University:
Massachusetts Institute of Technology
Course:
15.501/516 | Accounting
Academic year:

2023
Views:

273

Pages:

52
Author:

Deborah Gentry

Coherent arbitrariness Stable preferences in an unstable world Pricing? How do we decide if we should buy a cup of coffee (for $3)? Ideally an internal representation of coffee will be compared to price. If value > price ‫ ؞‬buy If value < price ‫ ؞‬don’t buy Fundamental Values Economic theories assume underlying “fundamental” values. Rarely, however, is it possible to measure fundamental values. Virtually all tests of economic predictions examine the effect of changes in circumstances on valuation. Such results are necessary but not sufﬁcient condition for fundamental valuation. A psychological perspective Reasons you love your spouse Schwartz, 2002 Sensitivity to anchors Kahneman & Tversky, 1974 Context effects Simonson & Tversky, 1992 ‫؞‬People often have an imperfect understanding of their own values How about new products? How much is this sound worth? How do we set such prices? How come the the sound is so different from coffee? Sound should be simpler… Some intuitions Can one map the pleasure of a chocolate-bar to money? Is this mapping immediate & direct? How about chocolate vs. ice-cream? The parts Economics Psychology (theory) (fuzzy preferences) Intuitions Market (mapping $ is hard) (behavior is predictable) A toy example 200,000 180,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 og s D al W sp ew N k ap oe Sh in e Sh So ng Si er s s ng se ou M ll Ki lS el Sm ce Bl ho es ue 0 Fa Variance 160,000 Experiment 1 Procedure Do people have fundamental values? 2 (Anchor) by 2 (order) by 3 (duration) by 3 (replications) Experiment 1 Procedure Introduction Subjects listen to sound Hypothetical question No anchor 10¢ / 50¢ Bidding for real X 9 Getting real payoffs Increasing 10 sec, 30 sec, 60 sec Decreasing 60 sec, 30 sec, 10 sec Experiment 1 Results I 6 High-(50c) Low-(10c) No-Q Ratios 5 4 3 2 1 0 R 60/10 R 30/10 R 60/30 Experiment 1: Interpretation 1 Money High-(50c) Low-(10c) No-Q Duration Experiment 1 Results II WTH over time 80 High-ExPrice Low-ExPrice No-ExPrice 70 60 50 40 30 T1 T2 T3 T4 T5 T6 T7 T8 T9 Experiment 1 Conclusions Subjects do not have internal value for novel hedonic stimuli Once a response is made, other responses follow Coherent arbitrariness Real products Mean value Low ss# High ss# Increase Trackball 16.25 10.38 21.52 107% Keyboard 32.47 21.81 42.03 93% $9 wine 15.80 11.62 19.55 68% $82 wine 22.89 17.42 27.76 59% Design book 18.81 14.15 23.00 62% Belgian Chocolates 13.31 10.04 16.24 62% High dependency between the prices of the 2 wines and 2 computer accessories Tom Sawyer 10 9 8 Poetry I Poetry, charge Poetry, pay 7 6 5 4 3 2 1 0 1 min 5 min 10 min 10 8 6 Poetry 2 Poetry, charge Poetry, pay 4 2 0 -2 -4 -6 -8 -10 1 min 5 min 10 min Decision making exp 14 12 DM study, charge DM study, pay 10 8 6 4 2 0 1 min 5 min 10 min Other examples Mountain climbing, giving public talks, performing on stage, cooking for others, taking pictures, etc. “The model” Self herding Evaluating self preferences is hard Behavior is a signal for preferences Frequent behavior is taken as a strong signal Experiment 2 Procedure Will the forces of the market correct arbitrariness? 2 (Anchor) by 2 (order) by 3 (duration) by 3 (replications) Anchors were 10¢ and 100¢ Groups (Markets), not individuals Experiment 2 Procedure Introduction Subjects listen to sound Hypothetical question 10¢ / 50¢ Bidding for real X 9 Getting real payoffs Increasing 10 sec, 30 sec, 60 sec Decreasing 60 sec, 30 sec, 10 sec Experiment 2 Results I Bids and wins over time 200 Bid High-ExPrice Bid Low-ExPrice Win High-ExPrice Win Low-ExPrice 180 160 140 120 100 80 60 40 20 0 T1 T2 T3 T4 T5 T6 T7 T8 T9 Experiment 2 Results II Standard Dev of WTA 200 160 120 80 40 0 1 2 3 4 5 6 Trial # 7 8 9 Experiment 2 Conclusions Arbitrary values are not “corrected” over time Arbitrary values are not “corrected” in marketplaces Learning to arbitrary values? Experiment 3 Known random anchors Use subjects’ own SS# Larger magnitude 100 sec, 300 sec, 600 sec Same basic procedure as in Exp 1 Plus ranking of small annoying tasks WTA ($) Experiment 3 results I 10 High Anchor 9 Low Anchor 8 7 6 5 4 3 2 1 0 100 sec. 300 sec. 600 sec. Experiment 3 results II 600 sec. WTA ($) 10 Increasing 9 Decreasing 8 7 6 600 sec. 5 300 sec. 4 3 100 sec. 2 1 0 First Sound Second Sound 100 sec. Third Sound Experiment 4 What is the role of the ﬁrst anchor? Information? Changes utility or mapping? 