Saturday, October 28, 2017

Decent criticisms of economics? Here are 111 of them.

Courtesy of a lengthy Twitter thread following the above tweet by @UnlearningEconomics, here are 111 criticisms of economics. Most are spot on. My personal favourites are 42, 51, 58, 66, 80, 87, 92, and 106.

1. Too many unobservable parameters that change every time they are 'measured'
2. The functions are all smooth, when everything interesting that happens in a capitalist economy is not
3. Defends unrealistic assumptions on the basis that they can make good predictions. Almost never makes good predictions.
4. Embeds libertarianism, consumerism and capitalism into models without questioning them
5. Excessively 'thin' conception of the environment as amenable to cost-benefit analysis, no acknowledgement of how ecosystems work
6. Obscene levels of professional arrogance
7. Bizarre obsession with optimisation models. Guys, there's other kinds of maths
8. Use of the word 'proof' for things that are either trivial or don't technically count as 'proofs'
9. Textbooks/classes that teach model first, reality second (or never)
10. Bizarre obsession with linear regression. Guys, there's other kinds of statistics
11. Literally no case of an acknowledgement that a model/theory is flat out wrong and should be completely discarded
12. No real concept of the social. Putting 'identity' in a utility function doesn't count
13. Ridiculously hierarchical journal system that stifles creativity
14. Articles that are way too long, despite pretension that maths allows one to be concise
15. Virtually no conception of power, exploitation, conflict
16. Possibly worst of all, has undue levels of influence on policy despite its huge shortcomings
17. Huge problem with under-representation of women and POC
18. Research largely centred on the USA and other western countries, who need it least
19. At best, only pays much attention to pressing social issues *after* a catastrophe
20. Is the source of 'you just don't understand economics', repeated ad nauseum by academics and internet knuckle-draggers alike
21. Defines itself by a methodology instead of the object of study
22. Is taught as ‘economics’, so that students don’t even realise they are only learning one perspective
23. In practice, has generally favoured the powerful through its influence on policy
24. Is rife with aggregation problems, and pretends they don’t exist
25. Absorbs concepts used by critics/non-mainstream economics, watering them down to the point of being unrecognisable
26. Spends large amounts of time, pages, volumes, on issues which are basically trivial
27. Related: has little conception of its own history. Has been reinventing the wheel (but with maths!) for a long time
28. Insists on approaching history largely through the use of statistics, preferring even bad statistics to the qualitative
29. Makes its students more selfish (don’t @ me)
30. Uses maths to represent quantities which don’t have measurable, scientific units
31. A lot of statistical results are essentially obtained through p-hacking. (Though this isn’t specific to econ)
32. Little in the way of professional ethics or duty
33. Irritating habit of defining all criticisms as someone else’s problem. “That’s sociology!” Or “I’m not even a macroeconomist!”
34. Overuse of equilibrium & comparative statics, little conception of how things actually change
35. Crises are not exogenous shocks
36. Dear microeconomists. Why do I care what people do in laboratory gambles? Besides, Gigerenzer did it better.
37. Behavioural economics: making economic man more realistic by having him solve more complex utility functions
38. Also behavioural economics: ‘nudging’ people to be more ‘rational’, because economists clearly have that idea down
39. Game theory weirdly convinces it’s practitioners it’s applicable almost everywhere when it’s applicable almost nowhere
40. Preference satisfaction & efficiency/output are the focal points of almost every model. The normative implications are rarely laid bare
41. Calibration. Wtf
42. Large areas of undergraduate economics are the same as books from hundreds of years ago, with no empirical reason for why
43. The EMH is either the most trivial or most ridiculous theory I’ve ever seen, depending on who is arguing for it
44. On the whole, mainstream textbooks and economists STILL don’t get why banks work
45. Rational expectations is blatantly absurd to even the most casual observer
46. Models are invented too fast, and used before they are fully understood (my latest medium post is about this)
47. Prices simply do not play the coordinating role attributed to them by mainstream economists in many markets
48. Abuse of labels like ‘dynamic’ ‘imperfect information’ or ‘bounded rationality’, when the models do not truly reflect these ideas
49. With the way it’s taught, people who learn it often cannot think any other way. It is difficult to do even if you want to
50. The ‘law’ of demand and supply is quite clearly no such thing. Counterexamples are easy to find
51. The certainty with which comparative advantage is propagated as an argument for free trade is proportional to its utter inapplicability
52. No appreciation of the social economy. I bet most economists don’t even know what it is
53. In general, workplace dynamics are absent (except in terms of contract efficiency)
54. ‘Intuitive’, ‘plausible’. What the hell do these mean and why are they in every econ paper?
55. The spectrum auctions were  flawed, please stop going on about them
56. The ‘empirical revolution’ aka we’ve got clever statistical methods and look at them no not over there the shiny bit here
57. Randomised control trials, because we like experimenting on poor people-how else do we determine which experiments on poor people work?
58. Instrumental variables, because the best way to deal with unverifiable assumptions about endogeneity is to introduce another one
59. Regression discontinuity design, because that thing that’s obvious from looking at one graph needs an entire paper
