<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Viet Vu on Medium]]></title>
        <description><![CDATA[Stories by Viet Vu on Medium]]></description>
        <link>https://medium.com/@vuviet?source=rss-3c636f3dcd72------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*nPofQaoFzoiWglAaf9Sn0g.jpeg</url>
            <title>Stories by Viet Vu on Medium</title>
            <link>https://medium.com/@vuviet?source=rss-3c636f3dcd72------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 16:38:57 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@vuviet/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Politicizing Mass Shootings — When we can talk about Gun Control]]></title>
            <link>https://medium.com/data-science/politicizing-mass-shootings-when-we-can-talk-about-gun-control-f196ebee2b0c?source=rss-3c636f3dcd72------2</link>
            <guid isPermaLink="false">https://medium.com/p/f196ebee2b0c</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[guns]]></category>
            <category><![CDATA[towards-data-science]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[politics]]></category>
            <dc:creator><![CDATA[Viet Vu]]></dc:creator>
            <pubDate>Tue, 02 Jan 2018 18:39:44 GMT</pubDate>
            <atom:updated>2018-01-02T19:35:31.916Z</atom:updated>
            <content:encoded><![CDATA[<p>On October 1st, 2017, a gunman rained bullets from a hotel room in Las Vegas, <a href="https://en.wikipedia.org/wiki/2017_Las_Vegas_shooting">killing 59 and injuring 546</a>. It was the deadliest mass shooting in US’s recent history.</p><p>After every mass shooting, the narrative has become familiar, almost scripted. Politicians on both aisles pray for the victims and their families, while Democrats using this chance to draw attention to gun control, and Republicans denouncing Democrats for “politicizing a tragedy”</p><p>Let’s take that politicization argument at face value and say that it’s not okay to talk about gun control immediately following a mass shooting. When is it okay to talk about it then? Let’s say the following two criteria are used to determine whether it’s okay to talk about gun control or not:</p><ol><li>No mass shooting on that date (4 or more shot or killed, not including the shooter)</li><li>At least 3 days have passed since the previous mass shooting</li></ol><p>Here’s what I came up with using 2017 mass shooting data from the US:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5KKElpXwOCGx-z-oTdfBCw.png" /></figure><p>Let’s start with some inferences:</p><ol><li>Under this criteria, you get at most 2 consecutive days to talk about gun control and gun issues before another shooting takes place.</li><li>You can’t talk about gun control in January, June, August, September, and October.</li><li>Overall, we have 15 days when it’s okay to talk about gun control — just over 2 weeks.</li></ol><p>3 days, though an arbitrary number, isn’t too far off from the truth, given that <a href="http://www.foxbusiness.com/politics/2017/10/04/left-is-politicizing-las-vegas-massacre-varney.html">dozens</a> <a href="http://thehill.com/opinion/criminal-justice/353908-politicization-of-tragedy-is-the-last-thing-we-need-from-our-leaders">of</a> <a href="http://dailycaller.com/2017/10/04/john-rich-on-politicizing-vegas-massacre-most-disgusting-thing-ive-ever-heard/">opinion</a> <a href="http://www.chicagotribune.com/news/opinion/commentary/ct-perspec-las-vegas-democrats-politics-1005-story.html">pieces</a> were published on October 4th (3 days after the Las Vegas shooting) bashing Democrats for politicizing the tragedy.</p><p>So, maybe it’s not about wanting to prevent a tragedy from being politicized. Maybe it’s about not wanting to talk about gun control, and critically examining the Second Amendment to adopt policies to prevent such shootings from happening again.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f196ebee2b0c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/data-science/politicizing-mass-shootings-when-we-can-talk-about-gun-control-f196ebee2b0c">Politicizing Mass Shootings — When we can talk about Gun Control</a> was originally published in <a href="https://medium.com/data-science">TDS Archive</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Virtue Signals or News Reading? An exploration on why Senators follow each other on Twitter]]></title>
            <link>https://medium.com/data-science/virtue-signals-or-news-reading-an-exploration-on-why-senators-follow-each-others-on-twitter-f9805ab381c2?source=rss-3c636f3dcd72------2</link>
            <guid isPermaLink="false">https://medium.com/p/f9805ab381c2</guid>
            <category><![CDATA[data-journalism]]></category>
            <category><![CDATA[politics]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[senate]]></category>
            <dc:creator><![CDATA[Viet Vu]]></dc:creator>
            <pubDate>Tue, 13 Jun 2017 10:21:28 GMT</pubDate>