3 (sounds) by 2 (order) by 3 (Anchor) Anchors were 10¢, 50¢ and 90¢ Duration was constant at 30 sec. Type of sound changed Fourth trail was the same as trial#1 + vise Experiment 4 Procedure I Introduction Subjects listen to sound Bidding for real Getting real payoffs Hypothetical question 10¢ / 50¢ / 90¢ 90¢ / 50¢ / 10¢ { { = = Vise Vise > $ > Why the vise? Experiment 4 Results WTA WTA-1 WTA-2 WTA-3 WTA 100 90 80 70 60 50 40 30 20 10 0 Increasing Decreasing Experiment 4 Results II Signiﬁcant effect of order No effect for anchors No effect for Vise/Sound choice Experiment 4 Conclusions The ﬁrst trial had a large role The ﬁrst response determines future responses Not an inference process Using the anchor as information Utility theory interpretation Traditional view on tradeoffs $ Time Utility theory interpretation “Incomplete preference” view on tradeoffs $ Time Utility theory interpretation Initial response and its consequences Utilities or mapping? Leftovers Experiment 3 (large payoffs & Random anchor) Rank order of small annoying tasks [Blood test, missing a bus, dropping your ice-cream etc.] Experiment 4 (3 different anchors) Choice of sound vs vise No effect for anchoring! The procedure Pair-wise preference Anchoring Y/N Question OR No measurement Pair-wise preference WTA Time Results Anchoring Bedpan: Butter: 2 [R = 0.46, F = 17.96, p < 0.01] 2 [R = 0.28, F = 5.8, p < 0.02] Shock: [R2 = 0.34, F = 8.8, p < 0.01] Drink: 2 [R = 0.29, F = 6.23, p < 0.02] Results effects on pair-wise Are the pair-wise preferences inﬂuenced by the ratios of the pair-wise anchors? 2 Shock/bedpan (R = 0.046) 2 Bedpan/butter (R = 0.023) Bedpan/drink (R2 = 0.002) Butter/shock (R2 = 0.002) 2 Drink/Shock (R = 0.011) Mapping or utilities? summary So far evidence for mapping Questions What makes mapping difﬁcult? Money? (sound vs. drink ‫ ؞‬no) Abstract attributes? Distance? Conclusions The coherence of the market seem to reﬂect the psychology of relative valuations People do not seem to have fundamental values even for simple experiences Is money a bad idea? The parts Economics (theory) Intuitions Psychology (fuzzy preferences) (mapping $ is hard) Market (behavior is predictable) Possible Applications / economics I Financial markets (Shiller, 1987) “Who would know what the value of the Dow Jones Industrial Average should be? Is it really "worth" 6,000 today? Or 5,000 or 7,000? or 2,000 or 10,000? There is no agreed-upon economic theory that would answer these questions. In the absence of any better information, past prices (or asking prices or prices of similar objects or other simple comparisons) are likely to be important determinants of prices today.” Market reactions are sensitive to performance relative to expectations (IBM is doing X better than expected) and to other relative changes (IBM is buying back X stocks) Possible Applications / economics II Labor markets Bewley (1998). “Non-union companies seemed to be isolated islands, with most workers having little systematic knowledge of pay rates at other ﬁrms. Pay rates in different non-union companies were loosely linked by the forces of supply and demand, but these allowed a good deal of latitude in setting pay” Well being within a company is related to relative pay to others and to former levels Possible Applications / economics III contingent valuation “valuations of any particular quantity [of a good] would be sensitive to its relative position within the range selected for valuation, but insensitive to which range is chosen, resulting in insensitive (or incoherent) values across studies using different quantity ranges” Frederick and Fischhoff (1998, p. 116) BUT, within subjects evaluations are coherent Possible Applications / economics IV Criminal deterrence (Ross, 1973) People seem sensitive to policy changes in deterrence (perhaps only short term) People seem insensitive to absolute levels of probability of punishment Reactions to crime also have a very strong cultural effects Questions for neuro- science Is there cardinal utility? When pain is experienced as pleasure (and the other way) what is the associated brain activity? Marketing implications First price has long lasting implications Products are comparative in nature Understanding what a product will be compared to is important Product lines and product extensions