-
Notifications
You must be signed in to change notification settings - Fork 1
/
mixed.rmd
3355 lines (2471 loc) · 270 KB
/
mixed.rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# The Mixed Economy {#mixed}
> The devil may indeed dwell in the details, but we first need to find an angel or two in the abstractions that govern [...].
>
> --- Edward J. @McCaffery2002 [K117]
<!-- alternative epigraph: taxation *is* the social contract -->
<!-- (this is our beef, not welfare state) -->
Welfare states are governed by the economic abstractions arising from the co-existence of market and command.
When you organize production and distribution of goods and services by market *and* by command, as welfare states do, the two contradictory systems "complexly interact" [@Perrow-1999-aa] and easily produce "unintended consequences" [@Merton-1936-aa] as for example, when ().
To this day, the *mixed economy* of Postwar Western Europe, in its idealized form, is the closest thing to an angel to ever emerge from this uneasy coexistence between market and plan.
To understand first-order desiderata of welfare state design, we need to understand the conceptual compromise of the mixed economy.
Let me first reiterate the logic of its two constituent systems.
#### Market vs. Planned Economy.
\phantomsection
[\[sec:market-vs-command\]]{#sec:market-vs-command label="sec:market-vs-command"}
We can organize economic production and distribution in at least one of two ways,
by centralized, coercive command or
by decentralized, voluntary exchange.
[^1]
1. In an ideal-typical [@Weber-1920-aa] *command economy*, whoever wields an effective monopoly on the use of force also directs the economy.
A worker constructs, say, a railroad (production) and is fed by a farmer (distribution), both in fear of --- however indirect --- bodily harm from the monopolist of violence.
[^2]
2. In an ideal-typical *market economy*,
[^3]
violence is threatened only to maintain property rights and enforce contracts.
People freely exchange goods and services at equilibrium prices that balance the costs to the producer and the utility to the consumer (, p. ).
A worker constructs a railroad (labor) in return for an enforceable promise to consume (property) a given amount (wage), which he then redeems in a similar exchange with a farmer for food.
#### Capitalist Welfare State is a Pleonasm.
\phantomsection
[\[sec:interface\]]{#sec:interface label="sec:interface"}
Welfare states combine elements of command and market in the service of *equity*.
[^4]
Specifically, welfare states coercively adjust the *distributional* outcomes of markets.
[^5]
[^6]
Welfare states insure their citizens against certain individual risks (disability, sickness, unemployment), fight "poverty" by instituting (unconditional or means-tested) minimal living standards and, sometimes, reduce inequality by compressing the income and wealth distribution of the citizenry.
To reach each of these goals, welfare states have to intervene in voluntary exchanges between buyers and sellers.
For better and/or for worse, welfare states change market equilibria.
No matter the legal structure, welfare state institutions never exist *outside* the market:
even "nationalized"
[^7]
health care needs to buy doctors (at what salary?) and drugs (at what price?) on free markets.
No matter the labeling, welfare state institutions never exist *independently* of the market:
even social "insurance".
[^8]
alters labor market outcomes (who should, and can bear the burden?).
Welfare state institutions can interact with markets in more or less attractive ways:
they can have a smaller or larger [dwls]{acronym-label="dwl" acronym-form="plural+short"} (, p. ), and they can have well or ill-defined incidences (, p. ), but they always interact.
Welfare state institutions can expand (UK socialized health care) or contract (Germany social health insurance) the scope of its command, but they will always interface with the market at some frontier between the two systems, as illustrated in (p. ).
This interface is ill-defined, as two incompatible logics collide, twice:
1. The state demand curve (dashed in green) for doctors or hospitals breaks down, as the marginal utility of an additional doctors to citizens (the ultimate consumers) cannot be known.
[^9]
The state therefore has to determine the citizenry demand, usually based on a [cba]{acronym-label="cba" acronym-form="singular+short"} or a related procedure --- all of which are really nothing but fancy names for sophisticated economic planning.
2. The state cannot *command* the required supply (of doctors or hospitals), but must instead *buy* the supply from state revenues on free markets.
[^10]
Again, the logic of the market breaks down:
the state as the only buyer (of doctors or hospitals) creates a monopsony, causing distributive effects (to the disadvantage of doctors or hospitals) and welfare losses (a [dwl]{acronym-label="dwl" acronym-form="singular+short"}).
Alternatively --- maybe more plausibly --- the producers of medical care may capture the planning body and extract rents, distributing away from the public and also causing a welfare loss.
Either way, pricing poses unavoidable problems:
somewhere down the line, producers require a price, but without atomistic demand, no pareto-optimizing price can be found.
This is not to say that if health care --- or some other welfare service --- *were* completely privatized, such pareto-optimal equilibrium *would* be reached.
In fact, government procurement of health and other services can be understood as a response to *for* these goods (p. ).
Arguably, government procurement replaces one (horrendous) market failure with another, (milder) form of failure.
<!-- TODO img here The Command-Market Interface in UK and German Healthcare -->
\scriptsize{
Market supply and demand are drawn in a conventional price-quantity diagram.
For a larger version, see \autoref{fig:supply-demand} (p.~\pageref{fig:supply-demand}).
I discuss the \hyperref[sec:market-solutions-production]{competitive market equilibrium} later (p.~\pageref{sec:market-solutions-production}).}
I aim here to distill for welfare, what Edward @McCaffery2002 urges us to do about tax:
"to find an angel or two in the abstractions that govern \[...\]" [-@McCaffery2002 K117].
I look for these angels in an *ideal* closed, mixed economy.
The account I provide in the (p. ) -- ) does not resemble any *real* existing economy, where abstractions are often shrouded in historical idiosyncrasies, and angels rarely found amidst imperfect policies.
But this is a question of the (p. ), and to know what is materially doable and normatively desirable we need (p. ), not a posteriori reality.
Even without the details, the abstractions alone need considerable space to be explained.
I urge readers to take the time, even if much will seem familiar and some things appear remote to welfare, let alone taxation or democracy.
They are not:
from (p. ) to network effects (p. ), (p. ).
Missing any one of these abstractions, we cannot know what welfare and taxation can, and should do.
Four disclaimers apply to my tentative answer on this very big question:[\[sec:disclaimers\]]{#sec:disclaimers label="sec:disclaimers"}
1. [\[itm:not-original\]]{#itm:not-original label="itm:not-original"} *Not Original*.
The perspective I take here is hardly original.
Many others have, in greater width [@Stiglitz2002] or depth [@Sinn2004], with narrower [@Scharpf1997] or different foci [@Zurn-2000-aa] discussed the first-order shortcomings of "negative" integration (in the [eu]{acronym-label="eu" acronym-form="singular+short"}:
@Scharpf1997), and economic liberalization everywhere [@Stiglitz2002], and have, in that context, defined the conditions of a mixed economy.
I aim here to review the works of others and to restate some fairly conventional economic concepts in order to build a first-order checklist of welfare state design.
2. [\[itm:no-test\]]{#itm:no-test label="itm:no-test"} *No Positive Test*.
I cannot myself muster the methodological rigor or provide the econometric data, to test the first-order theories of welfare state design, but rely on mainstream literature instead.
The economics of the welfare state are vastly complex, incompletely understood and any policy initiative requires careful (empirical) investigation to balance the often contradictory imperatives of economic policy.
Moreover, as I later find for taxation, reforms of the magnitude implied herein may strain against the current limits of economic science:
experimental designs lack in external validity of a vastly complex modern economy and simulations often lack the data, or even computing power to model such inframarginal changes.
3. [\[itm:no-calibration\]]{#itm:no-calibration label="itm:no-calibration"} *No Calibration*.
I offer no calibration of the mixed economy and its institutions, and, for the purpose of this chapter, advocate no *particular* balance between market and state, efficiency and equity or any of the other trade-offs a mixed economy may face.
Instead, I highlight the capacities and dysfunctions of markets and *potentially* able to mitigate these shortcomings (p. ).
I consider under (p. ) and hypothesize how they might explain the (, p. ).
4. [\[itm:little-macroeconomics\]]{#itm:little-macroeconomics label="itm:little-macroeconomics"} *Little Macroeconomics.*
I limit this discussion to very basic concepts of the real economy, and ignore many of the more complex models, especially of finance and money.
Modern macroeconomics, including such powerful frameworks as the IS/LM model are important [originally @Hicks1937], but would go beyond the already lengthy treatment here.
I also suspect and hope that macroeconomics is best investigated by experts and its policy imperatives safely implemented by technocrats.
