Gettin' a little dusty around here...

It has been almost a year since I've posted, and a lot has happened in that time. First, I have been pursuing a MS in statistics in addition to my full time work. This has unfortunately led to putting my new business on hold until further notice. If/when I get a better handle on my own schedule and can start devoting the requisite time again, the business will hopefully take off.

The purpose of this post is actually just to share a link to something that I'm planning on pursuing in addition to my current educational efforts. Will Wolf has a great post on his blog about doing an "open source machine learning masters" that I am going to start working through. In my current role, I mostly do financial risk analysis but I'm working with a group of really bright and talented people on expanding our scope. We are creating an analytics team that will build out some tools to comb through data with the goal of integrating machine learning techniques to our current processes. This move toward data science and statistical learning is why I started on my stats MS, and it's something I'm really excited about. I'm hoping this blog can help keep me honest and focused on working through the program as Will describes in the link below. I am probably going to modify some of this as I go (it is open source, right?!), but for now I'm just hoping to get started.


http://willwolf.io/2018/02/03/practical-guide-open-source-ml-masters/

Starting a new business! And some thoughts.

I've been very busy lately as I have decided to start my own business, but I wanted to get some thoughts down here that I've had lately. Some of these thoughts stem from my research in launching a new business. I'm starting a financial advisory that will cater to younger, less affluent individuals to help them improve their financial situation. The reason I'm starting this company is that, after working in the financial services industry (I was a portfolio manager and also assisted in creating financial plans and advising 401k participants), I realized that the cost of advice is simply out of reach for most people. The company I worked for had a soft $300,000 minimum for assets. I say "soft" because if you were young and had $300,000, you were in. If you were over 60 and had $300,000, you were probably going to get referred somewhere else. The kicker? That $300,000 portfolio would land you in the 1.25% fee bracket, meaning you are paying $3750 per year out of your investments for advice. So if you were one of the very few who amass a large amount of wealth at a young age, you are still going to get killed on fees. My own plan is to charge an affordable monthly subscription instead, which I hope to open the doors to sound financial advice to anybody who is interested.

This leads me to my next thought. Much has been said about the dissatisfaction among the working poor and middle class. When Trump was elected, the reasoning that I saw (and continue to see) is that people wanted and outsider, someone to burn down the current institutions because, as they saw it, the current institutions weren't working for them. The middle class seems to have turned on "the elite" (I hate that term, but I'm referring to public figures in media, politics, and generally recognized experts in the private sector) because they feel betrayed. So, they now are turning to other sources of information, which mostly turn out to be uninformed iconoclasts. I laughed when Trump tweeted "who knew health reform was so hard." EVERYBODY WHO KNOWS ANYTHING ABOUT HEALTH MARKETS KNEW THAT IT WAS HARD. Sorry, I get a little excited about this stuff. Even after expert after expert talked about the difficult trade-offs that must be considered when creating health insurance regulation, they seriously had no clue that this stuff is complicated.

One of the biggest issues I found when preparing to launch this business is that with so much information available, it is more important than ever to have trusted experts to sift through all that information and discern right and wrong. Basically anybody can write anything and use the megaphone that is the internet to reach a large audience, but popularity is a poor proxy for accuracy. Unfortunately, experts are held in pretty low esteem right now. While part of that is on us, I think the experts are at least as responsible. Why? Well, if you are an expert on polling data, you should not try to use your expert platform to talk about infrastructure. And if you are an expert on infrastructure, you should use your platform to rationally and reasonably inform people about infrastructure. The hyperbole has gotten so awful (even when people talk about things they are well qualified to talk about) that it seems like every conversation turns into an ideological battle. These conversations end up so far removed from facts, they do nothing but misinform and create deeper divisions.

I'm not sure how this business will turn out. Because I'm going to price this service at a very low cost, I will be doing it part time. Even if it is a booming success, I would have to take a steep pay cut to do it full time. So why am I doing this? Because I think it's important. The term "death spiral" gets thrown around a lot when people talk about health insurance. Well, people's personal situations are starting to teeter on the edge of a death spiral. Poorly informed financial choices combined with distrust of actual experts could spell ruin for all but the wealthiest individuals. If you are interested in learning more, feel free to contact me (@thetillett on twitter).