            <atom:updated>2017-06-14T16:53:08.067Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*INMexPgSx3m2XOnx3E62MA.png" /></figure><p>When you go on Donald Trump’s Twitter, one of the first things you notice is how few accounts he follows. As of June 13th 2017, that number stands at 45. A <a href="http://edition.cnn.com/2017/03/06/politics/trump-twitter-list-trnd/index.html">number</a> of <a href="https://www.aol.com/article/news/2017/04/12/every-account-president-donald-trump-follows-on-twitter/22037246/">news</a> <a href="http://mashable.com/2016/11/22/donald-trump-twitter-follows/">articles</a> have reported on the account he follows.</p><p>So I wondered, why is this the case? Why does he follow only a handful of Twitter accounts? But I realized quickly that I can’t really answer that question — I don’t know Donald Trump and I don’t know what goes on in his head. I decided to tackle a slightly easier question instead: why do US Senators follow each other on Twitter?</p><p>Following someone on Twitter is a relatively costless action. Therefore, one might think that there aren’t anything interesting going on here. However, when a Senator (say, Bernie Sanders) follows another Senator (say, Susan Collins) , two things change. Firstly, everyone can now observe the fact that a Senator follows another Senator. Maybe this can be considered as a signal that these two Senators like to work together. Secondly, Sen. Sanders’ newsfeed will now include Sen. Collins’ activities. This allows Sen. Sanders to gain valuable information about Sen. Collins’</p><p>The question this post explores is: on average, which effect dominates? Do Senators follow each other to get information, or to signal cooperation? Notice that if Senators follow each other to gain information, the value of a twitter follow as a cooperation signal decreases substantially because the receiving Senator expects they received the follow because the other Senator wants to monitor what they tweet.</p><h3><strong>The Twitter Network</strong></h3><p>In order to find the answer to this question, I needed to construct a dataset that maps out how Senators are connected on Twitter. I used Python and Twitter API to collect this information. I cleaned it up a little, then mapped it out using Gephi. The resulting network can be seen down here:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7l_EettsVbIL3ED9M2_mGA.png" /><figcaption>Senate Twitter network: Who follows who in US Senate</figcaption></figure><p>I created this representation using the Force Atlas algorithm. Intuitively nodes connected to each other are pulled together and nodes not connected to each other are pushed apart. I then imposed each Senator’s political party as colours on each nodes: Red for Republican, Blue for Democratic and Grey for Independents. A blue line means that the connection originates from a Democrat nodes. Red and Grey lines are similar for Republicans and Independents. Finally node size indicates how many Senate accounts they follow.</p><p>You can see a clear divide between the Democrats and the Republicans in terms of Twitter follows — most people tend to follow people within their own party. No single Senator disproportionately follows twitter accounts of the other party without also following accounts from their own party.</p><p>Moreover, the two independents: Angus King and Bernie Sanders behave relatively differently. Sen. Sanders look much more like a Democrat in this graph and Sen. King looks like a Centrist. Senators who are well known for bi-partisanship such as Sen. Collins[<a href="https://www.collins.senate.gov/about">1</a>] are also represented near the middle here.</p><p>At first glance, this looks great. The Twitter network does divide well according to party line and Senators who we know to be bi-partisan are also centrally located.</p><p>But does this hold generally? In order to test that question, we need a measure of how Senators work with each other. The central hypothesis is this: If on average, Senators follow each other to signal cooperation, then the network structure in Twitter would resemble the measure of cooperation. If Senators follow each other to gain news, then the network structure in twitter would not resemble the measure of cooperation. We’re not (yet) interested in causation here.</p><h3>Bill Co-Sponsorship</h3><p>What’s a good measure of cooperation? One of the most costless signals a Senator can give to signal cooperation is to co-sponsor bills. Bill co-sponsorship just requires you to sign on to the bill without a firm commitment that they’ll have to vote yes or even support the bill.</p><p>I went over to Congress’s website using python (and scrapy) to collect Sponsorship data on all the bill that has been considered by the Senate so far (over 1200 bills). I then used Gephi to graph this network, using a logic identical to the Twitter network:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zqiDblU702-3NmJhgjtx6g.png" /><figcaption>Senate Co-sponsorship network: Who co-sponsor with who in Senate</figcaption></figure><p>An additional note is needed on how I generated this graph. You might notice that some arrows are thicker than others. This is because of the way the co-sponsorship data is structured. The technical term is that we’re dealing with a bipartite network. Two Senators could have co-sponsored multiple legislations and the weight on each connection is simply the sum of legislations that these Senators have sponsored together.</p><p>We notice some of the same trends we saw earlier: we can see somewhat of a separation between Republicans and Democrats. Bernie Sanders is acting as a Democrat, Angus King is being a Centrist and Susan Collins is also appearing to work across party lines (even appearing more like a Democrat than a Republican.)</p><p>You also see other dynamics. Sen. McCaskill was positioned in the middle of the Democratic party in the Twitter graph but is now on the centre of the co-sponsorship network. We need to be able to quantify these two network structures more generally.</p><h3>Putting Them Together</h3><p>What I propose is to compare the node-specific centrality in each network. Centrality in graph theory is a measure of how central a node is — how easily you can reach any other nodes when you only use existing connections. There are many centrality measure used but for today, I’m going to focus on two measures: betweenness and closeness.