Monetary policy, for instance, may not raise deep normative questions or offer vexing trade-offs in need of democratic adjudication:
its imperatives hinge on contested and imperfect, but merely first-order, positive findings on a mass psychology of price and cost signals (see , p. ).
To a lesser extent, finance, too, may be politically epiphenomenal:
money and other property rights move in tandem *with*, and are *secondary* to material economic exchanges in the ideal mixed economy (see p. ).
To the extent that polities can agree on specific and measurable objectives (such as price stability, or risk diversification), macroeconomic policy really can be delegated to independent central banks or other regulatory bodies.
By limiting the discussion to a few rudimentary, but deeply understood concepts of the real economy, I also hope to reconnect regional integration and the welfare state to an *econonomic imagination* [paraphrasing @Mills-1959-aa] of our material affairs as a household --- only with a cast of billions.
[^11]
Inevitably, much of the detail and complexity that policy makers have to consider, will fall by the wayside.
If the defining characteristic of a welfare state is its uneasy union of market and plan, we must first understand the broader interplay of exchange and command in the mixed economy.
(p. ) summarizes exchange (or market) and command (or state) institutions to address five material dimensions of the human condition:
(p. ), (p. ), (p. ), (p. ) and (p. ).
This is a slightly expanded set, inspired by @MusgThet1959's [-@MusgThet1959] seminal definition of basic public policy functions:
allocation/efficiency, distribution and stabilization (for example, as cited in [@Bordo2011 4].
It also corresponds to @Samuelson-1954-eu's authoritative textbook [recently -@Samuelson2005] desiderata of a mixed economy:
1. to let scarce resources be efficiently allocated by *competitive markets*,
2. to improve the equity of market outcomes through *redistribution*,
3. to provide *public goods* by government procurement and
4. to limit inherent market instability by financial regulation and well-directed monetary and fiscal policies
As stands, the game appears stacked against the market, as there is no list of government failures.
In fact, failures abound in the command economy:
without credible information about individual utility [confer @Hayek1931], and an elegant mechanism for their aggregation [confer @Lerner1944; @Lange1934; @Debreu1954] resources are easily wasted and misallocated and even command components in market economies are prone to *government failure* [@Coase1964].
Moreover, a market economy seems to be closely related to liberal democracy,
[^17]
and bloated command economies may corrupt politics,
[^18]
or even threaten the very constitution of freedom [@Hayek1944; @Friedman1962].
Here again --- to "economize on moral disagreement" [@GutmannThompson-2004-aa K226] and to lend credence to my later conclusions --- I conservatively place the burden of proof on the state:
production and distribution by markets should *only* be replaced or altered by command when the market can demonstrably not achieve the desired outcomes, that is, when problems materialize (row 4 in ).
I now discuss the five material dimensions of human need in turn.
<!-- very important: we are here interested only in the quality of ONE of state/market interfaces, namely redistributive and general revenue tax.
emphatically *not* in other aspects of this coexistence (say, spending).
they matter only in so far as they illustrate that taxation is actually fairly important and a worthy candidate for deliberation
(by contrast, monetary policy may not be)
NOT in what the balance should be (state vs market) (though it should be possible).
NOT in what the redistribution should be (though it should be possible)
emphatically NOT whether the interface should exist at all, that is whether we should have an entirely different way of living together.
this is Because
a) keep the question manageable
b) be pragmatic; keep most institutions constant, vary one at a time
c) there is plenty enough here already; in fact, it's maybe the most controversial/clever if you only have an interface institution that is so badly designed that any of the other decisions don't really matter -->
<!-- Designing in the real world is hard.
Economic and administrative complications beset taxation and easily effect undesirable, unintended consequences [@Merton-1968-aa].
Economic complications arise where market participants interact with tax policies.
Administrative complications arise where transactions become complex, or where axiomatically assumed in real life (page ). -->
<!-- taxes vs fees vs pigou (tijnbergen!) -->
<!-- Finally, @Scharpf1997 even falls for the old social-democratic smoke grenade of "parity" contributions to social insurance, nominally shared by employees and employers, but whose incidence really always falls entirely on labor (all of the above [-@Scharpf1997 30-34] -->
<!-- [^18]: @Offe2003 reminds us that there is no "hyper-rational" answer of a *best* balance between efficiency and equity [-@Offe2003 445]:
that is an essentially political question, and should be decided by democratic sovereigns.
There are, however, objectively *better* or worse institutional designs under which these trade-offs are made.
For example, experts cannot know an optimal progressivity in a tax code, but they may well show that whichever progressivity the democratic sovereign desires will cost less in efficiency under a consumption than an income tax (for example, @McCaffery2005, @Frank2005) -->
<!-- this is about economic imagination, akin to sociological imagination, gans 1988 public Sociology
a household, but with a case of millions -->
<!-- i don't even mention the issues of imperfect economic integration here (EU!), though much of them can be read as extensions of the malaise described here -->
<!-- first, you have to show here that tax is a worthwhile issue; it *is* the social contract, and it is currently in bad shape, and it *is* the best way to address a lot of our problems (not spending, not printing money, no nothing) -->
<!-- also, crucially, the point here is *not* how inequality is positively (that is pretty complicated), nor how how the contravening institutions should be calibrated.
The point is that, *no matter* the desired redistribution, the institutions are bad. -->
<!-- this is also, when talking about mixed economy, strictly about institutions which *reconcile* command and market economy (otherwise, somewhat misleadingly, called democracy and capitalism)
it is at this interface that this diss is, and the question is whether the institutions governing this interface are good or bad, whether they can somehow satisfy the demands of *both* an efficient market economy, and a democratically powerful sovereign
the question, emphatically, is not about *where to steer*, so to speak.
but just what kind of steering mechanism we should have
the engineering metaphor doesn't fully work of course, because, as it turns out, it's not a well-defined engineering problem that you can outsource to experts (maybe not even in engineering, who knows)
there are *already* deep, last-reason kind of question *in* the design questions for the steering mechanism, so to speak
that is what this is about
some people criticise this as "market-compatible democracy", but that's misleading.
neither mechanism is somehow inherently superior, or more legitimate.
or at least there is no reason to assume this here. -->
<!--
why so careful, one might ask?
because a) it's really unclear. b) economize on moral disagreement
and really, we *should* agree that the sovereign should have institutions available to do this, as he/she pleases, if that is in fact possible (that is to be decided! make a footnote to the epilogue here!)
this is, fundamentally, neutral and open-ended: it is possible that the resulting, better institutions will strictly limit the amount of redistribution (however loosely defined!) that is possible.
or it could give the sovereign a much more powerful lever to redistribute as she sees fit.
or lastly, it could be completely impossible.
-->
### Circular Flow
### Positional Races
<!-- if you need stuff on positional races, go here, otherwise KILL IT
%garrard 2012
% briefly reference Rousseau "obsessive being-for-others" with whom "one is forced to compare onself at each instant".
% 382: ``The possession of positional goods has also been associated with measurable levels of increased well-being, while their absence has been linked with increased stress, status anxiety and other adverse psychological and physical effects such as depression, heart disease and reduced life expectancy''.
%382``There is some evidence of a connection between blood concentrations of the neurotransmitter serotonin—the so-called ‘happiness hormone’—and dominance. Unusually high levels of serotonin have been found in officers of college fraternities, athletic team captains and officers and crew members on an extended sailing voyage (Frank 1985: 23–8). Richard Wilkinson has linked being in a subordinate position with heightened levels of stress and depressed levels of serotonin (Wilkinson 1996). Increased levels of the central stress hormone cortisol, which can be measured in blood and saliva, have been found in individuals whose self-esteem and social status are threatened, and stress has been positively linked to The status of happiness 383 poor health and increased obesity. Low or declining social status has also been associated in some studies with poor cardiovascular health (Wilkinson and Pickett 2009: 38, 191; Marmot 2004).
%383: Rousseau may be right, but it might be nature already: While there is some evidence that he was more or less right about this, it seems as though our preoccupation with status and relative position is rooted in our natures which, according to evolutionary psychology, are ‘adaptive responses shaped by man’s biological nature and situation on earth’ (Hirschleifer 1978: 321).