A quickie - thoughts on DeLong's article in Vox

This is just a quickie (both in thought process and writing, so forgive the brain dump), I'm entirely too busy with work/school/family right now and so this is going to be neglected a lot more than I would like. But I did see this article that I wanted to share - it is from Brad DeLong on Vox about trade and it's a pretty good piece.

The comments I hear following the election is that the common wisdom regarding trade is wrong; that trade isn't the boon that everybody has always that it is. In 2016, nobody wanted to come out and defend free trade. Instead, we had both parties pledging a reduction in trade deals, and even potentially increasing trade barriers. In the linked article, DeLong pushes back against trade's critics. From the article, trade has not been the main cause of manufacturing job loss. Technology has displaced more workers than trade, and it isn't even close.

Noah Smith has recently attacked a lot of "foundational economic principles" as not really working in the real world. Among those, Smith has criticized trade, saying that it may actually produce a net negative benefit. I have a lot of difficulty believing that. The argument Smith provides is that trade is reducing income and personal/social well-being in exchange for cheaper stuff. The value of the cheaper stuff, according to Smith, is not enough to offset the reduction in jobs and reduced social status for displaced workers.

Isn't this just the opposite of the diffuse cost, concentrated benefit argument? In this case, we have diffuse benefit and concentrated cost. Everybody gets cheaper (and generally higher quality/dollar spent) goods, but some people get a really awful outcome, even with cheaper goods, because their previously well-paid skills are no longer valuable. Isn't that still an overall better outcome than before? It sounds awful, but wouldn't we want to generally raise everybody's standard of living a little, even if a small minority's standard of living is reduced by a lot? Another issue with this argument against free trade is that even if we don't allow companies to offshore labor or import foreign goods, those jobs are going away regardless. The reason is simple - people are expensive and inefficient. So now, not only are we not going to get cheaper, better goods, but people are going to lose their jobs anyway. The problem that free trade has created is a workforce training problem - we turned skilled labor to unskilled labor. The solution, then, to free trade is that it requires a system to mitigate the negative labor externalities.

Megan McArdle - Those Pesky Principles

It has been a month since I've posted, and it may be another month before I'm on here again the way things are going, but I just read Megan McArdle's article in Bloomberg View and found it interesting. Read the article, it is worth the few minutes.

As I wrote in the last post, I think Trump is a travesty. He represents so much of what is wrong in America, and he does so in a way that almost seems like it should be a joke - that he is some sort of caricature of a Fox News/Rush Limbaugh politician. I feel pretty confident in saying that it is not a joke, he is exactly as awful as he seems. But McArdle makes a really good point in the article - for as awful as Trump is, he knows how to spin the right story. Before even taking office, he has been "making deals" and putting on a show. The self dealing with foreign governments and Trump properties are disgusting, but I think the Carrier deal is probably even worse. If Trump continues to enrich himself, his businesses, and his family through the power of the presidency, he will probably end up impeached, perhaps sued, and maybe even imprisoned. The Carrier deal is just as wrong, but will likely have much farther reaching and longer lasting effects. And the worst part is that Trump's fans love it.

McArdle describes this very problem. Why do so many Americans support this type of protectionism and interventionism? From the article:

Trump’s tiny deal with Carrier was far more politically effective than all the Democratic railing about traitorous companies moving their headquarters abroad. After all that railing, what Democrats were proposing to do in the end was just to create even more impersonal laws, some incomprehensible amendment to an already opaque tax code. What Trump gave them was direct action -- action that said “I’m on your side, and I’m not going to sell you some nonsense about abstract economic principles; I’m going to go out there and save your jobs right now.”

There is an old saying in writing - show, don't tell. Democrats told voters they would fight to keep companies in America, Trump showed them he would. Unfortunately, both of these are bad positions to take, though you wouldn't know it considering the rhetoric throughout both campaigns. But both of those policies sound more appealing than talking about market forces and long run increases in total welfare. Basically, sound policy has a messaging problem.