</p><p>Closeness centrality measures how close a node is to other nodes. Intuitively, your closeness centrality score is high when you can reach other nodes using a shorter path.</p><p>Betweenness centrality measures how well placed a node is to connect other nodes. Intuitively, betweenness centrality score is high for you when you’re on a high number of paths that connect other nodes.</p><p>I calculate the Closeness and Betweenness of the two networks and compare them. If they’re correlated with each other, that would support our hypothesis that twitter follow is used more as a signal of cooperation. Otherwise, it is more likely that people follow twitter to get information.</p><p>I ran a simple regression here. Notice that I don’t particularly care too much about endogeneity (somewhat interpreted as “causation” though this interpretation is a bit problematic) because I’m really only concerned about whether there is some strategic aspect to twitter following that we can also see in how Senators co-sponsor bills. The result of this regression is shown here:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/607/1*SNMl3YOeevJgSKjVTIZCjw.png" /></figure><p>Focus on the closeness_cosponsor line. The “Estimate” column roughly tells me how closeness in the twitter network follows closeness in the cosponsor network. It’s not significant. Let’s try to look at the result for the betweenness centrality:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/622/1*PA05_TRU2X0p_SazqkDwEA.png" /></figure><p>Not even close either. What we can say from these two tables is that Centrality between the two network doesn’t resemble each other on average. On average, Senators tend to use twitter as a way to gather information as opposed to signal cooperation. That’s pretty neat.</p><h3>Caveats</h3><p>Unfortunately that’s not the entirety of this story. Some important assumptions and caveats need to be discussed.</p><p><strong>Assumption 1: </strong>Co-sponsorship is costless. One of the central reason why we can treat twitter follow with co-sponsorship of bill is because we think of both as costless. If for example a significant cost is introduced to co-sponsoring a bill, the decision surrounding how you co-sponsorship bill may be different from the way you use twitter to signal cooperation. This means that cooperation signalled in Twitter may not always translate to this more costly form of cooperation.</p><p><strong>Assumption 2: </strong>Party line is irrelevant. Another important assumption here is that Senators treat each other similarly regardless of party. This is probably the most unrealistic assumption used in this analysis. The decision to follow someone in the same party as you may look differently from the decision to follow someone in a different party than you. For example, the value of a twitter follow between two Democrats may be extremely low since by virtue of them being in the same party, they already have a strong cooperation signal. On the other hand, a decision by a Democrat to follow a Republican may act as a stronger signal. In future posts, I hope to explore this idea.</p><p><strong>Assumption 3: </strong>Bill co-sponsorship is “complete”. We’re barely 6 months into the current Congress. Perhaps the cooperation represented by bill-cosponsorship has not been able to present itself fully. This can be solved relatively easily by looking at this analysis again using a more comprehensive set of legislations in the future.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f9805ab381c2" width="1" height="1" alt=""><hr><p><a href="https://medium.com/data-science/virtue-signals-or-news-reading-an-exploration-on-why-senators-follow-each-others-on-twitter-f9805ab381c2">Virtue Signals or News Reading? An exploration on why Senators follow each other on Twitter</a> was originally published in <a href="https://medium.com/data-science">TDS Archive</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Does same sex marriage reduce childbirth? No no and no.]]></title>
            <link>https://medium.com/@vuviet/does-same-sex-marriage-reduce-childbirth-no-no-and-no-c4b75e14fa88?source=rss-3c636f3dcd72------2</link>
            <guid isPermaLink="false">https://medium.com/p/c4b75e14fa88</guid>
            <dc:creator><![CDATA[Viet Vu]]></dc:creator>
            <pubDate>Fri, 19 Dec 2014 23:25:00 GMT</pubDate>
            <atom:updated>2014-12-19T23:31:05.287Z</atom:updated>
            <content:encoded><![CDATA[<h2>Does same sex marriage reduce childbirth? No.</h2><p>One argument we often hear from proponents of same sex marriage is that legalizing same sex marriage destroys the “fundamental” reason why people marry — to have kids. Proponents argue that same sex marriage makes people less likely to marry each other and as a result, reduce childbirth (and increase out-of-wedlock childbirth.)</p><p>Is this true?</p><p>Our experience in Netherlands show that this is not the case:</p><p>Netherlands legalized civil union in 1990 and same sex marriage in 2000, becoming the first country in the world to do so. As same sex marriage has been in place for the past decade (two if you count civil union) and any effect that same sex marriage would have would have been experienced by the Dutch.</p><p>But it hasn’t been the case:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tCZP6h1bl8rfj-OPnEeIaQ.png" /><figcaption>Birthrate per 1000 in the Neatherlands</figcaption></figure><p>Here, you can see that birthrate has already fallen quite dramatically between 1960 and 1980 — before this same sex marriage thing became an issue.