60. Subjective well-being research has yet to tell us anything we didn’t already know
61. Utility as a concept has always been circular. Revealed preference doesn’t help, it just makes it more obvious
62. Functional forms are only chosen for tractability reasons, and every popular functional form has counterfactual implications
63. Excessive use of mathematical notation in explicit detail for god knows what reason. Makes it seem more scientific, I guess?
64. Similar love of graphs as apparently making ideas that are obviously wrong seem right. Laffer, Kuznets, environmental Kuznets, etc.
65. Convergence, as implied by the Solow model, is obviously wrong. 'Conditional convergence' just makes it unfalsifiable
66. Total Factor Productivity is an artefact of accounting, it doesn't measure productivity
67. Pareto optimality/efficiency are close to useless concepts, unverifiable and unobtainable in the real world
68. The market for lemons is so clearly wrong I'm not sure how it's so popular. The original referee rejections were right
69. Specific version of #27, but worth saying: I can't believe what the mainstream has done to Keynes. Sorry mate
70. Abuse of the term 'fallacy', which means an *objective* error in logic, to mean 'doesn't fit my little story about hotdogs'
71. Virtually ignores household work, care work, and anything else that isn't directly counted in GDP
72. Regional inequality buries GDP as a measure of national welfare. The UK is a case in point, the US isn't much better
73. Overly complicated statistical models hide assumptions and obscure more transparent relationships in descriptive data
74. The concept of 'marginal' anything is completely alien to most people, firms, governments, I don't know why it's used
75. The 'but there is a paper that does X' defence. 1. It will never be part of 99% of the mainstream 2. It's probably a crap attempt anyway
76. Kaldor-Hicks compensation is such a poorly thought out idea. People who've lost their jobs don't want 'transfers', they want jobs
77. 'Heterogeneity' in macroeconomics means 'people differ by one or two parameters', which is limited heterogeneity, to say the least
78. 'Models help us be logical and scientific'
79. EU was invented when mathematicians didn't know the difference between time averages & ensemble averages. Now they do, so get rid of it.
80. Markets don't clear, and they don't 'try' to clear. Businesses deliberately keep stocks to deal with uncertainty/change
81. Idea that maths is necessary for being logical has 2 issues (1) maths isn't always (Godel) & (2) words can obviously be logical
82. The basis of public choice theory - that political actors are selfish - has been convincingly falsified
83. Coase's theory of firms is either so vague as to be useless or wrong, otherwise all activity would be subsumed under a single firm
84. Where are the activist economists? Global slavery, meat-eating, the environment, are all both huge economic and huge moral issues
85. Has any economist ever been dismissed from the profession/respectability for doing terrible things? Scholes, Schleifer, etc.
86. Regulation policy is insufficiently systemic, overly focused on individual firms. Huge problem in the run-up to the crisis (see VAR)
87. There is generally little in the way of legal and institutional understanding when economists discuss policy
88. The overarching idea that 'competition is good' ignores the many cases where it is negative (such as arms races)
89. Unnecessary level of deference to existing models/literature, regardless of how wrong it is
90. Repetitive education. No I do not want to learn oligopoly theory 5 times
91. Lack of habitual use of case studies, survey methods, interviews, and other non-statistical research methods
92. Too much use of downloadable statistics without really knowing where they came from and their limitations
93. Time series, with the exception of finance, has data which verge on useless
94. The Lucas critique is a devastating critique of mainstream economics. Too bad it's somehow been interpreted as support for it.
95. Economists got Brexit wrong, and they barely even understand how. Hint: 'but our forecasts were right!' isn't the point
96. Econometrics "assumes independent, identically distributed populations when modelling unique & interdependent individuals" HT @BruceMcF
97. Excessive conviction in 'human capital' as a way to solve social problems and inequality
98. Any discipline which not only produces but *rewards* people like Robert Barro, Ed Prescott and Eugene Fama clearly has problems
99. It is common to see accounting identities interpreted as causal, probably due to economists mistaking them for equilibrium conditions
100. The causality paradigm in applied micro has completely overreached. Many social phenomena do not have 'causes'
101. 'Yes, we know this idea is wrong but we use it as a benchmark'
102. Inflation targeting (esp. CPI) played a huge role in blinding policymakers to the brewing financial crisis. No idea why it's still used
103. Monetarism is the single most ridiculous and most repeatedly falsified doctrine to have ever made it into policy. Still it lives on.
104. Becker-esque economic imperialism is so obviously counterfactual it's hard to believe it ever had a place in the discipline
105. Well-worn, but yes-it *is* sad/strange that economists invented their own 'Nobel'
106. The idea that the 'supply side' and 'demand side' are independent is wrong, and a source of (very) bad policy
107. Economists are largely responsible for Uber's surge pricing policy. 'nuff said
108. Major disciplinary institutions like the AEA were formed during in the red scare, hugely biasing them. These biases persist.
109. The Chicago School of anti-trust has taken over the discipline and competition authorities, doing irreparable damage
110. Ideas like 'government intervention' and 'externalities' only make sense if you assume the status quo is neutral. Which it isn't.
111. OK one more. Economic forecasts are really bad. Everyone knows it so why do we carry on making them??