These well-worn arguments, however true, have the unsavory flavor that Anatole France aptly summed up in an epigram: “The law, in its majestic equality, forbids the rich as well as the poor to sleep under bridges, to beg in the streets, and to steal bread.”

This is the nature of principles-based systems. They are cold and impersonal, and ostensibly neutral machinery often produces results that look grossly unfair by any common-sense moral standard: rich people getting better treatment from the legal system than poor ones, bankers who gambled on mortgage bonds getting a safer retirement than truck drivers who gambled on an overpriced house. If the “fairness” of following principle departs too widely from the “fairness” of following our intuitions, then people are going to start asking what’s so great about those principles.

Who is going to fight for principles that seem to disproportionately benefit wealthy over poor? McArdle brings up the case of banking executives such as Dick Fuld of Lehman Brothers and how well they all seem to have fared after the 2008 crash. While many of those executives have lost substantial amounts of money, power, and prestige, they are still wealthier than 90%+ of Americans. That isn't a satisfying consequence for people who came out of the financial crisis without a job, home, or any savings. But, as McArdle points out, there have been plenty of eager prosecutors that would happily make a name for themselves by throwing the Fulds of the world in jail. The problem is that there isn't enough compelling evidence that a crime was actually committed.

For people who work in finance, this probably isn't that surprising. Financial instruments can be very complicated - there is a reason so many math and physics PhDs work for investment firms. But to compare it to something more tangible, suppose you have a company that makes cars. Well, what if you wanted to make a self charging electric car. This new car uses a new charging system and battery that can go 1,000 miles without stopping. It is a brand new technology, but your engineers assure you that it is safe. They have run all kinds of tests in the lab and based on the test results, the car would only fail if subjected to temperatures of over 120 degrees for more than six hours. Based on simulations, the probability of that happening is 0.01%, and hasn't happened in the US since 1974. So you start selling these new cars and they are immensely popular. Several years later, however, you see on the news that a heat wave is rolling over the southwest with temperatures well above 120 degrees in several large cities. What those engineers didn't think about was that when that extremely unlikely heat wave does come, it will impact a lot more than just a few cars. Within a week, you have a class action lawsuit saying that your cars have been linked to thousands of fires and $500 million in damages and your company goes bankrupt.

This is (mostly) what happened in the financial crisis. Banks and investment firms had developed a new product: credit default swaps ("CDS"). A CDS is basically like insurance on a loan. For example, let's say I owe the bank $1 million. My bank might rightly see that as a lot of risk concentrated on a single person. So they get an insurance company to write a policy that basically says the insurance company will pay off any debt I am unable to pay in case I declare bankruptcy. The bank pays the insurance company a premium and feels all warm and fuzzy because they don't have to worry if I go belly up. The problem in the financial crisis, is that insurance companies forgot to account for how risks relate to one another. They wrote all these insurance policies (CDSs) thinking that the possibility of each policy having to pay out as unrelated. They ran some simulations and found that it might be something like a 0.1% chance that they would get a claim and priced the insurance accordingly. The problem is that these entities don't exist in a vacuum. As it turns out, the conditions that caused one company to fail caused a lot of companies to fail, and suddenly companies like AIG are on the hook for billions of dollars.

In the case of the self charging car, I don't think people would be parading down Pennsylvania Avenue asking for the car engineers' and CEO's head on a platter. But they did come out with their pitchforks and torches ready for AIG, Lehman Brothers, and Bear Stearns. There were even calls for Obama to throw someone in prison even just as a symbolic gesture. Why? People wanted to see action. Explaining the complexities of derivatives and financial contracts wasn't going to pacify anybody. Unfortunately, McArdle doesn't have a great answer for this problem either. As the world becomes more complex, this issue is likely to get worse, not better.

For the post-election blues.

This is as much an exercise in therapy as it is a thought exercise, but last night at about 11:30pm, I went to sleep with near certainty that Donald Trump would be the next president. From the beginning of the election, I had prepared myself for another disappointment on election night. As a sort of bleeding heart libertarian-type, nobody in this election would have implemented the type of policy agenda that I can really get behind. I have spent the year becoming more and more dismayed at the type of rhetoric and (only very occasionally) policy coming from all sides. Even still, this election felt like a real gut punch. Yes, the sun still rose this morning, but the world felt a little darker, a little colder. (I realize this is November, and the world actually is a little darker and a little colder every day.) My first thought was, how in the world can we, the greatest nation on earth, elect Donald Trump, a man who embodies our nation's dark underbelly?