</p><p>Using <a href="http://equaldex.com/">Equaldex</a>, we know that Netherlands introduced civil union in 1990. So if the proponent is true, the birthrate will have a higher drop after 1990 than before. Graphically this is not the case and analytically this wasn’t the case either. A statistical correlation could not be detected and even if the correlation was there, the difference was so small that it would not have registered as being significant.</p><p>Let’s put this debate to rest.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c4b75e14fa88" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[One graph that explains why bitcoin is nowhere close to real “money”]]></title>
            <link>https://medium.com/@vuviet/one-graph-that-explains-why-bitcoin-is-nowhere-close-to-real-money-a790e3bb289e?source=rss-3c636f3dcd72------2</link>
            <guid isPermaLink="false">https://medium.com/p/a790e3bb289e</guid>
            <dc:creator><![CDATA[Viet Vu]]></dc:creator>
            <pubDate>Tue, 16 Dec 2014 22:59:09 GMT</pubDate>
            <atom:updated>2014-12-16T23:02:18.601Z</atom:updated>
            <content:encoded><![CDATA[<h2>One graph that explains why bitcoin is nowhere close to “real money”</h2><h4>1 bitcoin can’t buy you the same jug of milk everyday.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BAee29ZlpQdJeYfl7mq3cQ.png" /></figure><p>Y axis on this graph is percentages. Specifically percentages deviation of daily closing prices from the mean closing price of a 1 year period (from December 12 2013 to December 12 2014.) The green line represents Bitcoin to USD exchange rate and the blue line represents Canadian dollars to USD exchange rate.</p><p>The CAD/USD line is “shorter” than the bitcoin line because bitcoin trading happens every single day without fail. This is not the case for CAD/USD. CAD/USD trading only happens when banks are open. Therefore, there are 252 data points used for CAD/USD whereas there are 366 data points used for Bitcoin/USD. This doesn’t affect the overall trend that one can observe in this graph.</p><p>Percentage deviation is used to normalize the magnitude. One thing is clear from this graph: Bitcoin is <strong>extremely unstable.</strong> To the extent that you have no way of being able to buy the same thing with the same amount of bitcoin day to day.</p><p>And that is why bit coin is not yet a form of currency. It is still a form of asset and will be for quite some time to come.</p><p><strong>Data sources:</strong></p><p>CAD/USD: FRED2</p><p>Bitcoin/USD: Bitcoin Price Index</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a790e3bb289e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Imagine that you’re a teenager living in North America. You get a job in an umbrella store and you…]]></title>
            <link>https://medium.com/@vuviet/imagine-that-youre-a-teenager-living-in-north-america-you-get-a-job-in-an-umbrella-store-and-you-ae213ac91914?source=rss-3c636f3dcd72------2</link>
            <guid isPermaLink="false">https://medium.com/p/ae213ac91914</guid>
            <dc:creator><![CDATA[Viet Vu]]></dc:creator>
            <pubDate>Thu, 27 Nov 2014 23:42:48 GMT</pubDate>
            <atom:updated>2014-11-27T23:44:08.726Z</atom:updated>
            <content:encoded><![CDATA[<h2>Executive Compensation: an evil capitalist or an out of control teenager?</h2><h4>The pay-for-luck principle. And how we can go around solving this mess.</h4><p>You work at a gas station. One day, international oil prices go up and you increase your station’s gas prices accordingly. On that day, your station shows a record sales of gas and petroleum. For some reason, you think that it is because of you that the gas station managed to do so well. So you convince your boss that you should get paid more. The boss agrees and does pay you more for that day.</p><p>Sounds ridiculous? Except according to economists, that’s how executives at top corporations are compensated. This phenomenon was first described in detail by Bertrand and Mullainathan in a 2001 study. Since then, a growing body of literature has been adding to their original analysis, discovering mounting evidence that executives may not “deserve” all of their pay.<strong> </strong>To a layman, this idea was only popularized in 2014 by Thomas Piketty, author of <em>Capital in the 21st century.</em></p><p>In the book, Piketty argues that the declining top income tax rate is to blame for this curious phenomenon. He argues that as lower tax rate allows the CEO to have more financial power, they’re more able to influence the board of directors to increase their own pay. He also argues that because top-income tax rate is to blame, it must rise again so that we can discipline these CEO into performing optimally. I’m doubtful. I’m doubtful that Piketty was able to objectively analyze this issue as he was entrenched in his ulterior motive to advocating for a global wealth tax.</p><p>Through this piece, I hope to paint another picture, identifying the real cause of the rise of this “pay-for-luck” phenomenon and hopefully bring to light an alternative approach to limit this unfair compensation method. (spoiler: it’s not increasing top income tax.)</p><h3>The rise of executive compensation: are declining top-rate taxes to blame for?</h3><p>The rise in executive compensation is a relatively new phenomenon. Up until the end of 1970, executive compensation was stable. It started increasing dramatically since the beginning of the 1980s in the United States.