Saturday, April 22, 2017

Lies, Damn Lies, and Phony Statistics Graphics

The graphic below showing the changing distribution of US income has making the rounds for over a year, including in this story in the FT. It sure looks scary, doesn't it? That big mountain of super rich people has just been growing and growing since 1971, while the middle class has been getting squashed down. Boy oh boy has the US become polarized! Yikes! What inequality!

The thing about the graphic is that it's complete nonsense. If you look at what's been happening. the income distribution has been stretching out as nearly every low-income group (except the very very poorest) has become smaller, while higher-income groups have become larger. But at what cost? If this graphic is to be believed, it's that we've let the benefits of growth become so skewed that we've created a class of super-rich who have become distinct from the rest of us. But it's nothing more that a graphical trick. By truncating the x-axis at $200k and collecting together all those above that income level into one group, this apparent growing upper class will happen even if there is no change in inequality. 

To illustrate this point, I've created a simple graphic that does exactly the same as the one above. The blue bars are represent the income distribution in year 1, and the red bars are the distribution in year 2. Median income rises, but there's disproportionate growth in the group of super-rich folks ($110k and above in the example). Awful, isn't it?

It turns out that the data underlying the above figure is nothing but good news. The figure below removes the arbitrary top group and displays the whole income distribution. Every group from the lowest to the median sees a decrease in the number of people, while all higher groups see smaller increases. Is this change in the income distribution bad? If you think it is, then you're nuts. But it's exactly the same as the one right above, which looked so evil.

Saturday, February 11, 2017

Silly and ignorant criticisms of economics

Economics, as does any academic discipline, has its share of problems. Unfortunately, most of the criticism one hears in popular publications or other non-academic places are based on a near complete ignorance of what economists do. A few years ago, Chris Auld came up with a handy list of the most common and dumbest criticisms. I've reproduced them here because they now reside on a site that has appropriated his name, and I want to send people to them.
Every mainstream science which touches on political or religious ideology attracts more than its fair share of deniers: the anti-vaccine crowd v mainstream medicine, GMO fearmongers v geneticists, creationists v biologists, global warming deniers v climatologists. Economics is no different, but economics cranks differ in that they typically make false claims about the content of economics itself, as opposed, or as a prelude, to false claims about the way the world works. That target sometimes making it hard for non-economists to differentiate crankery from solid criticism.
Here, then, are some symptoms of bad critiques of economics:
  1. Treats macroeconomic forecasting as the major or only goal of economic analysis.
  2. Frames critique in terms of politics, most commonly the claim that economists are market fundamentalists.
  3. You believe that “gambling” is a bad word. Not knowing the difference between risk vs. reward and therefore never depositing at the online casino of life. If you don’t play, you can never win.
  4. Uses “neoclassical” as if it refers to a political philosophy, set of policy prescriptions, or actual economies. Bonus: spells it “neo-classical” or “Neo-classical.”
  5. Refers to “the” neoclassical model or otherwise suggests all of economic thought is contained in Walras (1874).
  6. Uses “neoclassical economics” and “mainstream economics” interchangeably. Bonus: uses “neoliberal economics” interchangeably with either.
  7. Uses the word “neoliberal” for any reason.
  8. Refers to “corporate masters” or otherwise implies economists are shills for the wealthy or corporations.
  9. Claims economists think people are always rational.
  10. Claims financial crisis disproved mainstream economics.
  11. Explicitly claims that economics is not empirical, or does so implicitly by ignoring empirical economics.
  12. Treats all of economics as if it’s battling schools of macroeconomics.
  13. Misconstrues jargon: “rational.”
  14. Misconstrues jargon: “efficient” (financial sense) or “efficient” (Pareto sense).
  15. Misconstrues jargon: “externality“.
  16. Claims economists only care about money.
  17. Claims economists ignore the environment. Variant: claims economics falters on point that “infinite growth on a finite planet is impossible.”
  18. Goes out of its way to point out that the Economics Nobel is not a real Nobel."

Friday, October 14, 2016

Voters, Ranked

1. (tie) Both Clinton and Trump are awful, but I'll hold my nose and vote for the one I think will do the least damage to the republic.

1. (tie) Both Clinton and Trump are awful, so I will vote for a third-party or independent candidate.

1. (tie) Both Clinton and Trump are awful and I will not be a party to this monstrosity.