%. ‘[T]he farther an individual fell in his local pecking order,’ according to economist Robert Frank, ‘the more serious were the threats to his survival,’ which typically provoked feelings of stress and anxiety affecting levels of testosterone and serotonin, as we have already seen (Frank 1999: 136). In such a setting, ‘low social status is an evolutionary dead end,’ to put it bluntly (Wilkinson and Pickett 2009: 204). For
%``What mattered for survival in this environment was one’s position relative to local rivals with whom individuals and groups were in direct, immediate competition for food and sex, ‘whereas comparisons with others who are distant in time and space are typically irrelevant’, from an evolutionary perspective (Frank 2007: 57). One of the central findings of Robert Frank is that most people today are relatively indifferent to the wealth and status of people far removed from them, whereas they ‘care deeply about where they stand in their various local hierarchies’ (Frank 1985: 107). Hence,''
%``For example, the voluntary simplicity movement seeks to overthrow ‘the god of positional consumption’ (Schor 1999: 61) in the name of reduced consumption, anti- materialism and sustainable development. A policy of deliberate ‘downshifting’ typically involves working less, earning less and consuming less so that there is more time to better enable people to pursue other goods. Such a lifestyle would be materially poorer than now but richer in non-material, non-positional goods and therefore happier. To the extent that it is possible to reduce the speed of the consumption treadmill so that less time and fewer resources are expended on the joyless pursuit of positional goods and status, we may be able to devote ourselves to more intrinsically satisfying activities and forms of consumption that actually improve our quality of life, such as higher air quality, more urban parkland, cleaner drinker water, reducing violent crime (Frank 1999: 11).''
%comment on this kind of anti-growth stuff; it might be anti-growth, but anti-spending might be even better. -->
## Taxation and Welfare
Why does taxation matter for human welfare?
As @Schumpeter knew, "in the modern world, taxation *is* the social contract" [@Martin2009a, 1, emphasis in original], even though social scientists have since paid little attention to it [@Tilly2009, K191].
Tax matters especially for current OECD-style welfare regimes, in which markets and states must co-exist [@Stiglitz2011].
In such a mixed economy, government must be able to draft some privately-owned resources to serve its --- ideally democratic --- *command*, without unduly altering the prices of --- ideally competitive --- *exchanges* [@Ardant1975, 165f].
<!-- TODO ref chapter on mixed econ here -->
Much of modern social integration occurs in the balance of these two contrasting logics to produce and allocate resources, "enmesh[ing] us in the web of generalized reciprocity that constitutes modern society" [@Martin2009a, 3].
State plans make the conditions *for*, and set limits *to* *individual* action, as when government builds roads or collects sewage fees.
Conversely, market exchanges produce much of the "fungible resources" used for *collective* choice, as when government pays road builders or procures sewage pipes [@Martin2009a, 4].
Of all conceivable institutions to govern the interface of states and markets, taxation --- not price controls, not expropriation, not debt, not printing money, not tariffs --- is the most equitable, efficient and sustainable [@MusgThet1959; @Stiglitz2011].
<!-- TODO ref tax-matters chapter -->
Welfare states, with their penchant for market interventions for equity, efficiency *and* sustainability, especially, rely on good taxes.
<!-- %bad überleitung, the instead doesnt work -->
Still, taxation everywhere in the OECD is in crisis.
As alternative sources of economic relief --- monetary expansion and sovereign debt --- are maxed out, structural misalignments persist, and previously forestalling (asset) bubbles have burst into their days of reckoning, public revenues appear to be strictly limited by the longtime coming contradictions of current tax regimes [@Streeck2013].
<!-- %not sure about this soure, must read it -->
The popular mixture of (progressive) income, (proportional) consumption and (regressive) payroll taxes appears to offer only harshly unattractive tradeoffs between equity and efficiency [@McCafferyHines2010], as bases have shrunk and schedules flattened [@Ganghof2006].
At the same time --- possibly partly as a result --- inequalities of incomes and wealth have widened [@Butterwegge; @Wagner2007; @Grabka2007].[^no_wealth_distro]
<!-- TODO mention here that inequality findings are contentious -->
[^no_wealth_distro]: Data on the distribution of wealth is conspicuously hard to come by [@Crouch2004, 158] or ordinally summarized in deciles, rendering much of the inequality invisible.
If governments cannot raise the resources necessary to meet democratic demands without incurring prohibitive costs, the social contract is fraying [@Crouch2004].
<!-- %so so shource -->
Whichever way governments now turn, absent better tax, they will violate the post-war capitalist compact of stable, widely shared growth [@Pierson2002; @StreeckMertens2010].
[\[sec:market-solutions-production\]]{#sec:market-solutions-production label="sec:market-solutions-production"}
Assuming, as I do, (p. ) and conditions for (p. ) markets have at least two attractive properties:
[^20]
1. when all participants have made all profitable exchanges,
[^21]
markets produce at the quantity and price where the costs to the producers equal the willingness to pay of buyers (see , p. ).
Consumers and producers in any given market enjoy maximum *surplusses*:
all consumers pay prices (at least incrementally) below the utility they receive (by area $A$), all producers receive prices (at least incrementally) above the costs they incur (by area $F$).
In this *competitive equilibrium* (see column 1, row 4 in , p. ), no one can be made better off without making someone else worse off:
it is *Pareto optimal*.
[^22]
[^23]
<!-- ![Market Equilibrium of Supply and Demand[]{label="fig:supply-demand"}](supply-demand){#fig:supply-demand width="100%"} -->
2. The *price system* (see column 1, row 4 in , p. ) makes markets into supreme "information processors" [@Hayek1931].
[^24]
Because decisions are decentralized, markets can elegantly aggregate dispersed information where a central planner would have to gather them by bureaucratic means (such as a [cba]{acronym-label="cba" acronym-form="singular+short"}).
Because market decisions are always backed by private costs, the market price system can reveal and communicate (some!) private information, where a central planner may face distorted (inflated) information about cost and utility.
##### Conditions for Perfect Competition. {#sec:perfect-competition}
The above properties of market equilibria hold only under the strict assumptions of perfect or atomistic competition.
In one formulation, this entails [@McDowell2006 157f.]:
1. *Infinite buyers and sellers*.
[\[itm:infinite-buyers-sellers\]]{#itm:infinite-buyers-sellers label="itm:infinite-buyers-sellers"}
There are so many consumers and producers in the market that an offer or bid by any one of them will have a negligible impact on prices.
Everyone is a price taker.
2. *Zero barriers to entry and exit*.
[\[itm:easy-entry-exit\]]{#itm:easy-entry-exit label="itm:easy-entry-exit"}
Firms can start or cease to produce a good at relatively little cost and effort.
All markets are wide open.
3. *Perfect factor mobility.*
[\[itm:perfect-factor-mobility\]]{#itm:perfect-factor-mobility label="itm:perfect-factor-mobility"}
In the long run, labor, capital and other inputs to production can move to wherever they earn the highest rents.
It is hire and fire.
4. *Perfect information*.
[\[itm:perfect-information\]]{#itm:perfect-information label="itm:perfect-information"}
Consumers know all prices and qualities of goods, producers know all prices and qualities of factor inputs.
People are omniscient and powerful calculators of utility.
5. *Profit-maximizing firms*.
[\[itm:profit-maximizing-firms\]]{#itm:profit-maximizing-firms label="itm:profit-maximizing-firms"}
Firms sell at the price and quantity that maximizes their profit.
[^25]
Other, exogenous criteria are not part of firm decision making.
Given the other assumptions of perfect competition, firms sell where marginal cost equals marginal revenue.
[^26]
6. *Homogeneous products.*
[\[itm:homogeneous-products\]]{#itm:homogeneous-products label="itm:homogeneous-products"}
Goods produced by one supplier are the same as those produced by another supplier of the same category.
Inputs provided by one factor owner are the same as those provided by another owner of the same factor.
All goods and factor inputs are completely commodified.
To this, one might add, according to @Wikipedia2012:
7. *Zero transaction costs.*
[\[itm:zero-transaction-costs\]]{#itm:zero-transaction-costs label="itm:zero-transaction-costs"}
Buyers and sellers can exchange goods, services and make contracts at zero cost.
Search, information, bargaining, policing and enforcement costs are assumed away.
There is no friction.
8. *Constant returns to scale.*
[\[itm:constant-returns-to-scale\]]{#itm:constant-returns-to-scale label="itm:constant-returns-to-scale"}
For any additional unit produced, costs rise by the same amount, no matter how much units are produced.
The cost function is *linear*, marginal costs are constant.
There are no (dis)economies of scale or scope.
[^27]
Constant returns to scale are related to, but distinct from assumption [\[itm:easy-entry-exit\]](#itm:easy-entry-exit){reference-type="ref" reference="itm:easy-entry-exit"} on .
Excessive economies of scale imply difficult entry.
Conversely, difficult entry implies varying marginal costs at low output.
9. *Property rights*
[\[itm:property-rights\]]{#itm:property-rights label="itm:property-rights"}
are well established.
