Title: The Ghost in the Machine
1The Ghost in the Machine
Laura Taylor Georgia State University
2The Charge
- Behavioral Economics is it relevant for
Environmental Policy? - not
- Behavioral Economics is it relevant for
Environmental Economists?
3Example Hyperbolic Discounting
- HD implies people make far-sighted decisions when
costs/benefits occur in future but make
relatively short-sighted decisions when some
costs/benefits are immediate - Creates a time inconsistency
- Prospectively you would wish to make far-sighted
decisions, but when future arrives, you choose
short-sighted decisions.
4Relevance forEnvironmental Policy?
- Prince and Shawhan
- Do individuals exhibit behavior consistent with
HD when acting in a group context (voting for
outcomes)? - Yes. (although a little less so)
- Lab evidence suggests that this could be
important for environmental policy contexts.
5Relevance forEnvironmental Policy?
- Can you think of an actual past instance of a
group favoring some kind of future investment,
but then favoring it less as its time approaches,
even though nothing changed to significantly
worsen the (undiscounted) costs or benefits of
the investment? When I say "investment," I mean
any action with an up-front cost (of any kind)
followed by benefits.
6Is this Phenomenon of 2nd Order Importance?
- A phenomenon may be robust, but trivial in
naturally occurring settings. - A phenomenon may be substantive, but not robust
to relevant environments.
7Is the anomaly of hyperbolic discounting
important?
- Della Vigna and Malmendier, QJE, 2004
- If consumers have time-inconsistent preferences,
profit maximizing firms should respond and tailor
contracts and pricing schemes in response. - Firms would not respond to consumer deviations
that are not systematic or limited to small
stakes.
8Setup
- Consider two types of individuals (as compared to
time consistent agents) - Quasi-hyperbolic discounting self-aware
- Quasi-hyperbolic discounting unaware
- Consider two types of goods
- Investment goods (cost in t, benefit in tx)
- Time inconsistency leads to under-consumption
- Leisure goods (benefit in t, cost in tx)
- Time inconsistency leads to over-consumption
9Contract DesignInvestment Goods
- Firms will charge a fixed fee and then price
consumption below marginal cost for
time-inconsistent consumers. - Naïve overestimate the value of the future
discount - Sophisticates demand commitment device to
increase future consumption low per-use fee is
that device. - Firms price at marginal cost for time-consistent,
rational (sophisticated) consumers. - Result holds for monopolistic or competitive
firms.
10Contract DesignLeisure Goods
- Firms charge low (even negative) initial fees,
and then price consumption above marginal cost
for time-inconsistent consumers. - Naïve underestimate future consumption
contracts exploit this - Sophisticates demand commitment device to
decrease future consumption high per-use fee is
that device. - Firms price at marginal cost for time-consistent,
rational consumers.
11Evidence?
- Contracts in several industries are consistent
with models of naïve (or sophisticated) consumers
with time-inconsistent preferences. - Investment Good Health Clubs
- Initial fixed fees with pricing below MC (p0)
12Evidence?
- Bridget Jones Diary
- Sat, 31 Dec. New Years Resolutions. I WILL
... go to the gym three times a week not merely
to buy sandwich. - Mon, 28 April. Number of gym visits so far this
year 1, cost of gym membership per year 370
cost of single gym visit 123. - Leisure Good Credit Cards
- Zero or negative initial fees
- Pricing above marginal cost
13Evidence?
- Strong Hypothesis Tests
- Model predicts
- If X is true, then Y would be observed
- If Z is true, then M would be observed
- Collect data and test it.
14Left wondering Are behavioral anomalies of
2nd order importance for environmental policy?
15Where is the heavy lifting?
- Carbon taxes / gas taxes
- Impact fees
- Development infrastructure
- Storm water utilities
- Removal of perverse subsidies
- use-it or lose-it water permits
- production subsidies
- Pricing water
16Simple Battle of Efficient Water Allocation
- Georgia Water Coalition
- Some lawmakers still believe that allowing the
sale of water is the best way to allocate it in
times of shortage. Theyre wrong. If the market
determines water use, water will be sold to the
highest bidders... - Georgia Sierra Club / Upper Chattahoochee
Riverkeeper - "This act put a dollar value on water, and
undoing that will be extremely important."
17Behavioral EconomistNot so fast! What about
Israel DC exp?
- Daycare problem
- Late parents
- Economist solution
- Price being late
- Randomized experiment, designed by the economists
- Outcome Late pick-ups went up
18How to Rationalize the Data?
- The fine normalized the behavior.
- Application to Environmental Policy
- Voluntary water restrictions versus water pricing
during droughts - Could you see water use increase with pricing?
19Not so fast!
- Fee instituted at day care was 3/child (flat
rate per day). - Pricing in the real world?
- Between 1 and 2 per minute for late pickup ? no
late pickups! - Lesson Learned? Dont let an economist pick your
pricing strategy.
20Lesson Learned?
- The luxury of feedback, learning, and multiple
adjustments absent in many policy contexts. - Assumption of marginal adjustments moving one
toward the optimum may be a bad one.
21Behavioral Economics is it relevant for
Environmental Policy?
22Vernon Smith
- Many psychologists appear to find irrationality
everywhere, and many economists appear to see the
findings as everywhere irrelevant. - most standard theory provides a correct first
approximation in predicting motivated behavior
23The Charge Reformulated
- Demonstrate the importance of Behavioral
Economics for environmental policy
24- In mainstream economics, the concept of rational
behavior is defined as a social construct an
economic choice within the social institution of
mercantile exchange not an isolated decision
made in a private bubble.
25The Ghost in the Machine
26Wikipedia Decision Making and Behavioral Bias
- Over seventy different cognitive biases are
identified, including - endowment effect, focusing effect, hyperbolic
discounting, impact bias, information bias, loss
aversion, rosy retrospection (not my dad!),
status quo bias, ambiguity effect, anchoring,
availability heuristic, hindsight bias,
egocentric bias (coauthor bias), halo effect,
projection bias, notational bias, recency effect,
primacy effect, polarization effect, illusion of
transparency, in-group bias, overconfidence...
27Is this Humanity?
28The Ghost in the Machine