Hope can be hard to find in a moment like this. More so when you think of the 50+ million people that are celebrating this colossal failure. Yet I know people who voted for this monster, and despite that fact, they are my friends and my family. These are people that stand up on Sunday to sing about a loving God and denounce sin. These are people that have fought overseas to protect freedoms that Donald Trump finds inconvenient. These are people that believe in community and helping your neighbor, both the one down the street and the one on the other side of the planet. They have faced real struggle, and instead of succumbing, they have risen. They are doctors, they are engineers, they are pastors, and they voted for a cruel predator. How can I reconcile these things?

Plenty has been written about the correlation between Trump voters and racism, misogyny, nationalism, authoritarianism, economic anxiety, and everything else. As someone who lives in the heartland, which has been the fertile ground that gave rise to the Trump presidency, I give my best shot at explaining what I see and hear from those that supported a candidate who clashes with many of the "American Ideals" that I and others hold dear.

• Trump is not Hillary. When the Democratic Party nominated Hillary Clinton, they signed off on a candidate that even their own base didn't want. Say what you will, but the DNC managed to assemble a primary in name only that served the nomination to HRC on a platter. They disregarded their own constituency as it clambered for ANYBODY other than Clinton. After being wholly rejected in 2008, Clinton walked into the nomination as one of the least popular nominees ever. With such an unappealing alternative, more independents ended up voting for third party candidates this year and disaffected Republicans felt justified in voting for an even worse candidate.
• They didn't know what they were doing. When I would tell people about Trump's ties to Russia, his shady business dealings, his flirtations with white supremacists, etc., people were often surprised. I read a lot, and I read a wide variety of sources. These were issues I thought were beaten to death, yet others hadn't even heard of this. I attribute this mainly to three factors: a lack of quality journalism, partisan echo chambers, and election fatigue. The decline of print journalism is pretty self-explanatory. Fewer people subscribe to newspapers and the entire medium is struggling to adapt to the new economics of a digital age. Part of this new digital age of journalism, however, seems particularly bad at providing quality reporting that reaches a wide audience. I fear that the days of objective investigative reporting are behind us. Instead, we have a country that clings to their partisan media. Fox News, Breitbart, Red State, and the rest of the hyper-right media pump out absurd stories that serve the interests of their investors, not their listeners. When a significant percentage of Trump supporters believe that Clinton is an actual demon, we have gone all kinds of wrong. The "alt-right" commentariat lies, conceals, and confuses on its way to burning down the political fabric that holds this country together. Short term profits for long term anarchy, but hey, we're all dead in the long run, right? And last but not least, election fatigue. I can sympathize here, as someone who got sick of hearing about every skeleton in Trump's closet (there are A LOT). The election cycle was way too long. As the contest dragged on, people stopped paying attention. Trump's campaign manager worked for a pro-Russian upstart in Ukraine? Trump abuses women? No tax returns? At some point, people just stop caring and walk into the voting booth with a willful ignorance of just how awful Trump is.
• Deeply rooted political tribalism. I spent two years living in Poland as a missionary and part of my responsibility was to convince the people to change religions. Poland is a staunchly Catholic nation, and Pope John Paul II is a hero there. One phrase that I heard often during that time was "I was born Catholic, and I will die Catholic." Some people said that to me directly after telling me that they don't believe in the supernatural. Consider that. They are Catholic, and yet they not only believe something different than the Catholic church, they fully reject the entire premise. So it is with politics. These people grew up in staunchly Republican households. That red elephant is as much a part of their identity as their last name. When Trump said he could kill somebody in Times Square and it wouldn't matter, he is probably right. As long as his name is on the line for the Republican ticket, he will get at least 40% of the vote, and probably more. As with pretty much everything on this list, the same is true for Democrats. Romney was demonized for his 47% comment, but, tone deaf as it was, he was right. American politics are a team sport, and both teams are in it to win.
• The enemy of my enemy is my friend. In exit polls, voters willingly admitted to pollsters that they thought Trump would be a bad president, but they voted for him anyway. Why? Because he hates blacks/women/Jews/immigrants/Muslims/etc. Trump magnifies the worst in everybody. Even good people have deep-seated prejudice, and Trump preyed on fears of terrorism, crime, economic anxiety, and religious fanaticism. His campaign speeches portrayed an apocalyptic wasteland that was either already here or well on its way. I heard a lot of "I really don't like Trump, but he would stop group X from ruining our country."
• One last hope for those that thought hope was lost. Trump appealed to those who have been left behind in the digital economy. Factory workers, coal miners, steel mill workers, etc. have all seen their industries collapse with the rise of global markets. For the displaced worker who doesn't have much hope of regaining their dignity in a world they no longer recognize, Trump seemed like a risk worth taking. Trump explicitly promised to revive the coal industry in West Virginia and put people back to work. He promised tariffs and reversals of free trade agreements to bring jobs back onto American soil. Spoiler: those jobs are gone, regardless of who is president. While the blue collar crowd generally supports Democrats, Trump took up their mantle with much more vigor (and magic fairy unicorn dust) than Clinton. With the Trans-Pacific Partnership, she was for it only until it became politically expedient for her to be against it. After eight years of losing under Obama, they convinced themselves to take a chance on Trump.