Lastly, --- if somewhat redundant, because axiomatically assumed away by neoclassical welfare economics --- I would add:
10. *Utility equals willingness to pay.*
[\[itm:same-budgets\]]{#itm:same-budgets label="itm:same-budgets"}
The first theorem of welfare economics --- that above specified perfect markets equilibrate at pareto optimality over *initial* distributions --- is often misrepresented to imply that individual *utility* is adequately expressed in willingness to pay, on which free market exchanges operate.
This shortcut works only under the most heroic assumption of all:
that all market participants have the same budget constraint.
If everyone invoking the first theorem were to represent it in full, including the crucial, qualifying "...over *given* allocations", this condition for perfect competition would be unnecessary.
Alas, many do not.
Moreover, equating individual utility with willingness to pay constitutes what might be called the of welfare economics (p. , because it *both* assumes equal *and* unequal budget constraints:
the former to maximize utility, and the latter as incentive.
For a thesis so thoroughly grounded in neoclassical orthodoxy as this, it seems appropriate to feature this contradiction prominently.
I note wherever I relax some of these strict (and rarely plausible) assumptions in the following sections (summarized in , p. ).
It should be clear then that I make the above, conservative or neoclassical assumptions less out of conviction, but for their logical elegance and to "economize on moral disagreement", as [@GutmannThompson-2004-aa K226] have suggested.
I hope these positions will be widely acceptable to readers on the political right, and provisionally tolerable to readers on the left.
If I can show taxation to be wasteful, unfair and unsustainable and democracy to be outmatched, assuming even such liberal orthodoxy, sweeping reform should be all the more obvious.
#### Market Failures.
\phantomsection
[\[sec:market-failures\]]{#sec:market-failures label="sec:market-failures"}
Markets operate efficiently on private goods:
people can be *excluded* from their *rival* use (see ).
In this case, social costs and utility match private cost and utility.
From some goods, people cannot be effectively excluded and/or are not rivals in their consumption.
[^28]
In these cases, social and private costs and utility diverge and markets may fail.
Goods are overproduced when social cost is higher than private cost (*negative externality*), and goods are underproduced when social benefit is higher than private benefit (*positive externality*).
This problem holds more broadly for *public goods*, *common goods* and *natural monopolies* (see column 1, row 5 in ),
[^29]
summarized in according to @Samuelson-1954-eu's typology of goods [-@Samuelson-1954-eu].
##### Failure: Public Goods
\phantomsection
[\[sec:public-good\]]{#sec:public-good label="sec:public-good"}
are *underprovided* by markets, because no one can be prevented from using them (non-exclusion), and they do not get used up (non-rivalry).
Potential buyers can always free-ride on other's purchase and are therefore unwilling to pay producers adequately.
Defense is a public good and fireworks are canonical examples of public goods.
##### Failure: Common Goods
\phantomsection
[\[sec:common-good\]]{#sec:common-good label="sec:common-good"}
are *overused* on markets, because they are rival but again, no one can be prevented from using them [@Hardin-1968-aa].
Potential buyers can free-ride without paying the adequate price for exploiting the rival commons [@Hardin-1968-aa].
[^30]
[^31]
Clean air is a common good, and so may be the enlightened understanding of the electorate [@Caplan2007].
##### Failure: Natural Monopolies
\phantomsection
[\[sec:natural-monopoly\]]{#sec:natural-monopoly label="sec:natural-monopoly"}
arise where economies of scale abound in the production or distribution of goods or services, such that only a single or very few suppliers can profitably exist.
Natural monopolies are *mispriced at the margin* because after initial fixed costs --- which few single consumers would be willing or able to pay --- marginal cost become negligible.
Excessive economies of scale often occur in businesses dominated by fixed, rather than variable cost.
Sewer systems, electricity grids or search engines can be natural monopolies with prohibitively high entry costs
[^32]
for basic infrastructure (sewers, electricity masts, web indices) and later, negligibly small marginal costs for adding an additional consumer.
Natural monopolies can incur welfare losses in two ways:
1. If only one market supplier exists, it may charge monopoly prices causing a deadweight loss of underconsumption.
2. In the extreme, given the high marginal cost for the first buyer, no first buyer may come forth and the otherwise
[^33]
pareto-improving natural monopoly may not be provided at all.
##### Failure: Principal-Agent Problems.
\phantomsection
[\[sec:principal-agent-problem\]]{#sec:principal-agent-problem label="sec:principal-agent-problem"}
Principal-agent problems are one market failure of the broader class of information asymmetry problems (row 5, column 2 in , p. ), where at least one party to a trade knows less about the service, good or risk being exchanged (pioneered by Nobel laureates @Akerlof-1970-aa, Stiglitz [-@Stiglitz1976] and @Spence1974).
In principal-agent problems, the asymmetrically known quality is the effort exerted by the agent on behalf of the principal.
When agent (say, manager) effort cannot be fully observed by principals (say, owners), and principals have interests (say, long-term returns) diverging from those of agents (say, a pet project), agents may be able to cheat on their contracts.
When agents shirk successfully, they will exert less (or ill-directed) effort than would be pareto-optimal.
In the extreme, the market between principals and agents breaks down completely, as principals anticipate shirking agents and forego the transaction altogether.
Applied game theory and related disciplines offer a host of incentive designs to realign interests of principals and agents, which I need not discuss here comprehensively (but see @Tirole2006).
Solutions include (p. ) --- which creates unemployment --- or deferred compensation and tournaments --- which invites risk-seeking (Holt 1995).
Efficiency wages and tournaments try to alter the probabilistic calculus of would-be shirkers by promising outsized instead of *marginal* rewards and punishment for whichever effort *is* (randomly) observed.
Deferring (part of) the compensation to later may make agents more far-sighted, but will still strictly cap their downside risk, especially when only bonuses are deferred (for example, stock options).
The worst that can happen to an agent in any of these schemes is to loose their job, the tournament promotion or their bonus.
By contrast, the worst that can happen to a principal, is to lose everything.
In addition, if the observations of effort on which such schemes are based are spotty or isolated --- as they often are --- agents can "game the system" and incentives may turn ineffective, or even perverse.
Both (p. ) and tournament compensation also increase economic inequality:
instead of marginal productivity, they reward (p. ) and threaten losers with unemployment.
All these incentive design schemes fall short of the one genuine solution to realign principal and agent interests:
to either sell agents stock or charge them a substantial sign-up fee [@Tirole2006], effectively making agents into principals, too.
Only then can they bear both the full upside and downside risk of the enterprise.
To be able to take on such risk, of course, agents must own substantial assets, which they may not have in an unequal economy.
Principal-agent problems fail markets if, and to the extent that, inequality in assets prevents people from taking equal risks in otherwise welfare-enhancing joint projects.
They are one of the cases, where inequity makes for inefficiency, too.
This is not merely a theoretical conumdrum, but a very real problem for postindustrial and knowledge-based economies (for example, Lisbon Strategy, EU 2020, @Bell-1973-aa).
[^34]
Almost by definition, an economy based on knowledge and innovation will display information asymmetries.
The effort a knowledge worker (say, a programmer) puts in, cannot easily be observed, because that would require the principal to acquire the exact same specialized knowledge (say, a programming language).
Similarly, a would-be innovator (say, an [ict]{acronym-label="ict" acronym-form="singular+short"} entrepreneur) will always know more about her nascent and uncertain innovation than any possible investor, because otherwise, the investor would have done the project herself, already.
In short, for competitive markets to do their magical "stochastic tinkering" [@Taleb2007 211], people need to be equipped and incentivized to act on their local and uncertain ideas, and to specialize.
For a *homo economicus* at least, there can be no *entrepreneurship* without some *ownership*, too.
#### State Responses to Market Failures
\phantomsection
[\[sec:state-responses\]]{#sec:state-responses label="sec:state-responses"}
States can respond to market failures by fiscal and regulatory interventions.
I discuss them for each type of good (rows 6--8, column 1 in ).
##### Fixing Public Goods.
\phantomsection
[\[sec:public-good-response\]]{#sec:public-good-response label="sec:public-good-response"}
States can step in to *provide public goods* or subsidize their private provision, both out of the public purse (fiscal policy).
There is no regulatory or monetary response to public goods failure.
[^35]
##### Fixing Common Goods.
\phantomsection
[\[sec:common-good-response\]]{#sec:common-good-response label="sec:common-good-response"}
States can protect overused commons by a regulatory policy of "fencing in the commons".
[^36]
By doing so, governments follow the [@Coase1960] theorem.