4. All others.

Saturday, September 10, 2016

Election 2016 in a Nutshell

As of today, I am undecided between SMOD and Cthulhu as my choice for President of the United States. Both are running very convincing campaigns, but I'm currently leaning SMOD because I think he'll administer the end with the level of pain and suffering that we deserve. I remain persuadable, however.

The arguments for SMOD and Cthulhu are obvious, so you can look at their twitter feeds linked above for information to help you decide between them. My summaries of the candidates offered by the four largest parties are outlined below.
  • The Republicans have put forth Donald Trump, an emotionally and intellectually insecure man-child with a sexual fascination for his daughter and a homoerotic fascination for men who wield power. As a person, he is a huge POS.
  • The Democrats have as their candidate Hillary Clinton, a congenital liar who ran the State Department as an international influence-peddling scheme, using her husband's charitable foundation to launder the proceeds. To cover the tracks of this operation, she used private e-mail servers that exposed classified information to hackers. She's managed to stay ahead of the FBI by a combination of obstruction, lying, destruction of evidence, and a deeply partisan Department of Justice.
  • The Libertarian Party, finally seeing its opportunity on the big stage, nominated Gary Johnson, former governor of New Mexico. He turns out to be a semi-retired stoner who, surprisingly for a self-proclaimed libertarian, is at best lukewarm about individual rights and the rule of law.
  • Finally, the Green Party, looking to tap into the millions of voters energized by failed Democratic candidate and noted socialist Bernie Sanders, nominated Dr. Jill Stein, who is still mourning the death of Harambe.

Thursday, September 8, 2016

Economics Research Rankings for Missouri

RePEc (Research Papers in Economics) is an extremely valuable resource for economists. Produced by volunteers from around the world, it's a database of working papers, journal articles, books, etc. that's become the go-to place for everything about economics research past and present.

It was inevitable that this data would be used to create rankings of journals, departments, and even individual economists. These rankings have been sliced and diced by field, country, region, state, type of item, sex, graduation cohort, and even deceasedness. Below is the current ranking of university economics departments in Missouri:

RankDepartment, UniversityScore
1Dept of Economics, Washington Univ in St. Louis3.19
2Olin School of Business, Washington Univ in St. Louis4.16
3Economics Department, Univ of Missouri5.05
4Center for Economics & the Environment, Lindenwood Univ7.92
5Dept of Economics, Univ of Missouri-St. Louis10.92
6John Cook School of Business, St. Louis Univ13.72

Note that I have excluded the state's two Feds in Missouri because of the difficulty in figuring out who actually works in which division and because their lists of people include some who are just visiting scholars. Thus, their RePEc rankings are not terribly informative. Given that the two Feds account for a significant portion of the economics research done in the state, this is unfortunate.

In addition, RePEc's overall ranking methodology suffers from a severe case of information kitchen-sinkism. That is, the department rankings are determined by the harmonic average of 33 rankings, some of which are of questionable importance, and many of which are little more than minor variations of each other. Specifically, the 33 rankings include seven variations on the number of works, 12 variations on the number of citations, six variations on the number of pages, two variations on the number of citing authors, two variations for each of abstract views and downloads from the RePEc site, a measure of the success of graduate students, and the department H-index.

The rankings of individual economists are burdened by four additional sub-rankings: the closeness and betweenness measures of one's co-authorship network, the breadth of citations across fields, and the Wu-index. The its credit, the site allows you to choose criteria of your own, but you only get the world's top 5% for each.

I've created a more-useful ranking of Missouri's departments using individual economist's H-indices from the CitEc project, which contains RePEc's raw data on citations. The H-index is a simple and popular measure that is meant to capture both the quantity and impact of an author's body of work. From Wikipedia:
The definition of the index is that a scholar with an index of h has published h papers each of which has been cited in other papers at least h times.
Note that I include each author's primary affiliation only, and, to fix inconsistencies in how authors chose their affiliation, each author was assigned to a specific department within a Fed or University. Here are the top 25% of research economists in Missouri according to their H-indexes:

1Anjan V. ThakorOlin School of Business, Washington U31
2William A. BrockDept of Economics, U of Missouri30
3Robert A. PollakOlin School of Business, Washington U29
4Michele BoldrinDept of Economics, Washington U23
5Christopher NeelyResearch Division, St Louis Fed22
6James BullardResearch Division, St Louis Fed19
6Bruce PetersenDept of Economics, Washington U19
8Christopher OtrokDept of Economics, U of Missouri18
8Ping WangDept of Economics, Washington U18
8Rodolfo E. ManuelliDept of Economics, Washington U18
8Goufu ZhouOlin School of Business, Washington U18
12B RavikumarResearch Division, St Louis Fed17
12Stephen WilliamsonResearch Division, St Louis Fed17
14Michael T. OwyangResearch Division, St Louis Fed16
14Steven FazzariDept of Economics, Washington U16
14Philip H. DybvigOlin School of Business, Washington U16
17Howard WallCenter for Econ & Environment, Lindenwood U15
17Christopher WallerResearch Division, St Louis Fed15
17David WheelockResearch Division, St Louis Fed15
17Michael McCrackenResearch Division, St Louis Fed15
21Ronald M. HarstadDept of Economics, U of Missouri14
22Costas AzariadisDept of Economics, Washington U13
22Marcus BerliantDept of Economics, Washington U13
22Jeroen SwinkelsOlin School of Business, Washington U13
25Jonathan WillisResearch Division, Kansas City Fed12
25Troy DavigResearch Division, Kansas City Fed12
25Rik HaferCenter for Econ & Environment, Lindenwood U12
28Craig S. HakkioResearch Division, Kansas City Fed11
28Don SchlagenhaufCent for Household Fin Stability, St Louis Fed11
28Bill DuporResearch Division, St Louis Fed11
28Cletus C. CoughlinResearch Division, St Louis Fed11
28Jeffrey MilyoDept of Economics, U of Missouri11
28L. Randall WrayDept of Economics, UM-Kansas City11
28Werner PlobergerDept of Economics, Washington U11
35Jordan RappaportResearch Division, Kansas City Fed10
35Richard AndersonCenter for Econ & Environment, Lindenwood U10
35David AndolfattoResearch Division, St Louis Fed10
35Yi WenResearch Division, St Louis Fed10
35Barton HamiltonOlin School of Business, Washington U10
35Glenn MacDonaldOlin School of Business, Washington U10
35Hong LiuOlin School of Business, Washington U10
42George A. KahnResearch Division, Kansas City Fed9
42Richard J. SullivanResearch Division, Kansas City Fed9
42Carlos GarrigaResearch Division, St Louis Fed9
42Christian ZimmermannResearch Division, St Louis Fed9
42S. BandyopadhyayResearch Division, St Louis Fed9
42Michael PodgurskyDept of Economics, U of Missouri9
42Selahattin DiboogluDept of Economics, UM-St Louis9
42Norman SchofieldDept of Economics, Washington U9

One way to obtain department rankings from individual rankings is to take an average of the H-index for the economists within a department. Doing so would give an upward bias to departments like mine, which is small and has a skewed distribution of H-indices (i.e., we skew old). So, instead, I also used the department H-index, which benefits large departments while also trimming the high end of the distribution. A department's score is then the sum of these two Hs relative to the maximum across departments. The result is ranking that includes the two Feds and the 11 departments in Missouri with at least 2 authors registered with RePEc:

1Olin School of Business, Washington U100
2Dept of Economics, Washington U92
3Research Division, St Louis Fed91
4Dept of Economics, U of Missouri76
5Center for Econ & the Environment, Lindenwood U62
6Research Division, Kansas City Fed55
7Center for Household Financial Stability, St Louis Fed53
8Dept of Economics, UM-St Louis47
9Dept of Ag and Applied Econ, U of Missouri34
10Dept of Economics, UM-Kansas City32
11Department of Economics, St Louis U29
12Bloch School of Business, UM-Kansas City27
13Dept of Econ and Finance, U of Central Missouri22

Thursday, August 25, 2016

Building a Successful Center on a Shoestring Budget

Back in April I gave a presentation at the annual APEE conference in Las Vegas. The session was one of several on the program which had directors of centers like mine (The Hammond Institute at Lindenwood University) exchanging best practices. I had found these sessions at previous APEE conferences to be very useful, so I was very happy to be a presenter this time around. 

The Hammond Institute happens to be one of many academic centers that receives support from the Charles Koch Foundation. Anti-Koch paranoia is the chemtrails of the loopy left, so a group called UnKoch My Campus decided to use hidden microphones to record the proceedings of my session, and of two other sessions. You can find the recordings and transcripts here. Despite the fevered rambling of UnKoch guys, the actual sessions were pretty innocuous, just what you would expect from academics discussing administrative practices in a forum open to the public. 

