Wallet Study Readers Digest over a period of

Wallet Study Reader’s Digest, over a period of years, “lost” more than 1, 100 wallets in various cities. Each wallet contained $50 in local currency and a name and phone number so it could be returned. The wallets were dropped in various places (phone booths, sidewalks, restaurants, etc. ).

Wallets Returned Norway 100% Denmark 100% Singapore 90% New Zealand 83% Finland 80% Scotland 80% Australia 70% Japan 70% South Korea 70% Spain 70% Austria 70% Sweden 70% U. S. 67% England 67% India 65% Canada 64% France 60% Brazil 60% Netherlands 60% Thailand 55% Belgium 50% Taiwan 50% Malaysia 50% Germany 45% Portugal 45% Argentina 44% Russia 43% Philippines 40% Wales 40% Italy 35% Switzerland 35% Hong Kong 30% Mexico 21% AVERAGE: 56%

U. S. Cities Wallets Returned Wallets Kept Seattle 9 1 St. Louis 7 3 Atlanta 5 5 Boston* 7 3 Los Angeles* 6 4 Houston* 5 5 Greensboro, N. C. 7 3 Las Vegas 5 5 Dayton, Ohio 5 5 Concord, N. H. 8 2 Cheyenne, Wyo. 8 2 Meadville, Pa. 8 2 (*wallets dropped in suburbs of these cities)

Neglected Topics Families Norms Social Outputs

Families are associations similar to the ones that Putnam and others measure. If all families were equal, omitting them from analysis would not matter. But families differ a lot: 1. Presence of one or both parents. 2. Inclusion of siblings, grandparents, cousins. 3. Number of siblings.

Norms are rules that exist independent of enforcement by the government. One way to view them is as equilibria of games such as the prisoner’s dilemma and the coordination game.

THE TAXONOMY • (1) Bilateral Costly Sanctions. One other person incurs the cost of punishing you. • (2) Multilateral Costly Sanctions. Many other people incur the cost of punishing you. • (3) Automatic Sanctions. Crashing into a driver on the road. • (4) Guilt. You feel bad about your sin, even though nobody else knows or cares. • (5) Shame. You feel you have lowered yourself. (disapproval) • (6) Informational Sanctions. Your action conveys unfavorable info about yourself.

SIGNALLING GAME • 1. Nature chooses 90 percent of workers to be steady, producing x, and 10 percent wild, producing x-y. • 2. Each worker decides to marry or not. Marriage adds m to the utility of the steady, and -z to the wild. • 3. Employers offer wages conditional on marriage. • (Spence won the Nobel Prize in 2001 for signalling)

EQUILIBRIUM • If z>y, so wild workers hatred of marriage is greater than their inferiority in output, then they stay single and the steady get married. • If z<. 9 y, everyone gets married, and the wage is pooling. There is a norm of marriage, enforced by an informational penalty. • In between is a mixed-strategy equilibrium.

WHAT CAN GOVERNMENT DO? • 1. Provide supplemental punishments. • 2. Provide info on who did what. • 3. Be careful about interfering with private sanctions. • 4. Supplement incentives for private sanctions. • 5. Help in norm creation. • 6. Fight bad norms.

Social Outputs Economic outputs depend on social capital, to be sure, but what crime, illegitimacy, loneliness, etc. ? Social outputs can also be modelled using production functions. Social outputs can also be assigned dollars values: how much would people pay to increase a social output? The variability in social outputs from differing levels of social capital may be much greater than the variability in economic outputs.

Insert a tex slide table Go to the State data tables

Regressions 1. Specification search (which variables? ) 2. Using simple bivariate correlations 3. Weighted regressions 4. Dropping outliers 5. Adding quadratic or log terms for curvature

Available Data pop malework murder cartheft illegit divorce metro retired midage unemp teachsal homeown income bachelor abortion black southern churchm voting

Bivariate Correlations with the Murder Rate malework cartheft illegit divorce metro retired midage unemp teachsal homeown income -. 25. 71. 79. 03. 27 -. 03. 14. 38. 15 -. 52. 23 bachelor abortion black southern churchm voting . 32. 84. 77. 38. 06 -. 27

Source: Robert Putnam: Bowling Alone, “Selected statistical trend data” http: //www. bowlingalone. com/data. php 3 (12/8/01)

Correlation Matrix | putnam orgs member trust --------------------------putnam | 1. 00 orgs |. 81 1. 00 member |. 73. 58 1. 00 trust |. 92. 73. 70 1. 00

Bivariate Correlations with “putnam” Social Capital malework murder cartheft illegit divorce metro retired midage unemp teachsal homeown income . 58 -. 74 -. 45 -. 56 -. 35 -. 36. 15 -. 05 -. 44 -. 13. 05. 09 bachelor abortion black southern churchm voting . 34 -. 30 -. 71 -. 62 -. 02. 69

Different Measures Murder = putnam -2. 75 (7. 55) R 2 =. 55 N = 48 Murder = trust -. 16 (6. 03) R 2 =. 48 N=41 Murder = orgs -3. 44 (5. 17) R 2 =. 35 N=50 Murder = member -1. 06 (3. 35) R 2 =. 22 N=40 Murder = putnam orgs -2. 60 . 08 trust member -. 03 . 91 (1. 89) (0. 07) (0. 47) (0. 71) R 2 =. 52 N=40