It holds that markets can pareto-optimally resolve externalities if transaction costs are sufficiently low, and if property rights are well-defined.
[^37]
The Coase theorem is often erroneously cited to argue against state intervention.
Maintaining and *issuing new property rights* are, of course, *regulatory* state interventions.
[^38]
Alternatively, states can resolve "Tragedies of the Commons" [@Hardin-1968-aa] fiscally by slapping a *Pigouvian tax* (@Pigou1912, popularized by @Baumol1972) on using the commons.
[^39]
The Pigouvian tax prices in the externality of using a common good.
[^40]
There is no monetary response to overused commons (see , p. ).
<!-- %here are two important notes from prospect theory, and Kahnemann 2012
%I think insurance requires only the weaker (contested by Kahnemann) expected utility hypothesis (bernoulli), who says that risk aversion can be explained merely by diminishing marginal utility (specifically, argues Bernoulli, utility is a logarithmic function of wealth).
%Maybe I can do here with risk aversion and don't need LOSS aversion (based on reference points), which is what Kahnemann is out to explain, and with him prospect theory.
%Consider a weird graph about that in Kahnemann.
% The mixed economy can be understood as those institutional crutches that help system-one humans to behave as system-two humans, because system-two can't be relied upon in the long run, it's too effortful to use.
%So we use institutions to get us there -- because system two is the better.
% Kahnemann raises a big stink with indifference curves:
%he says neoclassical economics is wrong there, because prospect theory shows that people have reference points from which they are loss averse, suggesting that there is no such thing as an indifference curve (I am not sure that is formally right, the indifference curve just wouldn't be linear, as it might be).
% my problem with Kahnemann is that he meanders between positive and normative theory.
%I would think prospect theory is a positive finding, but it need change normative theory.
%Of course, Kahnemann might say, well if utility is experienced according to prospect theory, than maybe THAT is the kind of utility that we should maximize.
%I disagree:
%I think we should let system two reign, and therefore also need not abandon neoclassical economics, we just must make sure that there is always good and enough crutches around. -->
<!-- %\subparagraph{Fixing PCA-Problems}
%\subparagraph{Entrepreneurship Needs Broad-Based Ownership.} \phantomsection \label{sec:Ownership} More deductively, market economies can be thought of as welfare-maximizing because they capture decentralized and/or privately known information, and let diverse solutions compete.
%Nassim Nicholas \cite{Taleb2007} has put this succinctly by praising ``aggressive trial and error'' (\emph{ibid.}:
%xxi) in free markets that ``allow people to be lucky'' (\emph{ibid.}:
%xxi)\footnote{Being a quantitative trader by profession, \cite{Taleb2007} actually abandons rational choice when faulting Karl \cite{Marx-1867-aa} and Adam \cite{Smith-1776-lq} for believing that free markets work because of rewards.}.
%This basic impetus for capitalist creativity is expressed in desideratum \ref{des:Entrepreneurship}:
%\begin{desideratum}[Entrepreneurship]
% A desirable tax will allow entrepreneurs to make their own production decisions according to their independent judgement of private and/or local information.
% \label{des:Entrepreneurship}
%\end{desideratum}
%But what is required for this ``stochastic tinkering'' (\citealt{Taleb2007}:
%211) to work?
%Principal-agent theory suggests that maximum effort may not be exercised when effort of agents is imperfectly or non-observable, or other information asymmetries prevail.
%This of course --- unobservable effort and information asymmetries --- are likely features of highly differentiated knowledge economies\footnote{Calls for more startups, patents and research spin-offs, particularly in Germany, may serve as evidence for suboptimal incentive design under the status quo.}, where people produce an idea, not a piece of welded metal (cf.~\citealt{Bell-1973-aa}).
%Game theoretic incentive design suggests that information asymmetry problems can be resolved either by making agents shareholders or by charging them substantial sign-up fees, both of which require substantial assets to begin with \citep{Tirole2006}.
%In short, for competitive markets to work their magic, people need to be equipped and incentivized to act on their local and diverse ideas.
%For \emph{homo economicus}, there can be no \hyperref[des:Entrepreneurship]{entrepreneurship} without some \hyperref[des:BroadOwnership]{ownership}, too.
%This ties in with the simple capitalism desideratum presented earlier, but it also adds a specification:
%\begin{desideratum}[Broad-Based Ownership]
% A desirable tax allows for or promotes a broad-based ownership of the means of production.
% \label{des:BroadOwnership}
%\end{desideratum}
%include footnote for all desideratum's of where they are leading.
%reference the same footnote
%\subsection[Welfare Gains]{Welfare Gains:
%How Taxes Can Make the Pie Larger} \label{sec:PurposesOfTaxation}
%The burden of proof is on the state for interventions in the market.
%Witnesses for the defense are summarized in \autoref{tab:ends-mixed-economy} and are entertained in the below.
%Taxes and other state interventions in the economy can in fact enhance market outcomes in several ways:
%this figure is obsolete, it is now \label{tab:ends-mixed-economy}
%\begin{description}
% \item[Redistribution.] \phantomsection \label{sec:fiscal-redistribution} Governments may respond to excessive \emph{inequality} by taxing people proportionally or progressively to redistribute resources.
%While inequality is further discussed in \autoref{sec:tax-justice} on \hyperref[sec:tax-justice]{equity}, it does bear on efficiency, too, as is argued in \autoref{sec:InequalityIsInefficient}.
% \item[Risk Pooling.] \phantomsection \label{sec:state-insurance} People can possess \emph{asymmetric information} about things of uncertain quality they exchange on the marketplace.
%Insurance of unemployment, health or disability are principal examples, where sellers of risk (insurants) typically know more about their own risks than buyers of risk (insurers)\footnote{This relaxes perfect competition condition {itm:PerfectInformation} (\hyperref[itm:PerfectInformation]{perfect information}).}.
%Less-informed buyers (insurers) of risks may expect bad risks for \emph{all} insurants, causing premiums to rise and driving low-risk sellers out of the market entirely.
%This mechanism may repeat until no exchanges are made at all, defeating the purpose of insurance.
% Governments can avoid these \emph{lemons markets} by forcing everyone to take out insurance \citep{Akerlof-1970-aa}.
%When risks are universal --- as is arguably the case for unemployment, health and disability --- the premiums for these insurances are effectively taxes.
% \item[Public Goods] \phantomsection \label{sec:public-good} can be enjoyed by an arbitrary number of people without exhaustion (non-rivalrous) and no one can be excluded (non-excludable) from its using (\citealt{Samuelson-1954-eu}, summarized in \autoref{tab:Types-Of-Goods}).
%National defense is one example.
%Because people in larger groups can always free-ride on the provision of public goods by others, they are likely to be \emph{underprovided} by self-seeking individuals or markets \citep{Olson-1971-aa}.
% Governments can improve welfare by providing public goods out of tax revenue.
%the types of good table happens earlier, as \label{tab:types-of-goods}
% \item[Common Goods] \phantomsection \label{sec:common-good} are rival in their consumption but do not allow exclusion:
%everyone can benefit, but they can be exhausted \citep{Samuelson-1954-eu}.
%A \emph{Tragedy of the Commons} occurs when people overuse the common good \citep{Hardin-1968-aa}\footnote{Elinor \cite{Ostrom1990} criticizes the canonically assumed failure of commons in social science and provides an empirically grounded account of their successful, non-coercive governing.}.
%\begin{quote}
% \emph{What is common to the greatest number has the least care bestowed upon it.}\\*\\*
% Aristotle, Politics, Book II, Chapter 3 (384 b.c.-322 b.c.)
%\end{quote}
% Government can avoid the \emph{negative externality} of exhaustion of commons by \emph{pricing in} the costs of its use, an approach also known as Pigouvian taxation (\citealt{Pigou1912}, popularized by \citealt{Baumol1972})\footnote{The alternative, non-tax solution of issuing property rights on the commons (for example through an Emissions Trading Scheme), thereby making it an ordinary private good is of course a ``government'' solution, too.
%Markets cannot maintain, let alone introduce new property rights.}.%check definitively whether it is Pigovian our Pigouvian.
%note from correction:
%add coase theorem in here.
%Note that some solutions are costly.
% \item[Natural Monopolies] \phantomsection \label{sec:natural-monopoly} arise where economies of scale abound in the production or distribution of goods or services, such that only a single supplier can profitably exist\footnote{This relaxes perfect competition condition \ref{itm:infinite-buyers-sellers} (\hyperref[itm:infinite-buyers-sellers]{price taking}) and {itm:easy-entry-exit} (\hyperref[itm:easy-entry-exit]{easy entry and exit}).}.