I don't claim that any of these reasons are satisfying, but at least I can understand them. And hopefully understanding where the Trump support comes from can help craft solutions that don't involve voting for another president that could destroy this great republic (should she survive this one).

For progressives, get over yourselves. The elitism and condescending attitude only further alienates those you obviously don't understand. "Let them eat cake" didn't end well last time, either. Get involved in your community, but also get involved in a community that exists outside of your urban trappings. Recognize your solutions aren't THE solutions. There isn't a right answer in politics, so quit with the stupid "mic drop" commentary.

For those that voted for Trump, get over yourselves. As we learned from Brexit, be careful what you wish for. And if you find yourself justifying your vote for one of the above reasons... I hate to be the bearer of bad news, but Trump isn't going to make things better. I hope for the future that Trump's campaign was an elaborate con just to get elected president, and now that he is there, he will simply hire good people and get out of their way.

I keep trying to find a silver lining in all of this, but the best I can think of is perhaps we have seen how fragile our freedoms really are. Maybe by opening this festering wound, we can finally treat it - maybe in exposing our weakness, we can become stronger. Let us be diligent now, lest we destroy the things that have always made America great.

Sumner - Non-materialistic millennials and the Great Stagnation

I saw this today, and because I'm a millennial and a sucker for anything people say about millennials (which is probably a very millennial thing to do), I read with much interest. Right at the beginning, he makes a pretty interesting point that I hadn't realized. Sumner:

When I was a teenager in the early 1970s, I looked forward to getting a nice stereo system, with a receiver, amp, turntable and big set of speakers. My 17-year old daughter also likes music, but has no interest in those things. I have no idea why. And I notice this patter in lots of other areas. ... When I look at her lifestyle, it seems shockingly unmaterialistic. (Or maybe she has all these things embedded in her iPhone.)

What I thought was interesting is that when I was a teenager in the early aughts (is that the term we have settled on? because it sounds decidedly like something a millennial would never say), I also wanted a nice stereo system, with receiver, amp, speakers, but no turntable. I'm not a monster. I also wanted a lot of other things, like a nice Les Paul Classic electric guitar like Paul McCartney, a Mitsubishi 3000GT, and a sick Pentium 4 computer with an nVidia GeForce4 graphics card and a gig of ram. What is interesting is that as I got older, I found that a lot of the things that I wanted as a teenager weren't that important to me, e.g. a sports car (I have a car, it gets me from point A to B at about 35mpg with comfort and ease); or that with a full time job, the expense of those things were relatively small compared to my earnings (I recently added up all the material things I really want and found that it would probably only take me about 2 years to save up to splurge and get them all). Part of that is my tastes have changed significantly over the last decade, but perhaps a larger part is that the cost of these materialistic wants has declined substantially. The speakers, computers, guitars, TVs, and cars that I wanted then are either much cheaper now or have cheap substitutes that work just as well or better.