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/764/1*CVAnoiw-hX7k23ENPRmjbA.png" /><figcaption>Frydman and Saks (2010)</figcaption></figure><p>Piketty, in his grand work, argues that executive compensation increases coincided with the decreases in the top income tax rate in the United States. As many will remember, this was the era of Reaganomics, who advocated for a stronger emphasis on supply-side policies which Piketty is not very fond of.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/729/1*KudtJK4_soTeV7inR5HWHQ.png" /><figcaption>Piketty 2014</figcaption></figure><p>His argument is as follows: when the top rate marginal tax rate was high, executives didn’t see a reason to fight for higher wage increases (as their realized benefits will not be significant.) However, as the top rate marginal tax fell, these same executives found it worthwhile to convince the board for higher pay (as they can get more out of each dollar increased.) This, added with how the board of director is chosen, contributed to the recorded increase in executive compensation.</p><p>The argument that Piketty puts forward, at first glance, seems convincing. However, correlation does not simply imply causation. Other factors may play a role in explaining this change in executive compensation.</p><p>Let’s stop here for a moment and think about why this correlation may not imply causation. Even though a sizable number of the top income bracket are executives, they only account for around 30–40% of the top 0.1% of the income distribution in some estimates.(Bakija et. al. 2012) This means that if tax decreases cause the rise in executive compensation, we should see similar increases in other professions’ income, especially those who have bargaining power over their pay.</p><p>In the same paper, Bakija et. al. show that between 1979 and 2005, the increase in compensation for executives and managers were responsible for 70% of the rise in compensation. Despite also being in that group of experiencing lower taxation, other professional groups (such as doctors who do have many bargaining power due to low supply) do not experience substantial income increases. This analysis breaks down Piketty’s argument that declining top-income tax rate is to blame for this explosion in executive pay.</p><p>Lemieux et. al. further explores this idea in their 2009 research and reached the same conclusion that much of the increase in wage inequality since late 1970s, especially in the top 20% income group, can be attributed to pay for performance.</p><p>This anecdote is backed up by real research. Frydman and Sacs, in their 2010 paper showed that even though the period of the change in executive compensation does coincide with the change in tax rate, the movement cannot be explained alone by the changes in tax rate. In fact, similar tax decreases occurred in Japan in the 1980s without having any significant effect on executive compensations where pay for performance was less pronounced. The following graph illustrates the difference between pre-tax and after-tax compensation.Although the size of the gap follows the tax changes, is not pronounced enough to explain all of the increases in compensation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/756/1*aFYMRRKEFI3O1b8UFuKAlw.png" /><figcaption>Frydman and Saks (2010)</figcaption></figure><h3>Pay for Performance is the real cause of the explosion of executive compensation</h3><p>Instead, Saks and Frydman attributes the rise in executive compensation to the prevalence of pay-for-performance in recent decades. Pay-for-performance is a philosophy of compensation where employees get paid according to how they perform. This philosophy was developed to address an economic problem called the Principal-Agent Problem.</p><p>Principal-Agent Problem was developed in the 1970s and 1980s, most famous with the 1983 paper by Grossman and Hart. It states that when the principal (the shareholders) have a different goal (maximizing firm profit) than the agent (the executive who wants to maximize sales and firm growth), suboptimal performance will follow. Therefore, pay-for-performance was developed as a way to align these goals, making sure that the CEO is trying to achieve the same goal as the shareholders’.</p><p>To date, the most popular form of performance based pay for executives is stock options. This form of compensation became popular starting in the 1950s. However, the full effect of stock-option compensation only took hold after the 1970s, when option grants caught up with option holdings. Figure one above illustrates this point as well, showing the wide gap between total compensation and salary+bonus starting in 1980s.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/754/1*nUq_VFdSSaqbgtAc8lgfXg.png" /><figcaption>Frydman and Saks (2010)</figcaption></figure><p>Now, this increase in executive compensation would be all fine and dandy (at least to an economist) if they were being compensated for their performance. In other words, if the pay increase can be attributed to all the extra value that these executives add to a company and to the economy, it would be considered “fair”. Unfortunately, as with the story of the gas station, economists have consistently found that these executives are being compensated for factors that are outside their control.</p><h3>The dark side of performance based compensation</h3><p>Bertrand and Mulainathan took a sample of petroleum companies’ executive compensations and benchmarked it against oil price changes. We know that oil company’s CEO does not have any control over changes in oil prices (as that is determined by politics and the market) so we shouldn’t see executive compensation to differ greatly as oil prices change.