%Excessive economies of scale often occur in businesses dominated by fixed, rather than variable costs.
%Sewer systems, electricity grids or search engines can be natural monopolies with prohibitively high entry costs for basic infrastructure (sewers, electricity masts, web indices) and later, negligibly small costs for adding an additional consumer.
% Natural monopolies can incur welfare losses in two ways:
%if only one market supplier exists, it may charge monopoly prices causing a deadweight loss of underconsumption.
%Conversely, if several suppliers exist in one market, each of them may be unable to invest at optimal levels (underprovision).
%%This ain't quite right.
%I'm missing the marginal vs.\ average cost problem.
% Governments can avoid the deadweight losses of natural monopolies by regulating them (for example, last mile ICT in Germany), franchising or outsourcing them (for example, local railway in Germany), enforcing common carriage (for example, electricity in Germany) or by nationalizing them (for example, public ownership of motorways in Germany).
% \item[Easy Market Entry.] Problems of prohibitive entry costs are not limited to natural monopolies\footnote{This relaxes perfect competition condition {itm:easy-entry-exit} (\hyperref[itm:easy-entry-exit]{easy entry and exit}).}.
%Competition can also be hampered by market players who enjoy excessive economies of scale by sheer size or past learning curves.
%Aside from regulatory responses (antitrust), governments can react proactively by means of infant industry protection or other industrial policy, the contested (de)merits of which are not under further consideration here\footnote{If infant industry protection \emph{is} considered welfare-enhancing, its medium-term welfare losses to the polity (either in the form of a DWL of subsidizing or protectionism) become a public good to the extent that a successful infant industry generates positive externalities for the rest of the economy.
%As such, infant industry protection should be partially financed out of general revenue.}.
%\end{description} -->
<!-- %\paragraph{Diminishing Utility is a Fact (Hard) to Observe} Diminishing utility is a plausible intuition grounded in our very nature\footnote{Some findings suggest that we are neurologically hard-wired to display diminishing utility in our feelings \citep{Ng-1997-aa}.}.
%Our quintessential evolutionary features, both our metabolism and propagation display starkly diminishing returns:
%you can only eat so much and rear so many children.
%Survey measures of self-reported happiness also support diminishing utility of wealth and income \citep{Veenhoven-2000-aa, Nickell2008}\footnote{If \hyperref[sec:positional-race]{positional consumption} is, in fact, rampant, survey measures may yet underestimate the diminishing utility of wealth and income.
%When people extract utility from levels of consumption \emph{relative} to others, their valuation of \emph{absolute} wealth and income is probably inflated.}.
%add positional cascades, Frank
%this is a great argument, look for empirical evidence -->
<!-- Some theories for self-reinforcing inequality
%\paragraph{Path Dependence or Cumulative Causation.} The third dynamic of \emph{path-dependent or cumulative causation} refers to situations where small initial state differences in performance lead to additional opportunities, reinforcing initial inequality.
%These additional, scarce opportunities may be awarded to individuals (or firms, or regions) based on easy but imperfect measures (think SAT scores).
%They may also be awarded based on probabilistic predictions on future performance (think past scholarships), further increasing a self-reinforcing dynamic.
%In the worst, most inequitable (and inefficient case), they are awarded based on meaningless, randomly occurring differences (think mental state on day of testing), haphazard selections (think first come, first serve) or systematic measurement bias (think habitus expectations by assessors).
%Malcolm \cite{Gladwell} illustrates this dynamic in his account of \emph{Outlier} hockey stars in Canada, whose birthdays are significantly more often in the early months of the year.
%\citeauthor{Gladwell} attributes this to very early elite selection in Canadian hockey and a cut-off point between the different leagues on each December 31st, giving children hockey players born early in the year a slight developmental advantages over their peers, amplified by the additional training they receive if initially selected.
%Path-dependent or cumulative causation of inequality are frequently observed in educational systems, particularly in those which track students early (as in Germany).
%These dynamics also apply elsewhere, where initially only slight differences lead to divergent experiences, reinforcing inequality and leading to further opportunities, for example when individuals or firms benefit from learning curves or economies of scale after initial jobs.
%\paragraph{Self-reinforcing Network Effects in Scale-Free Distributions.} The fourth dynamic are self-reinforcing network effects in homopholous networks with a scale-free graph distribution.
%Networks are formalized as \emph{graphs} comprising of \emph{nodes} (individuals, firms) interconnected with (directed, undirected and/or weighted) \emph{edges} (aquaintance, contracts) \citep{Kleinberg-2009-oz}.
%The \emph{degree} of a node is given by the number of edges emerging from it.
%I will first illustrate the graph theory of innovation diffusion \citep{Bass1969} with a simple example, the diffusion of fax machines.
%Assume that individuals (nodes) decide on whether to purchase a fax machine based on the technology's inherent value (the \emph{innovation coefficient}) and the value they realize from the number of their peers (degree) also owning a fax machine (the \emph{imitation coefficient}).
%Consider first what would happen in a grid network, where every individual is connected to each adjacent individual (all nodes have the same degree).
%Assume next that initial adopters are randomly distributed.
%Fax machines would, largely determined by their inherent value, proliferate (or not) relatively homogenously over the entire network.
%It will be unlikely that any given individual (node) has many more peers with fax machines than any other individual, rendering the imitation coefficient relatively inconsequential.
%illustrate this -->
##### Fixing Natural Monopolies.
\phantomsection
[\[sec:natural-monopoly-response\]]{#sec:natural-monopoly-response label="sec:natural-monopoly-response"}
Governments can avoid the deadweight losses of natural monopolies fiscally, by nationalizing them (for example, public ownership of motorways in Germany) and ideally charging users a fee at *average* cost (for example, trucks and coaches on federal motorways in Germany).
[^41]
Governments can also procure natural monopoly goods and services from private firms and charge users a fee at average cost (for example, local railway in Germany).
Alternatively, governments can allow privately held natural monopolies but tightly regulate them to enforce pricing at average cost and avoid a monopoly [dwl]{acronym-label="dwl" acronym-form="singular+short"} (for example, *last mile* [ict]{acronym-label="ict" acronym-form="singular+short"} or electricity in Germany).
There is no monetary response to natural monopoly problems (see footnote [\[fn:monetary-commons\]](#fn:monetary-commons){reference-type="ref" reference="fn:monetary-commons"}).
#### Monetary Policy for Price Stability.
\phantomsection
[\[sec:price-stability\]]{#sec:price-stability label="sec:price-stability"}
Monetary policy contributes best to efficient production --- and almost all other dimensions of material human need --- by staying out of the way of markets with stable prices to allow efficient exchanges in the first place.
[^42]
[^43]
When prices rise (inflation) or fall (deflation) overall, otherwise pareto-optimal exchanges may be hampered.
The welfare losses of inflation include
hoarding (of real assets),
drowning out relative price changes (noise),
increasing transaction costs (such as shoeleather costs
[^44]
and menu costs
[^45]
), as well as
general uncertainty and possibly, unrest.
Cost-wage spirals
[^46]
make inflation self-reinforcing, potentially escalating into hyperinflation.
Inflation is also believed to set off the business cycle [@Friedman1970].
Deflation, while scarcely observed in the western world in the post-Bretton-Woods regime,
[^47]
is similarly damaging.
It also makes
transactions more costly and
heightens uncertainty.
Deflation can also
cause the hoarding of cash and
trap liquidity
[^48]
and self-reinforce into a deflationary spiral.
In the long run, monetary expansion (or contraction) should follow the output of the economy $G$, so that price levels stay stable.
[^49]
Whether inflation and deflation are, as the monetarists would have it, "always and everywhere a monetary phenomenon" [@Friedman1970]
and therefore caused by an over-expansion or contraction of the money supply in the first place, or whether it has it roots in the real economy as the Keynesians would argue, is a very complicated empirical question but --- luckily for the author --- irrelevant to the state job of ensuring an efficient market place.
No matter "who dunnit", monetary policy should aim at price stability, and, perhaps, counteract real price shocks --- if and to the extent that they occur --- with (p. .).
Pooled Risks: Saving the Pie {#sec:risk}
----------------------------
[^50]
#### The Human Condition of Risk.
\phantomsection
[\[sec:human-nature-of-risk\]]{#sec:human-nature-of-risk label="sec:human-nature-of-risk"} Humans inhabit a volatile environment, marred by low probability, but high impact events (*black swans*, according to @Taleb2007), for example a serious work accident.