For my part, I thanked the UnKoch people for providing the recordings. There were some people who couldn't make my session to receive my sage advice. I also sent them a better photo to include on their web page, which is called KochiLeaks, of course:

I also made sure that they had a copy of the last slide of my presentation. There was a lot of laughing at that stage of my talk, so I didn't want them to be left out of the joke:

I'd like to thank the UnKoch people once more for providing the recordings, which I have synced up with my PowerPoint slides and made into a movie. I thought I gave some great advice, so now Koch-funded centers around the world can operate at least a bit better now because they can watch my presentation. Enjoy:

Saturday, March 26, 2016

Dispelling some myths about the earnings taxes in St Louis and Kansas City

The cities of St. Louis and Kansas City are alone among their neighboring cities in levying a 1 percent tax on earnings. Given the ease with which people can change where they live and/or work within a metro area, it’s not difficult to see how the taxes put St. Louis and Kansas City at a disadvantage relative to their neighbors. Specifically, in some empirical research that I did a couple of years ago, I estimated that St. Louis lost 14,700 residents between 2000 and 2010 because the city levied an earnings tax when its neighbors did not. For Kansas City my estimates indicated that population growth over the period was 18,700 lower than it would have been in the absence of the tax.

More accurately, what the study found that these were the effects of not having the same tax structure as the average city in the United States. Because of the relative harmfulness of earnings taxes, most cities are relatively reliant on property taxes. In 2011, on average, property taxes accounted for 17 times the amount collected in earnings taxes. But in St. Louis and Kansas City, earnings taxes accounted for 2.5 and 1.6 times the revenue collected from property taxes. In short, my implicit policy proposal is that St. Louis and Kansas City would not have lost these thousands of residents if they had been like most cities and not taxed earnings.

These findings are very similar to those from an earlier paper of mine, as well as from a paper by Joe Haslag at Mizzou. All three papers offer detailed descriptions of our research methods and data, so I think I speak for Joe in saying that it would have been great if our research had resulted in an intelligent debate about the relative merits of various tax systems. Despite the fact that our research was laid open for all to see, no one has refuted our findings. It’s possible that some researchers have tried and failed, but such research would never see the light of day. It’s also likely that professional economists simply have little to gain from engaging in local policy debates instead of scrambling for journal publications.

At any rate, I think it’s fair to say that there has been no serious attempts at addressing the question of whether or not St. Louis and Kansas City should keep their earnings taxes despite the significant population losses they cause.  Instead, the counter-arguments, which have most often been made by editorialists and politicians, have tended to be equal parts hysteria, ignorance, and hatred. Given that residents of the two cities will vote on April 4th on whether or not to keep their earnings taxes, the hysteria, ignorance, and hatred have been appearing with increasing frequency in the editorials and news stories of local media outlets.

It should be noted that there are some perfectly respectable and intelligent arguments for keeping earnings taxes: They are more-stable sources of revenue, the two cities have too many non-profits (hospitals, churches, universities, etc.) who are typically exempt from property taxes, and the taxes are more fair and equitable. Although some have made these arguments, including the city of St. Louis itself and at least one St. Louis alderman, I haven’t found them to have been at the forefront. I don’t happen to buy these arguments, but that’s not the standard for respectability and intelligence.

An editorial in the St. Louis Post-Dispatch provides a handy compendium of the state of the public debate against the repeal of the earnings tax. Because it’s an almost perfect representation of the hysteria, ignorance, and hate surrounding the issue, I’ll use it to address the most common myths and misrepresentations that have been made. The lines from the editorial are in italics, followed by my responses.

1. (Q)uirky beliefs of one very wealthy man.
The belief that it is harmful for a city to impose an earnings tax when its neighbors do not ought to be a fairly standard belief for anyone with any competence at economics or math.
Along these same lines is the claim that it is perfectly normal for earnings taxes to be levied by cities. For example, the city of St. Louis has claimed that 1,000s of jurisdictions have earnings taxes. Even if we ignore the sleight of hand in using the word “jurisdictions” rather than “cities”, this claim completely misses the point. The vast majority of these jurisdictions are in one state, Pennsylvania. And, while technically true, the fact is not relevant because the question is not whether or not a city has an earnings tax, but whether it has a tax that is higher than what is levied by surrounding areas.
I should note the stupidity of attaching relevance to the wealth of someone putting forth an idea. But such is the vitriol routinely spat out by the Post-Dispatch editors at those who disagree with their progressive mission. Hate is not a substitute for reason.

