Graduate Experimental Economics

1
Graduate Experimental Economics

• Lecture 9
• John Hey

2
New Developments

• Macroeconomics Experiments
• Field Experiments
• Neuro Economics Experiments

3
Macroeconomic Experiments

• See the survey by Roberto Ricciuti and references
  therein.
• We will just look at one example Riedl, A. and
  Van Winden, F. (2001) Does the wage tax system
  cause budget deficits? A macro-experiment,
  Public Choice, 109, 371-394.
• But the same principles apply elsewhere.

4
Riedl and Van Winden

• In this paper we investigate experimentally the
  economic functioning of a wage tax system for
  financing unemployment benefits in an
  international economy, in particular in reaction
  to budget deficits and tax adjustment. Our
  results support the hypothesis that due to
  out-of-equilibrium price uncertainty producers
  are reluctant to employ inputs.We also observe a
  downward pressure on wages exacerbated by an
  over-supply of labor by consumers. These
  observations can explain the budget deficits
  found. Furthermore, we find that tax adjustments
  in order to facilitate a balancing of the budget
  has strong adverse effects on unemployment and
  real GDP.

5
(No Transcript)
6
Broad Structure

• Two countries Big and Small
• 3 consumers in each country with log-linear
  utility/payment function supply labour and
  consume.
• 2 type X (traded) producers and 3 Type Y
  (protected) producers in each country demand
  labour and produce. CES prod.fn.
• Multiple Unit Double Auction.

7
Purpose to compare two tax regimes

• Result 1. For the constant tax regime in both
  countries substantial and persistent budget
  deficits are observed, which do not vanish over
  time.
• when taxes adjust to the previous period budget
  deficit or surplus...
• Result 5. For the dynamic tax regime the budget
  deficits converge to zero from below in both
  countries.

8
Unemployment

• Result 2. For the constant tax regime (periods 1
  to 8) in both countries, unemployment converges
  to equilibrium unemployment from above, in the
  sense that the asymptotic (long-term) values,
  though larger, are statistically not
  significantly different from the predicted
  values.
• Result 6. In the long run the unemployment rates
  measured as deviation from equilibrium
  unemployment rates increase from 6 to 12 in
  the small country and from 4 to 18 in the large
  country. Long-term RGDP decreases by 8 in the
  small country and by 13 in the large country.

9
Field Experiments

• See Harrison G and List J A (2004), Field
  Experiments", Journal of Economic Literature, 62,
  1009-1055.
• Six factors that can be used to determine the
  field context of an experiment
• the nature of the subject pool,
• the nature of the information that the subjects
  bring to the task,
• the nature of the commodity,
• the nature of the task or trading rules applied,
• the nature of the stakes, and
• the nature of the environment that the subject
  operates in.

10
Field Experiments A Classification

• a conventional lab experiment is one that employs
  a standard subject pool of students, an abstract
  framing, and an imposed set of rules
• an artefactual field experiment is the same as a
  conventional lab experiment but with a
  nonstandard subject pool
• a framed field experiment is the same as an
  artefactual field experiment but with field
  context in either the commodity, task, or
  information set that the subjects can use
• a natural field experiment is the same as a
  framed field experiment but where the environment
  is one where the subjects naturally undertake
  these tasks and where the subjects do not know
  that they are in an experiment.

11
A Field Experiment

• Camerer, Colin F. 1998. Can Asset Markets Be
  Manipulated? A Field Experiment with Racetrack
  Betting, Journal of Political Economy, 106,
  45782.
• To test whether naturally occurring markets can
  be strategically manipulated, 500 and 1,000
  bets were made, then canceled, at horse racing
  tracks. The net effects of these costless
  temporary bets give clues about how market
  participants react to information large bets
  might contain. The bets moved odds on horses
  visibly (compared to matched-pair control horses
  with similar prebet odds) and had a slight
  tendency to draw money toward the horse that was
  temporarily bet, but the net effect was close to
  zero and statistically insignificant. The results
  suggest that some bettors inferred information
  from bets and others did not, and their reactions
  roughly canceled out.

12
Parimutuel (win) Betting

• Let Bn be the total amount bet on the horses that
  did not win and Bi be the amount bet by better i
  of the set of bettors who bet on the winning
  horse (i1,2,,I)
• Then bettor i gets back his bet plus
• k Bi/(B1 B2 BI) Bn
• where (1-k) is the share that the track takes.

13
What Camerer did

• He placed bets on a particular horse and then
  cancelled the bet just before the start.
• He wanted to see if his temporary bids changed
  the prices of the horses
• and also whether this was a way of manipulating
  the market.