Unfortunately, humans also tend to ignore precisely such low probability, but high impact events [@Taleb2007], and overestimate the probabilities of favorable outcomes [@Baron2000 44], especially when they have few cognitive resources available.
When they give it some thought, most people *avoid grave downside risks*:
for example, most will prefer the certain but low cost of car liability insurance over the rare but high cost of paying for a car accident (column 2, rows 1--3 in ).
[^51]
#### Market Solutions to Risk: Insurance.
\phantomsection
[\[sec:insurance\]]{#sec:insurance label="sec:insurance"}
Markets provide *insurance* as a ready-made institution to address this human need for down-side risk aversion (column 2, row 4).
Insurants can buy protection from their grave downside risk, say, a car crash, by pooling their individual risks.
An insurance deal stipulates that all insurants will regularly chip in a small amount to cover the few unlucky car wreckers, in return for the promise that they, too, will receive a payout if they crash their cars.
[^52]
Aside from car insurance, markets do sometimes, to some extent, provide insurance against the four big life risks commonly associated with the welfare state:
unemployment,
sickness,
accident and
disability.
[^53]
Market insurance of these life risks *does not* involve a redistributive component, though that is easily assumed for unemployment insurance.
The *voluntary* exchange of premiums for coverage, in car and all other insurance, is a pareto-improvement:
all risk-averse insurants are better off, by hedging against downside risks.
Insurance is not obligatory and whoever finds it unnecessary (as may be the case for rich individuals who face limited downside risks) is not affected.
#### Market Failure in Insurance: Asymmetric Information. {#sec:asymmetric-information}
Insurance markets may fail when buyers and sellers possess asymmetric information about the risks to be insured.
[^54]
##### Ex ante,
\phantomsection
[\[sec:adverse-selection\]]{#sec:adverse-selection label="sec:adverse-selection"}
insurants with privately known high risk may disproportionately take out insurance.
Insurers, anticipating such *adverse selection*, but unable to tell high-risk ("lemons") from low-risk ("cherries") applicants,
[^55]
may expect bad risks for *all* buyers, causing premiums to rise and further driving low-risk insurants out of the market.
This *lemons market* mechanism may repeat until no exchanges are made at all, defeating the purpose of insurance [@Akerlof-1970-aa].
Adverse selection abounds in the insurance of the big life risks.
Ex ante, insurants know more about their likelihood to become unemployed, sick, to be in an accident or become disabled than their insurers.
[^56]
##### Ex post,
\phantomsection
[\[sec:moral-hazard\]]{#sec:moral-hazard label="sec:moral-hazard"}
sellers of risk (insurants) might take on more risk than they otherwise would have, causing moral hazard and in turn drive up premiums.
Moral hazard, too, may occur in the insurance of big life risks.
Ex post, insurants may be willing to live less healthily, or more recklessly than they would without insurance.
#### State Responses to Asymmetric Information in Insurance.
\phantomsection
[\[sec:state-insurance\]]{#sec:state-insurance label="sec:state-insurance"}
##### Ex ante,
states can resolve lemons markets by *regulatory* means if they force everyone at risk to take out insurance (@Akerlof-1970-aa, @Barr)
(column 2, row 8).
[^57]
States can also resolve lemons markets by *providing benefits* out of the treasury (column 2, row 9).
As the risks of unemployment, health, accident and disability are near-universal,
[^58]
the contributions for such state-run insurance are effectively taxes.
Some states (such as Germany) outsource insurance to quasi-fiscal organizations.
This "social insurance" may make a difference in administrative, legal or rhetorical terms, but its economics are that of a state-run insurance and its revenues are taxes.
Financing public insurance out of dedicated "social contributions" (usually regressive payroll taxes) instead of general revenue only adds a specific distributive component (a regressive tax on labor) to the overall tax schedule.
[^59]
[^60]
Mandating insurance or, equivalently, providing benefits out of the public purse need not redistribute resources over and above the Kaldor-Hicks improvement from resolving a lemons markets.
Properly understood, state interventions to hedge *individual* risks are meant to improve the efficiency, not equity of outcomes.
Real policy often differs from this (p. ) and layers distributive components on top of the Kaldor-Hicks improvement.
In Germany, for instance, social contributions are proportional, but capped and public health insurance covers an insurant's children at no additional cost.
These additions to schedule and benefits may or may not be desirable, but they properly belong in the realm of (p. ).
##### Ex post,
states and markets have essentially the same, clumsy method to reduce moral hazard:
[^61]
they re-individualize some of the risk by asking for co-payments or provide incentives for prudent behavior.
Exploiting moral hazard is a (p. ), and can therefore be resolved either by (partial) property rights (co-payments) or by Pigouvian taxation (incentives).
Equitable Distribution: Slicing the Pie Fairly {#sec:distribution}
----------------------------------------------
#### The Human Condition of Inequality
\phantomsection
[\[sec:human-nature-of-inequality\]]{#sec:human-nature-of-inequality label="sec:human-nature-of-inequality"}
Humans, the social animals, can deal with material scarcity in two ways [@Pickett-2009-kx]:
"because members of the same species have the same needs as each other, they have the potential to be each other's worst rival" [-@Pickett-2009-kx 197] but also,
"the potential to be each other's best source of cooperation, learning, love and assistance of every kind" [-@Pickett-2009-kx 198].
Dominance Strategies.
: The first, @Hobbes-1651-aaian [-@Hobbes-1651-aa]
strategy is one of dominance:
*homo homini lupus est*, man is a wolf to his fellow man.
[^62]
Following *dominance strategies* (column 3, row 1 in ), humans --- much like their primate cousins, the chimps --- maximize their classical evolutionary fitness (@Darwin1859, recently @Dawkins1976) by relying on individual power to secure access to scarce resources (food, shelter, females).
Affiliative Strategies.
: The second strategy is one of *affiliation* (column 3, row 3) and mutuality:
to be your "brothers keeper" ([kjv]{acronym-label="kjv" acronym-form="singular+short"} Bible, Genesis 4:9).
[^63]
Following affiliative strategies, humans ---as their other primate cousins, the bonobos --- maximize inclusive fitness (@Hamilton1964, popularized in @Wilson1975), or fitness emerging at the group-level [@Wilson2012] by cooperation, trust and reciprocal altruism [-@Pickett-2009-kx 202ff].
*Dominance hierarchies* (column 3, row 2) arise as dominance strategies prevail and (usually male) members of a species fight for higher status to successfully monopolize resources.
All but the highest status individuals are held to suffer from such heightened *relative* inequality (*ibid.*).
Instinct does not determine us to follow dominance strategies (as the chimps, according to *ibid.*) or affiliative strategies (as the bonobos, according to *ibid.*), and so we need culture and institutions to strike that balance.
*Both* markets and states can strike that balance, and reign in on dominance hierarchies.
#### Market Equity
\phantomsection
[\[sec:market-equity\]]{#sec:market-equity label="sec:market-equity"}
On competitive markets, people enter into voluntary exchanges that, at least, make everyone better off (if not necessarily by the same amount).
[^64]
Interactions under dominance hierarchies are *no such* pareto improvements;
the (physically) stronger will extract from the weaker all she can, possibly short of killing the weaker party if prolonged extraction is in the interest of the stronger party.
[^65]
*Real* existing market economies have sometimes --- but not always --- created sharp inequality.
Still, against the backdrop of dominance hierarchies, *ideal* market economies with , institutionalizing pareto-improving, voluntary exchange must be considered a civilizing achievement.
[^66]
#### Excessive Inequality
\phantomsection
[\[sec:inequality-dynamics\]]{#sec:inequality-dynamics label="sec:inequality-dynamics"}
Real existing, imperfect, modern markets display some excessively inequitable dynamics that may compromise this ability to at least somewhat compress dominance hierarchies.
In the following, I discuss some of these dynamics and further relax some assumptions of (p. ).
##### Efficiency Wages.
\phantomsection
[\[sec:efficiency-wages\]]{#sec:efficiency-wages label="sec:efficiency-wages"}
Efficiency wages are, counterintuitively, wages *above* the market-clearing rate.
At efficiency level, wages are so high, that some people cannot find gainful employment and must remain unemployed.
[^67]
Wages may be above market clearing rate because employers want to attract more applicants to choose from, because local traditions demand it, or even to feed malnourished workers.
Two other suggested reasons stand out:
1. *Reducing turnover.*
Employers may pay above-clearing wages because they want to avoid costly turnover.
When faced with both potential unemployment, attractive, and, especially, seniority-graded wages, workers may not look for a job elsewhere (for example, @Salop1979, on [ldcs]{acronym-label="ldc" acronym-form="plural+short"} @Stiglitz1974a).
