saw under Obama on the left).  I guess it’s better late than never, but the media needs to be far more adversarial in the future long after Trump is gone.