14
(No Transcript)
15
(No Transcript)
16
Conclusions

• The answer to my title question is "no."
  Pari-mutuel racetrack odds could not be
  systematically manipulated with a sample of 50
  500 and 33 1,000 bets on randomly chosen
  temporary-bet horses, compared with matched-pair
  control horses in the same race. There is no
  evidence during any period-while the bet was
  "live" and lowered the odds on the bet horse,
  after it was canceled (raising the odds), or
  over the entire pre-bet to post-bet period - that
  other bettors responded systematically to the
  temporary bets. The bets also did not increase
  the variance of pre- and post-bet changes on the
  temporary-bet horse, relative to the controls, so
  it is not the case that temporary bets worked
  strongly in opposite directions.

17
A Natural Experiment

• Deal or No Deal aka Affari Tuoi
• Contestants sequentially choose between an offer
  of a sure amount of money and a gamble over the
  remaining boxes.
• Can be modelled either static or dynamic.
• Data can be used to estimate the risk preferences
  of the contestants.
• See Conte et al LUISS (plus many others)

18
Neuro Economics

• This is perhaps the most active (as well as being
  a very expensive) field.
• Major players are Colin Camerer, Ernst Fehr, Aldo
  Rustichini, David Laibson, George Loewenstein.
• Jockeying for a Nobel Prize.
• Neuroeconomics combines neuroscience, economics,
  and psychology to study how we make choices. It
  looks at the role of the brain when we evaluate
  decisions, categorize risks and rewards, and
  interact with each other. (Wikipedia
  definition)
• Opponents include Gul and Pesendorfer ("The Case
  for Mindless Economics") and Ariel Rubinstein,

19
Colin Camerer and Ernst Fehr

• Both are very good at PR.
• Camerer is at Caltech http//www.hss.caltech.edu/
  camerer/camerer.html and has a neuro site there
  http//www.neuro-economics.org/
• Fehr is at Zurich (and MIT) http//www.iew.uzh.ch/
  chairs/fehr/team/fehr.html.

20
A Camerer Risk/Ambiguity experiment

• Ming Hsu, Meghana Bhatt, Ralph Adolphs, Daniel
  Tranel, Colin F. Camerer (2005), Neural Systems
  Responding to Degrees of Uncertainty in Human
  Decision-Making, Science, 9, 310, 1680 - 1683

21
A Camerer Risk/Ambiguity experiment
22
Fig. 2. Regions showing greater activation in
response to ambiguity than in response to risk.
Random-effects analysis of all three treatments
revealed regions that are differentially
activated in decision-making under ambiguity
relative to risk (P   0.001, uncorrected
cluster size k   10 voxels). These regions
include (A) left amygdala and right
amygdala/parahippocampal gyrus (coronal section
shown at y 7 in MNI space heat map represents
t statistic with 42 degrees of freedom) and (B)
bilateral OFC. (C) Mean time courses of amygdala
and OFC (time synched to trial onset, dashed
vertical lines are mean decision times error
bars are SEM n 16).
23
Fig. 3. Regions showing greater activation in
response to risk than in response to ambiguity.
Random-effects analysis of all three treatments
revealed brain regions that are differentially
activated in decision-making under risk. These
regions include (A) dorsal striatum, as well as
precuneus and premotor cortex (table S8) (P  
0.001, uncorrected cluster size k   10
voxels.) (B) Mean time courses for risk regions
(time synched to trial onset, dashed vertical
lines are mean decision times error bars are
SEM n 16). (C) Regions of the dorsal striatum
significantly correlated with expected values of
subjects' choices in risk condition of Card-Deck
treatment (red) and both risk and ambiguity
conditions of Knowledge treatment (blue) (P lt
0.005, uncorrected cluster size k   10 voxels)
24
A Recent Fehr Paper

• Ernst Fehr, Manfred Spitzer, Urs Fischbacher,
  Bärbel Herrnberger and Georg Grön (2007), The
  Neural Signature of Social Norm Compliance,
  Neuron 56, 185-196.

25
(No Transcript)
26
Conclusions on Neuro

• Interesting to know which parts of the brain are
  active when we are doing things.
• and to know that different parts are active when
  we are doing different things.
• But how does this information help us to do
  economics? To formulate new theories? If
  behaviour is different we knew that before if it
  is not different then why might we be interested?

27
The End

• of the module
• though hardly the end of experimental economics.
• I hope that you have enjoyed the module.