What I most want now is something entirely different than what I wanted as a teenager: time. Time to spend with my awesome wife and three amazing children. Time to actually listen to those speakers, play on that Les Paul, or boot up a game on my sick computer. When I'm working now, I actually think of income as a means of exchange for future time. Every dollar I save is buying back future time. This sounds a lot like slavery, and in some ways, I think that accurately describes how many millennials view their working lives. The rise of extreme early retirement and the gig economy are both ways that millennials are moving away from the traditional American Dream of a 9-5 job and a nice house with lots of stuff. Sumner talks about the rise of camping as another rejection of materialism. I think part of the rise of camping is actually a way for millennials to travel on a budget. Instead of paying $200/night at a hotel, pay $40 for a camping pass. Modern camping is usually a drive in affair with bathrooms and a hot shower nearby. Since technology travels so well, you can be as (un)connected as you want to be.

Popping in to say hello...

Slow month for posting, but busy month for me. My wife and I welcomed our third baby this month and I am finally settling back into my normal routine of husband/father/worker/student/blogger/programmer. At least, I think that it is a normal routine.

Much has happened over the last few weeks, but one of the most interesting things I saw was the presidential debate. While I don't much respect either candidate and thus didn't have high hopes for the debate, I was still surprised and disappointed at just how economically illiterate both candidates seem to be. Hillary was better than Donald, but that is like saying being punched is better than being stabbed. Both are painful and neither one is good for you. From what I heard, you would think the general consensus is that trade is ruining the world. If this kind of thinking gains traction, I genuinely fear for our future. Markets and trade have done more to lift people out of poverty than every charitable and humanitarian effort in history combined. No, I don't have research to back up that statement, but I don't know many serious people who would disagree with it either.

As for what I've been reading lately, I just saw a post this morning that I thought was interesting on Andrew Gelman's blog about air rage on planes with first class seating. The article talks about a poorly done statistical analysis of air rage incidents that managed to get published in a journal despite all the obvious issues. One part I really liked:

When posting on this study, I threw gratuitous shade at one of America’s most trusted news sources in my “tl;dr summary”:

NPR will love this paper. It directly targets their demographic of people who are rich enough to fly a lot but not rich enough to fly first class, and who think that inequality is the cause of the world’s ills.

The next day I posted a roundup of media outlets that’d fallen for this story, including CNN, the LA Times, and ABC News, along with respected tech sources Science and BoingBoing. I discussed the selection bias that occurs when the best science reporters realize this study is empty and don’t report it, while everyday journalists just follow the PPNAS label and don’t even think there could be a problem. All jokes about “stat rage” aside, this is a big problem in that consumers of the news only see the sucker takes, never the knowledge.

I think Andrew points out something here that doesn't get enough thought. News outlets report what they think will be interesting to their readers, but often don't have the technical knowledge about some of the subjects they cover to really know if what they are reporting is true. They rely on gatekeepers, such as field journals, to weed out good information from bad. So when these journals drop the ball and let shoddy work through, we end up with gross misinformation. It goes back to my concern with the presidential debate. When the two most visible candidates for the highest office in the US are spouting nonsense about trade, that matters. People will assume these candidates are smart, or at least have smart people advising them on these subjects, so what they say will heavily influence people.

Part of the reason I decided to write this blog, and why I agree with Miles Kimball that blogging is important, is to have another voice to contradict bad information and influence other thinkers. This also provides a forum for me to put ideas I have out in public, which makes me think about them more. An example is a post I have probably rewritten a half dozen times (or more) about the guaranteed basic income ("GBI"). Morgan Warstler had a post a while back that proposed an alternative to GBI that I thought was interesting. But whenever I have put together a post to comment on it, I realize I have more questions that I want to dive into before I post. That process continues to sharpen my own thinking on the subject and I feel more confident that when I do finally post something, I will at least have taken the time to think through many of the issues with such a policy change. From the Andrew Gelman piece, it seems like more journalists need to go through that process as well.