1 comment:

  1. Lethal criticism of economics? Here it is!
    Comment on Cameron Murray on ‘Decent criticisms of economics? Here are 111 of them.’

    Economists are scientifically incompetent. In colloquial terms, both orthodox and heterodox economists are stupid or corrupt or both.

    Economics is a failed science since 200+ years. The four main approaches ― Walrasianism, Keynesianism, Marxianism, Austrianism ― are mutually contradictory, axiomatically false, materially/formally inconsistent, and all got the pivotal concept of the subject matter ― profit ― wrong. With the pluralism of provably false theories, economics is what Feynman called a cargo cult science.#1

    What the representative economist does not understand until this day is that economics is NOT a science of behavior and that both Walrasian microfoundations and Keynesian macrofoundations are false.#2 Methodologically it holds: if it isn’t properly macro-axiomatized, it isn’t economics at all.#3

    The decent criticism of economics is insufficient and counterproductive: “The moral of the story is simply this: it takes a new theory, and not just the destructive exposure of assumptions or the collection of new facts, to beat an old theory.” (Blaug)#4

    Economics is a failed science and incompetent critics like fresh-thinker Murray are an essential ingredient of the mess.

    Egmont Kakarot-Handtke

    #1 Economics: 200+ years of scientific incompetence and fraud

    #2 For details of the big picture see cross-references Failed/Fake scientists

    #3 Ten steps to leave cargo cult economics behind for good

    #4 The stupidity of Heterodoxy is the life insurance of Orthodoxy