</p><p>Interestingly enough, every time there is an increase in oil prices, you observe an increase in executive compensation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/810/1*o937YTVVsq3T1BRvEtVOxg.png" /><figcaption>Bertrand and Mullainathan (2001)</figcaption></figure><p>This may be justified if the executive compensation goes down when oil prices go down. However, in the 10 years that oil prices did go down, executive compensations only went down in 6,with only 3 out of those 6 appearing to have any significant decrease.</p><p>If this visual example doesn’t persuade you (and it shouldn’t), Bertrand and Mullainathan performed several other tests with other industries and found the same effect.</p><p>Despite Bertrand’s result that the pay for luck effect appears in all forms of compensations (salary + bonus, stock option grants…), the most significant increase is still the option grants and long term pay. Salary and bonus pay for luck has existed since before the tax decrease.</p><p>Think of a high schooler (or even undergraduate students) who get to grade their own paper and have that grade appear on their transcript. Now apply that same image, while replacing the student with a CEO and the grade as their compensation package, especially those based on peer reviews.</p><h3>Tax the rich. Tax the rich. Tax the rich. Wait… really?</h3><p>Piketty, in his book, argues that the only solution to curve this “unjustified” increase in executive compensation is to increase the progressivity of income taxation. He argues for a return into the era where top-income brackets are taxed at the level of 80–90%, before Ronald Reagan and Margaret Thatcher took over.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*DQOqWWAGjnL1pn6g6o3qWQ.jpeg" /></figure><p>This is not an optimal solution. In the sense that Piketty’s solution would only work if executive compensation was caused by the decrease in top-income tax decreases. It wasn’t. It was a result of a shift in culture of how executive incentives should be constructed and how they should be compensated. Focusing on the tax rate misdirects extremely needed political capital to the wrong solution. Furthermore, increasing the taxes of the top income population also unfairly punishes those who are not executives (or the remianing 50–60% of the top 0.1% income group.)</p><p>Furthermore, by increasing taxation, total revenue that could be generated by the government may also take a major hit and increase the prevalence of tax avoidance (whether through tax havens or creative compensation methods.) This will hurt the quality of many public services.</p><p>We should instead, focus on reforming the corporate governance structure. We already have evidence that it works. In 2002, United States passed the Sarbanes-Oxley Act that made it mandatory for all publicly listed firms’ board of directors to have a majority independent directors, diminishing the power of the executive to set their own pay. Research from the Bank of Canada in 2008 shows that this strategy has been successful. Firms where pay-for-luck was prevalent before this regulation showed a decrease in their prevalence of pay-for-luck while firms where pay-for-luck was not prevalent maintained their business as status quo.</p><p>In fact, we don’t even have to look anywhere other than the original work done by Bertrand and Millainathan for further evidence. They show that by having one significant independent director, pay for luck decreases by 26 percent. This is a result of this independent director caring about firm performance and act as a “parent” to the executive.</p><p>As many others have demonstrated, executive compensation is only but a small piece of the puzzle in the universe of inequality. However, these executives are being over-compensated and not entirely for their performance. This is simply inefficient and is not fair. Implementing a blanket increase in top-income tax rate is inefficient and have the potential to damage many other groups whom this increase was not intended to damage. I propose that we focus more on the governance side. We need stronger and more independent directors to make sure executives’ performance are properly evaluated.</p><p>Who knows, with time, these out-of-control teenagers might finally grow up to be an adult. And if they don’t, unlike parents who do die, the government will outlive all parents, providing the necessary guidance for these eternal children of capitalism.</p><h3>References</h3><p>[<a href="http://qje.oxfordjournals.org/content/116/3/901.full.pdf">1</a>]Betrand, Marianne and Sendhill Mullainathan, “Are CEOs Rewarded For Luck? The Ones Without Principles Are.”, Quarterly Journal of Economics, August 2001, 901–932</p><p>[<a href="http://web.mit.edu/frydman/www/trends_rfs2010.pdf">2</a>]Frydman, Carol and Raven E. Saks, “Executive Compensations: A New View From a Long Term Perspective, 1936–2005&quot;, Review of Financial Studies, 2010</p><p>[<a href="http://www.bankofcanada.ca/wp-content/uploads/2010/02/wp08-20.pdf">3</a>]Paligorova, Teodora, “The Effect of the Sarbanes-Oxley Act on CEO Pay for Luck”, Bank of Canada, 2008, Working Paper.</p><p>[<a href="http://en.wikipedia.org/wiki/Capital_in_the_Twenty-First_Century">4</a>] Piketty, Thomas, “Capital in the 21st Century”, Harvard University Press, 2014.