2. *Avoiding shirking.*
Employers may also pay above-clearing wages because they want to deter incompletely contracted and incompletely observed workers from shirking.
Because employers (principals) can observe effort only sporadically and imperfectly, they will catch shirking workers (agents) only some of the time.
Would-be shirkers face some probability of getting caught, a reward for continued shirking, and a punishment for getting caught and may optimize their behavior accordingly.
Efficiency wages can thereby solve this (p. ):
by increasing both wage and unemployment rate, employers widen the spread between the two probabilistic outcomes for would-be shirkers (getting caught, not getting caught).
Risk-averse employees may then work hard to avoid the inflated downside risk of unemployment and loss of a generous wage [@Stiglitz1984].
Both when they increase wages to reduce turnover, and to avoid shirking, employers move the labor market out of equilibrium.
Otherwise pareto-improving employment is lost, and economic welfare foregone.
Still, efficiency wages may persist, because they are efficient --- individually utility optimizing --- for employers, if inefficient for the economy as a whole.
Employers enjoy lower turnover and shirking, at some price of higher wages, but they also push some of the social cost of above-clearing wages to everyone else.
If and to the extent that principal-agent problems are otherwise unavoidable, its socially costly remedy may not be considered a market failure:
there may be no way to make anyone better off (the unemployed) without making someone else worse off (well-paid employed and satisfied employers).
But even if efficiency wages were to destroy no welfare, they certainly redistribute it.
Here again, as in (p. ) or (p. ), people will be rewarded and punished *probabilistically* (not deterministically) and *out of proportion* to the marginal contributions they made --- or could make --- to the market economy.
##### Winner-Take-All.
\phantomsection
[\[sec:winner-take-all\]]{#sec:winner-take-all label="sec:winner-take-all"}
Five paradigms and stylized dynamics of today's economy point to the possibility of runaway social inequalities that may result in distributions much akin to dominance hierarchies, where whoever is at the top reaps most or all of the benefits [@Frank1996].
1. \phantomsection
[\[itm:non-linear-returns\]]{#itm:non-linear-returns label="itm:non-linear-returns"}
*Non-linear returns to scale* in indivisible human capital may disproportionately reward highly-skilled workers, as demand for their skills increases in the knowledge economy and they cannot be replaced by several less-skilled workers.
[^68]
[^69]
Similarly, some highly-skilled work can easily by scaled up (an algorithm can easily be deployed millions of time), but many other occupations cannot be easily scaled as production reaches a physical limit (a hairdresser can only do so many haircuts a day) (originally @Rosen1981, recently @Taleb2007).
Given the [eu]{acronym-label="eu" acronym-form="singular+short"}'s proclaimed goal to become the leading knowledge economy in the world, it is to be expected that non-linear returns to scale in indivisible human capital will further increase [@Commission2007].
*Baumol's cost disease* is a related, but inverse concept.
According to [@Baumol1965], some sectors (such as manufacturing) enjoy faster productivity growth than others (such as nursing), but, competing for the same laborers, both sectors must raise salaries.
In violation of (neo)classical dictum, the wages of, for example, nurses rise, even though they have --- supposedly --- not concomitantly gained in productivity.
Much of these sectoral productivity gains are reflected in similarly increasing non-linear returns to scale in human capital:
as the indivisible innovations (say, laser welding) of engineers are easily scaled up in manufacturing, such innovations are absent in nursing and would resist scaling.
If labor is, as [@Baumol1965] assume, in fact, (p. ) workers will be free to choose jobs with above-linear returns to scale until pay equilibrates at the same level across scalable and non-scalable work.
In this scenario, diverging productivities are priced into *sectoral costs*, not *worker pay*.
As a flip-side to diverging pay from above-linear returns, unscalable sectors will either disappear or become *relatively more* expensive to *consumers* (as seems to be the case with nursing).
is, of course, an implausible assumption.
If workers cannot, in fact, freely choose to work in scalable (computer engineer) or unscalable occupations (hair stylist), those in unscalable occupations will bear the brunt of diverging productivities in lower relative pay.
The truth, as often, will lie somewhere in between, and greatly depend on circumstance.
Some of the divergence between scalable and unscalable occupations will fall on workers, some on consumers and much will be split.
[^70]
2. [\[sec:cumulative-causation\]]{#sec:cumulative-causation label="sec:cumulative-causation"}
*Path-dependent rewards and cumulative causation* may also let winners take all or most.
If small initial state differences in performance lead to additional opportunities, initial inequality will be reinforced (@Jackson1968 [@Merton1988] recently popularized by @Gladwell).
[^71]
This pattern of path-dependent or cumulative causation is often observed in educational systems, particularly in those which track students early (as in Germany) or where social permeability is low (as in much of Europe) [@OECD2006].
3. [\[sec:network-effects\]]{#sec:network-effects label="sec:network-effects"} *Self-reinforcing network effects*
occur where economic activity occurs along homopholous networks with a scale-free graph distribution (for an introduction to graph theory, see @Kleinberg-2009-oz).
[^72]
As actors (nodes) opt to interact with similar people (homopholy, for example, @Mcpherson2001), and opportunity (innovation) spreads (cascades) only among tightly nit groups (clusters) [@Bass1969] resulting utility (degree) distributions will be decidedly non-normal (scale-free, or fractal [@Mandelbrot2004]).
Whenever features of economic consequence
[^73]
permeate through these networks of tightly clustered, self-similar nodes, opportunities and rewards will be a function of that same power-law distribution.
Inequality, by the very structure of modern society, will be excessive and self-reinforcing [@Cozzi2009; @Keller2005; @Andriani2007].
A similar dynamic, applied to economic activity in space, is implied in the agglomeration and scale effects of (p. ).
##### Different Budget Constraints.
\phantomsection
[\[sec:different-budget-constraints\]]{#sec:different-budget-constraints label="sec:different-budget-constraints"}
The magic of the (p. ) works over *given* allocations.
In the real world, wealth and income disparities cause market participants to have different budget constraints.
A higher budget constraint will inflate their willingness to pay and a lower budget constraint will deflate their willingness to pay, both at constant levels of utility.
As distributions in the real world are not a blank, egalitarian slate, the demand and supply curves are distorted by differential budget constraints.
Voluntary exchange no longer necessarily equilibrates at the pareto-optimum of *utility*, but at the pareto-optimum of *willingness*, and thereby *ability to pay*, a very imperfect and distorted proxy.
If anything, this glossed-over difference between absolute *utility* and budget-dependent *willingness to pay* is the original, logical sin of neoclassical welfare economics.
On the one hand, different budget constraints distort prices without any informational gain for @Hayek1931's superior information processing system:
budget-distorted willingnesses to pay are *misinformation*.
The old computer science adage applies here, too:
*garbage in*, *garbage out* (GIGO).
A market that equilibrates at multi-million yachts for people with outsized budgets and at malnutrition for others with very small budgets may, be formally pareto-optimal, but it may --- among other things --- not be utility-efficient in any meaningful way.
[^74]
On the other hand, such distortions of individual utility are precisely the sticks and carrots that incentivize homo economicus in market economies.
Market economies reward individuals by letting them amplify their utility signals with a larger budget constraint and they punish individuals by forcing them to subdue their signals with a smaller budget constraint.
These informationally useless spillovers from one exchange (Steve Job's inventions) to other, unrelated exchanges (Steve Job's consumption of yachts) are not merely a side-effect of market economies, they are their motivational essence.
As original sins go, they can never be redeemed in full --- at least not in this world.
So it is with the dirty little secret of neoclassical welfare economics:
to equalize all budgets at all times would be to abandon a market economy to the fullest.
Still, neoclassical economists and everyone else who relies on the first theorem must at least repent this sin by always confessing to it, and by highlighting its normative and policy implications.
Else may not await purgatory, but lies ideology.
We must not seal off, but open up an intellectual edifice to its criticism, we must not assume away but render transparent its inherent contradictions.
##### Diminishing Marginal Utility.
\phantomsection
[\[sec:diminishing-marginal-utility\]]{#sec:diminishing-marginal-utility label="sec:diminishing-marginal-utility"}
With each additional unit of goods and that people gain, the added utility may fall (theoretically by @Lerner1944 [23], recent empirical support from @Ng-1997-aa [@Veenhoven-2000-aa; @Nickell2008]).
Highly inequitable market outcomes will still be formally pareto optimal,
[^75]
but may leave great Kaldor-Hicks improvements unrealized as the poor stand to gain greater marginal utility than the rich would have to give up at their higher levels of consumption.