Facebook Bias, Bernie Sanders, and Ubers in Austin

Does Facebook Have a Liberal Bias?

I feel like this story is a lot less relevant than it sounds, but here is some background.

From CNN:

An anonymous former Facebook contractor says he witnessed colleagues suppress news about “popular conservative topics” from the website’s “trending” section.

Three former Facebook workers who spoke with CNNMoney echoed what some of Gizmodo’s other sources said — that personal biases might creep into the day-to-day “trending” work, but they never detected institutional bias for or against conservative information.

And here is the NYT:

Facebook denied the allegations after a backlash — from both conservative and liberal critics — erupted. “It is beyond disturbing to learn that this power is being used to silence viewpoints and stories that don’t fit someone else’s agenda,” read a statement from the Republican National Committee. “NOT LEANING IN… LEANING LEFT!” blared the top story on The Drudge Report, a widely read website.

The journalist Glenn Greenwald, hardly a conservative ally, weighed in on Twitter: “Aside from fueling right-wing persecution, this is a key reminder of dangers of Silicon Valley controlling content.” And Alexander Marlow, the editor in chief of Breitbart News, a conservative-leaning publication, said the report confirmed “what conservatives have long suspected.”

Intrigued yet? Here are some things to consider in defense of Facebook even if the allegations are true:

  1. Facebook is a company, not a government organization. It does not have an obligation to be politically balanced.
  2. As a company, Facebook’s main goal is to generate profits, and if it does so better by instilling a liberal bias, then so what? In the scale of moral quandaries, this is pretty benign compared to what other companies do.
  3. Facebook is a social media site, not a news agency. And neither has an obligation to be politically balanced.
  4. There is some argument that the news media already has a liberal bias. If so, how is Facebook’s liberal bias different from that of other forms of media?
  5. Liberals tend to post more political things on Facebook than conservatives. So even if nobody working at Facebook is tweaking knobs, you should see more liberal posts than conservative ones, and liberal posts should trend more often.

liberal-conservative-facebook

Facebook denies having any intentional bias. Here is the WSJ:

Facebook Inc. on Tuesday denied allegations from former workers who said the social media site suppressed news about conservative issues on its popular “trending” news feature.

Tom Stocky, Facebook’s vice president for search and the person responsible for the trending feature, defended the company’s practices and said it found “no evidence that the anonymous allegations are true.”

I wouldn’t be surprised either way, if there was or was no manipulation of trending news. Not that I’d notice—my Facebook newsfeed is nearly all liberal posts.

Also, I did warn that this topic is less relevant than it sounds. The best thing to do is recognize that there might be some liberal bias and move on.

Bernie Sanders and Efficient Markets

The headline for this section is not a joke, haha.

Some primary ago, Bernie Sanders won a state but his chance to win the Democratic nomination went down as a result. And on my social media feeds, people (esp. supporters of Bernie Sanders) didn’t seem to understand how this could happen. I also often heard things like “Sanders is projected to win the next 5 states, but the media is saying Clinton has practically won the nomination, so therefore the media is biased against Sanders.” This is of course nonsense. So here is my PSA to explain how markets work.

Test your understanding of markets by figuring out which statement is correct:

  1. Sanders is 5% to be the Democratic candidate according to prediction markets, but polls show him far ahead of Clinton in the next 5 states. Therefore, the real chance he has to win overall is something higher than 5%, maybe more like 10% or 20%.
  2. Sanders is 5% to be the Democratic candidate according to prediction markets, but polls show him far ahead of Clinton in the next 5 states. These polls don’t matter, and his chance to win overall is still 5%.

Naively, the first statement sounds better. But anyone who has caught on to the concept I am getting at would know that the first statement is wrong and that the second statement is correct. Why? The prediction markets already take into account the results of the polls. That is, the knowledge gained from the polls is already incorporated into the levels of the market. (This is, of course, assuming Efficient Market Hypothesis.)

You can apply the same concept to stocks. Let’s say Apple is $90 a share on the market, but you think it has a 10% chance to be bankrupt and worth $0. Thus, Apple is 90% to be worth $90 and 10% to be worth $0, for a fair value of $81 per share. [90%*$90 + 10%*$0 = $81.]

So should you sell Apple at $90? No. In fact, the calculation to get $81 is wrong.

Assuming you obtained your information legally, other people in the market, including those who have presumably spent far more time than you have researching the fundamentals of Apple, also think that Apple is 10% likely to be bankrupt. So here is what’s actually going on in the market:

  • People think Apple is worth $100 per share if it is not bankrupt.
  • It has a 10% chance to be bankrupt, in which case it is worth $0.
  • Thus, the fair price is 90%*$100 + 10%*$0 = $90.
  • Thus, it is trading at $90 in the market.

An interesting thing to consider is, what if Apple comes out on the news and releases some sales numbers, and now everyone thinks there is 5% chance Apple is bankrupt? The price actually increases, to $95 [95%*$100 + 5%*$0 = $95].