</p><p>[<a href="http://brousseau.info/pdf/cours/grossman_hart_83.pdf">5</a>]Grossman, Sanford J. and Oliver D. Hart, “An Analysis of Principle-Agent Problem”, Econometrica(51)(1), 1983, 7–46</p><p>[<a href="https://web.williams.edu/Economics/wp/BakijaColeHeimJobsIncomeGrowthTopEarners.pdf">6</a>]Bakija, John, Adam Cole, and Bradley T. Heim, “Jobs and Income Growth of Top Earners and the Causes of Changing Income Inequality: Evidence From US Tax Return Data.”, 2012.</p><p>[<a href="http://www.economics.ubc.ca/files/2013/05/pdf_paper_thomas-lemieux-performance-pay-wage-inequality.pdf">7</a>]Lemieux, Thomas, W. Bentley Macleod, and Daniel Parent, “Performance Pay and Wage Inequality”, Quarterly Journal of Economics(124)(1), 2009.</p><p>[<a href="http://www.law.harvard.edu/programs/olin_center/papers/pdf/Bebchuk_et%20al_510.pdf">8</a>] Bebchuk, Lucian, and Yaniv Grinstein, “The Growth of Executive Pay”, Discussion Paper, Harvard Law School, 2005.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ae213ac91914" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[6 lessons I learnt about communication and life from speaking at TEDxTT]]></title>
            <link>https://medium.com/tedx-experience/6-lessons-i-learnt-about-communication-and-life-from-speaking-at-tedxtt-88bc646956c7?source=rss-3c636f3dcd72------2</link>
            <guid isPermaLink="false">https://medium.com/p/88bc646956c7</guid>
            <dc:creator><![CDATA[Viet Vu]]></dc:creator>
            <pubDate>Sun, 09 Nov 2014 21:05:04 GMT</pubDate>
            <atom:updated>2015-06-10T13:37:16.907Z</atom:updated>
            <content:encoded><![CDATA[<h4>And I only learnt one of them during my 15 minutes talk.</h4><p><em>photo by Steven Durfee</em></p><p>On September 29th, 2014. 9:02pm, I received the following email:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/748/1*yNumzQ_znZhg475fF7GGDA.jpeg" /></figure><p>This email confirmed that I was going to be a speaker for this year’s TEDx Terry Talks (TEDxTT) — <a href="http://www.terry.ubc.ca/tedxterrytalks/">unofficially the longest running TEDx event in the world</a>.</p><p>The video is now available here: <a href="https://www.youtube.com/watch?v=XMbfDV55kmc">https://www.youtube.com/watch?v=XMbfDV55kmc</a></p><p>Needless to say, I was ecstatic. I remember not being able to sleep that night, talking long walks and thinking about what this means.</p><p>I officially spoke on October 25th, 2014. And it has been 2 weeks now since that has happened. Two weeks have passed since then, and I still haven’t been able to figure out exactly what I learnt through this journey.</p><p>And so I’m writing this to make sense of what I ended up learning speaking in front of a shit load of people about coming out.</p><h3>Lesson 1 — If you have an idea, share them.</h3><p>When I got the email about speaking in TEDxTT, I did not know exactly what I was talking about. I knew that it was about bisexuality, I knew that it was about coming out, I knew that it was about the work that I was doing with Equaldex but I had no idea how to convey any of those ideas.</p><p>And yet the curators invited me to speak at the event.</p><p>To this date, I didn’t know why they chose me exactly amongst the applications they received. My video was 41 seconds long and didn’t really have much to it(This is the first time this video is made public and oh god was I awkward):</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FhimS4RgdnIo%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DhimS4RgdnIo&amp;image=http%3A%2F%2Fi.ytimg.com%2Fvi%2FhimS4RgdnIo%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="640" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/be73889ad3aa5c2b91269995a04cae59/href">https://medium.com/media/be73889ad3aa5c2b91269995a04cae59/href</a></iframe><p>But the acceptance email read: “we wanted to hear more about your topic”</p><p>Too often we don’t talk about the ideas that we have because we think the idea is not worth talking about.</p><p>Stop. Share that idea. I want to hear it and your friends want to hear it.</p><p>Because an idea stays an idea until you share it. Once you have shared it, it stops being an idea. It becomes a goal, a movement, a project and things that are much much more.</p><p>If you have an idea, share them.</p><h3>Lesson 2 — Know who you’re talking to, and why they care.</h3><p>The biggest struggle I had with my TEDxTT talk was on what I was trying to accomplish with it.</p><p>During my coming out, my parents threatened to disown me and I know that this is an experience I wouldn’t wish upon my worst enemy. And I wanted to craft my talk to make sure that doesn’t happen to anyone.</p><p>But it was an experience that not a lot of people have experienced before.</p><p>5 minutes after I told my best friend about the TEDxTT and the message I was trying to convey, he made a comment: “Make sure the message is also translatable to non-LGBTQ+ people in the audience so that we all know what we can do to help :) )</p><p>I realized that the majority of my audience is going to be straight and has never gone through the “coming out” experience. If my talk is just going to be about how I came out, and my struggle through this experience, it won’t be able to resonate with anyone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*SusNTpuw4M71JqIQp9qlow.jpeg" /><figcaption>Photo: Sruthi Sreedhar</figcaption></figure><p>So I flipped the message and included the audience in my talk. I talked about how my friends, most of whom are straight, helped me through the experience of coming out. I made the story more relatable and tried to help the audience realize the power that they possessed.