2. In St. Louis, the $162 million in e-taxes is 33 percent of the general revenue budget. Plus, (y)ou want to see a real exodus, fix it so there won’t be enough money for cops or firefighters.
Technically, the 33% number is accurate, but it’s just not particularly relevant. As is usual, the number is accompanied by a list of the good things that are paid for with money from the general fund, with the implication that these things will somehow disappear if the e-tax were to be repealed. All of this is a bunch of hooey, for several reasons:
  • The idea is for the city to replace the e-tax with other revenue sources that are much less deleterious to the city’s health.
  • The percentage of general revenue funded by the e-tax is simply not that important. It might be important to know the percentage of the city’s total spending that is paid for with the earnings tax, which is a a not-so-scary 15 percent. At any rate, the distinction between the general fund and the other 52 percent of the city’s revenue is largely artificial. About $143 million dollars of the city’s revenue from outside the general fund was used to help finance spending on departments also financed by money from the general fund. 
  • Similarly, despite claims to the contrary, it is simply untrue that money outside the general fund is not discretionary from the city’s standpoint. Receiving money that has strings attached, such as federal grants to purchase fire equipment, frees up money to pay for anything else. In other words, the city’s budget is fungible. So-called “tied” money simply pays for one thing you want, allowing you to pay for another thing you also want.

3. About 55 percent of the e-tax revenue in St. Louis comes from nonresidents. If it goes away, the whole load of replacing it would fall on residents.
This statement belies a complete misunderstanding of how tax burdens are distributed, which has little to do with who physically pays a tax. Once taxes are levied, the prices of the taxed items change and the distribution of the effects depend on a variety of factors. Take, for example, a tax on the value of an office building. Although he physically pays the tax, the owner of the building does not bear the entire burden of the tax. Rents will be adjusted to pass on all or some of the tax to renters. The companies renting the space will not bear all of the taxes either because they might pay their workers less and/or charge their customers more. Unless all of the owners, workers, and customers are city residents, some or all of the burden of the tax will be shifted onto nonresidents.
It should be noted that the advantage of property taxes over earnings taxes is that the latter is a direct tax on those who live or work in the city. It is because of this directness that earnings taxes have larger negative effects on population and employment than do other taxes.

4. Who knew it was the earnings tax and not white flight that emptied out the city?
This sentence is simply an especially stupid non sequitor. No one has said that the earnings tax is the sole cause the decline of St. Louis City. What has been said is that it is one of the contributors, to the tune of 1000s per year. In fact, my own research says clearly that 1/2 of the decline between 2000 and 2010 is due to the tax. Given that there there should have been natural population growth as has occurred elsewhere, I suspect that other factors are responsible for several times the effect of the earnings tax. The existence of even worse things does not preclude fixing this one thing.

5. It would be one thing if Sinquefield had a reasonable plan for replacing the e-tax revenue. He doesn’t. It would be phased out over 10 years and replaced by … well, he’s vague on that.
It’s not clear why a private citizen should do the job of the mayor, who is supposed to be the one who prepares the city for eventualities. Even so, it is simply false that Sinquefield, or the Show-Me Institute, has not put forward alternatives. In fact, Patrick Ishmael recently wrote a piece called How Would You Pivot From the Earnings Tax? Let Me Count the Ways. See also Joe Haslag’s two very detailed pieces How to Replace the Earnings Tax in Saint Louis and How to Replace the Earnings Tax in Kansas City. Given the obviousness of these titles, any 10 year old with access to Google could have found these pieces.

6. The mayor acknowledges that earnings taxes might be a disincentive to living and working in the city, but he says studies show higher sales and property taxes are bigger disincentives.
There are simply no such studies. They do not exist. Because there are few cities dumb enough to levy earnings taxes when their surrounding areas do not, little effort has been expended estimating their effects. The studies cited above are the only ones that actually do this and, as already stated, my 2014 study estimates the effects of having an earnings tax relative to the average tax structure of other cities. Thus, what research there is finds that earnings taxes are more harmful than the average mix of taxes.