So people thinking Apple has a chance to be bankrupt can increase the price of the stock. The key is that the chance it is bankrupt is less than what everyone expected before.

Similarly, if prediction markets are already expecting Bernie Sanders to win a state with 60% of the vote but he wins with only 55%, his chance to become the nominee goes down.

You can see what prediction markets are saying about the 2016 race here.

Uber and Lyft in Austin

As a fan of free markets and as someone who grew up in Austin, TX, I am kind of sad to see Uber and Lyft shutdown in Austin over regulation. Here is TechCrunch:

Today voters in Austin went to the polls to weigh in on Proposition 1, an attempt to overturn a bill requiring mandatory fingerprint-based criminal background checks for new Uber and Lyft drivers in the city.

The results are in, and with 56 percent of total voters voting against Prop 1, the proposition failed to pass. This means that the bill requiring fingerprint-based background checks will proceed, with new drivers needing to pass the check before being able to drive.

In response to the news, Uber and Lyft have announced that they will be shutting down operations in the city — at least temporarily.

One idea of markets is that you wouldn’t do a transaction if you thought it was bad for you. So the fact that many people were using Uber in Austin before this meant that even without fingerprint checks, people preferred Uber to older cab services. And who benefits from this regulation? Older cab companies.

Of course, safety is good, but it seems ironic that a city that prides itself on being kept weird and being the new Silicon Valley, the tech capital of the south, is the one to introduce this kind of regulation that will just slow down tech development. The WSJ reports, “Austin is now the largest U.S. city where Uber isn’t currently available.”

It also seems like this is a weird political issue in that people who use Uber are generally young people, and young people are generally more liberal, but in this case they are more anti-regulation.

Here is Kristen S. Anderson, in one of the best passages from The Selfie Vote (Ch. 4):

Suffice to say, cabs hate Uber, as well as Lyft and other “ride-sharing” companies, as they came to be known. Ride-sharing companies are absolutely eating the cab industry’s lunch. Almost nothing about the existing taxicab system in most major cities resembles a free market, and the results are exactly what you’d expect. In a properly regulated free market, customers can make informed decisions about whether or not to engage someone in services for hire. They can vote with their wallets, choosing between a variety of competitors to patronize only quality vendors with good pricing, and those who offer poor service wind up failing.

The cab market in most cities is the opposite of this in nearly every way. When I’m standing on a street corner in a city, trying to hail a cab, I’m not choosing between quality drivers; I’m hopping in whatever cab pulls up and hoping for the best. I’ll pay the same rate whether the driver is kind  and funny and highly competent (as many drivers are!) or reeks of cigarette smoke and blasts the heat in July and drives like they’re running away from the cops. There’s no choice or competition, on price or quality. The lack of market forces makes the taxicab industry in many cities function like something out of the Soviet era: inadequate supply and mediocre quality.

Ride-sharing companies add competition, and there’s nothing an existing cartel hates quite like competition. In cities across the U.S., ride-sharing companies entered into battle with local regulators and taxicab unions, with the regulators and unions trying to keep the newcomers out of the market. In DC, for instance, cab protests have involved drivers swarming particular major thoroughfares and loudly honking their horns to draw attention to their plight (likely perturbing their intended audience in the process). But it wasn’t just free-market types and libertarians championing the cause of ride-sharing companies: I remember being amused at the many left-of-center writers I saw dropping their pro-union and pro-regulation posture when the unions and regulations were going to keep them from being able to get a cheap, quality ride home from work or happy hour (or perhaps Whole Foods!). The ride-sharing battle in many cities pitted entrenched interests against an upstart. Overwhelmingly, the young—moving to denser areas, eschewing cars when they can—were on the side of the upstart. And the battle over ride-sharing companies and regulations created a near picture-perfect example of the power of the oft-maligned free market to do great things, to encourage innovation, and to improve people’s quality of life.

Misc

MarketWatch on “Hillary Clinton reaping donations from Wall Street“:

The Democratic front-runner has raised $4.2 million in total from Wall Street, $344,000 of which was contributed in March alone. According to a Wall Street Journal analysis of fundraising data provided by the nonpartisan Center for Responsive Politics, the former secretary of state received 53% of the donations from Wall Street in March, up from 32% last year and 33% in January through February, as the nominating contests began.

Trump, by contrast, hasn’t garnered more than 1% of Wall Street contributions in any month through March.

The Atlantic on why empathy is bad:

Paul Bloom, psychologist and Yale professor, argues that empathy is a bad thing—that it makes the world worse. While we’ve been taught that putting yourself in another’s shoes cultivates compassion, it actually blinds you to the long-term consequences of your actions.

The Washington Post on being interrogated for doing math on a plane:

The curly-haired man laughed.

He laughed because those scribbles weren’t Arabic, or another foreign language, or even some special secret terrorist code. They were math.

Disclosure: I have done math on a plane before.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s