Weighting Murder = putnam income -2. 69(7. 51) -. 00012(1. 75) Murder = putnam income -2. 48(5. 58) -. 00014(2. 05) (weight = pop) Murder = putnam income -2. 46(8. 00) -. 00018(2. 38) (weight = 1/pop) Murder = putnam income -2. 69(9. 97) -. 00012(1. 60) (White robust standard errors)

Residuals from Illegit, Homeown 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. state v 1 AL 1. 313196 AK -. 0083414 AZ -3. 002368 AR -. 3776368 CA. 1270525 CO 4. 011913 CT. 6871804 DE -8. 302368 DC 20. 62919 FL -3. 582326 GA -1. 276142 HA -4. 394481 ID 5. 293575 IL. 9854771 IN -. 028383 IA -. 3358477 KA 2. 801039 KE. 0886737 LA -3. 010255 ME -3. 217509 MD 1. 957757 MA. 5488002 MI. 6701256 MN 1. 619589 MS -5. 27337 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. MO -. 1145223 MT -. 0974675 NE 2. 119589 NV 1. 030039 NH 1. 85349 NJ. 9795045 NM -5. 145649 NY -4. 276142 NC 1. 391659 ND -1. 135848 OH -3. 414522 OK -. 6083413 OR -. 7113263 PA -1. 172947 RI -5. 171453 SC -3. 259298 SD -4. 209833 TN. 614687 TX 1. 582491 UT 9. 338348 VT -1. 134354 VA 2. 237926 WA 1. 193363 WV -1. 231369 WI. 0302502 WY 1. 387181

Residuals from putnam, income 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. AL -. 87 AK. AZ 2. 44 AR. 35 CA. 83 CO. 64 CT. 66 DE -2. 40 DC. FL -. 25 GA -. 58 HA. ID -2. 94 IL 2. 70 IN 1. 76 IA -1. 22 KA 1. 26 KE -3. 59 LA 4. 01 ME -2. 42 MD 4. 29 MA -1. 98 MI 1. 78 MN. 974 MS 1. 86 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. MO MT NE NV NH NJ NM NY NC ND OH OK OR PA RI SC SD TN TX UT VT VA WA WV WI WY 1. 85 1. 40. 63 -1. 57 -1. 60 3. 75 -. 65. 18 -. 31 -2. 08 -. 44 -. 26 -. 62 -3. 05 -. 40. 11. 13 -. 35 -1. 59. 28. 01. 42 -4. 27 -. 37. 87

Dropping Outliers Murder = putnam income -2. 69 (7. 51) -. 00012 (1. 75) R 2 =. 55 Murder = putnam income -2. 81 (8. 20) -. 00016 (2. 33) R 2 =. 63 (WV dropped) Murder = putnam income -2. 61 (8. 37) -. 00016 (2. 66) R 2 =. 65 (WV, MD, LA dropped) Not much happens if outliers are dropped.

Dropping Outliers Murder = illegit homeown -0. 68 (7. 78) -0. 25 (2. 89) R 2 =. 67 Murder = R 2 =. 53 illegit homeown -0. 37 (7. 31) 0. 05 (0. 98) (DC dropped) The effect of illegitimacy becomes smaller, and home ownership becomes insignificant.

Squared and Log Terms Murder = illegit homeown -0. 68 (7. 78) -. 25 (2. 89) R 2 =. 67 Murder = log(illegit) homeown 19. 20 (5. 56) -. 33 (3. 33) R 2 =. 55 Murder = illegit 2 homeown -1. 61. 03 -. 07 (6. 27) (9. 14) (1. 23) R 2 =. 88

Murder Regression 1 R 2 =. 55 Putnam -2. 75 (7. 55) Regression 2 R 2 =. 68 Putnam Income Black Metro Southern -1. 62 (3. 06) -0. 00028 (2. 75) 0. 15 (3. 45) 0. 024 (1. 18) -1. 58 (1. 83)

Illegitimacy Regression 1 R 2 =. 34 Putnam -4. 23 (4. 91) -1. 82 -0. 00046 0. 34 0. 021 -2. 79 (1. 40) (1. 81) (3. 16) (0. 42) (1. 32) Regression 2 R 2 =. 50 Putnam Income Black Metro Southern

Male Labor Participation Regression 1 R 2 =. 35 Putnam 2. 60 (5. 08) 2. 45 0. 00012 0. 007 0. 013 -0. 76 (2. 89) (0. 74) (0. 11) (0. 42) (0. 55) Regression 2 R 2 =. 42 Putnam Income Black Metro Southern

Car Theft Regression 1 R 2 =. 21 Putnam -103. 86 (3. 53) Regression 2 R 2 =. 61 Putnam Income Black Metro Southern -37. 03 -0. 02 0. 13 8. 03 -47. 85 (1. 04) (2. 92) (0. 04) (5. 76) (2. 97)

Aristotle contra Kelvin: ``When you measure what you are speaking about and express it in numbers, you know something about it, but when you cannot express it in numbers your knowledge about is of a meagre and unsatisfactory kind. '’ Aristotle: ``. . . it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits; it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician scientific proofs. '’ Kelvin: ``In science there is only physics; all the rest is stamp collecting. '' Kelvin: ``I can state flatly that heavier than air flying machines are impossible. '’