</p><p>I focused on how the audience made a difference rather than focusing solely on what difference they made and what difference they can make.</p><p>And it worked. I received a standing ovation.</p><h3>Lesson 3 — Your first draft is shit. Let people tell you how shit it is.</h3><p>One week after our first speaker meeting, the first rehearsal was scheduled. I worked on it for a few days, thinking it will be amazing and I will be one of the most prepared speaker ever.</p><p>Oh I was so wrong.</p><p>My talk was shit. It had way too many slides, it was confused, it was too fast, my pacing was that of a bullet train… in short, it was bad.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mQPtnLvw7Q9oVwMLKnky-A.jpeg" /><figcaption>What I did to prepare for my talk</figcaption></figure><p>So I revised it. Multiple times. The version that ended up being on the stage is the third major revision of the talk with numerous little details edited throughout the two weeks I had after the first rehearsal.</p><p>If I didn’t let those people to tell me how bad my talk was, it would have been bad on the day of.</p><p>The first edition of anything that anyone makes is painfully bad. Accept that fact and move on. There are people around you who will help you realize how bad it is and how you can improve it. Embrace them and make a unicorn out of your shit.</p><h3>Lesson 4 — You’re a representative. Represent well.</h3><p>During the coffee break after my talk, 2 people came up to me and said that they were both bi and they appreciated what I had to say.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BHlMHH_ch__vcA2vDakjbQ.jpeg" /></figure><p>Those were two of the best compliments I have ever received in my life.</p><p>I never realized it, but once I was on the stage, intentional or not, I was the ambassador of the bi community. Without knowing it, I was talking not only for me but for a group of people. That’s scary and yet humbling somehow.</p><p>In an ideal world, you will just represent you, or, as any TED(or TEDx) curators will tell you: “The views expressed by the speakers are their own and do not reflect the views held by TED, TEDx or organizations that the speakers represent.” Despite all these efforts to isolate the individual from a group, I was still a member of that group, talking about that group and people associated me with that group.</p><p>Represent them well. You’ve been given a platform and you should strive to use that platform well. It’s a big responsibility but you can do it.</p><h3>Lesson 5 — When you’re vulnerable, you forget irrelevant things. And it’s nice.</h3><p>When I have to recount the 15 minutes I spent up on that stage, it would have to be this song:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fn1Q08xO7Zsg%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dn1Q08xO7Zsg&amp;image=http%3A%2F%2Fi.ytimg.com%2Fvi%2Fn1Q08xO7Zsg%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="640" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/841ceda8bddd22123d5699a0747ca731/href">https://medium.com/media/841ceda8bddd22123d5699a0747ca731/href</a></iframe><p>Once I was up there, I forgot everything. I took off my glasses so I couldn’t see the faces. And I talked. I made my butt joke(yes there was one), my experience coming out and my call to action so everyone share their happiness, be empathetic and be vulnerable with each other.</p><p>It was surreal but yet really nice. I didn’t worry but I wasn’t excited either. All there was was what I had to talk, the next slide and the emotion I felt as I said those words.</p><p>On that stage, I was so completely vulnerable. And it felt nice.</p><h3>Lesson 6— Appreciate everyone who allowed you to be here.</h3><p>Without my friends, I would not be here, today, alive.</p><p>For that, I am grateful.</p><p>If you were ever doing something, the same thing applies. You wouldn’t not be where you are without those who are around you who helped you get there.</p><p>Appreciate them and thank them. They really did deserve it.</p><p>So I want to thank the following people (I’m only going to mention their first names for anonymity) who have helped shape my talk, practiced with me, organized the event and allowed me to give the talk of my life.</p><p>Logan | Gary | Michelle | Lina | Rob | Kamil | Charlotte | Matt | Aliya | Ekat | Frances | Dave | Joanne | Sam | Gordon | Urooba | Alex | Abeer | Serbulent | Joel | Swinzle | Dan | Durfee | Saarah | Kathleen | Carol | Josh | Peter | Sarah | Hailey | Milkshake</p><p>and many others who I can’t list here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*nUXtTvdl-fvhztVMvhx09Q.jpeg" /><figcaption>Photo: Sruthi Sreedhar</figcaption></figure><p>For that, I am grateful</p><h3>Bonus lesson — you don’t become a super star after your TEDx talk; mind your ego.</h3><p>As an example, I am still very fond of butts. Nothing really changes after your talk. And despite this being a bonus lesson, this should be <strong>the lesson.</strong> A TED talk is not the end. <strong>It’s a mean to spread your ideas to a lot of people.</strong></p><p>So here’s hoping for the next few months/years to be the best time of my life.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=88bc646956c7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tedx-experience/6-lessons-i-learnt-about-communication-and-life-from-speaking-at-tedxtt-88bc646956c7">6 lessons I learnt about communication and life from speaking at TEDxTT</a> was originally published in <a href="https://medium.com/tedx-experience">TEDx Experience</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>