Skip to main content

Advertisement

Table 3 CPI bias, robustness check

From: An estimation of CPI biases in Argentina 1985–2005 and its implications on real income growth and income distribution

  Dep. Var.: Share of food
Small set of control variables Extended set of control variables
Using Expenditure Using Income Using income as instrument of expenditure Using Expenditure Using Income Using income as instrument of expenditure
(1) (2) (3) (4) (5) (6)
Dummy for ENGH 96/97 − 0.110*** − 0.086*** − 0.115*** − 0.099*** − 0.076*** − 0.104***
(0.004) (0.004) (0.004) (0.004) (0.004) (0.004)
Dummy for ENGH 04/05 − 0.111*** − 0.101*** − 0.115*** − 0.100*** − 0.084*** − 0.105***
(0.005) (0.004) (0.005) (0.005) (0.006) (0.006)
Ln of per household expenditure − 0.118***   − 0.130*** − 0.097***   − 0.108***
(0.002)   (0.003) (0.003)   (0.004)
Ln of per household income   − 0.100***    − 0.072***  
  (0.003)    (0.003)  
Food prices/non-food prices 0.038** 0.050*** 0.032** 0.046*** 0.061*** 0.041***
(0.015) (0.015) (0.015) (0.015) (0.015) (0.015)
Observations 10,380 10,364 10,364 10,380 10,364 10,364
R-squared 0.407 0.35 0.405 0.424 0.382 0.423
Adj. R-squared 0.406 0.349 0.404 0.421 0.379 0.420
Cumulative Bias in CPI from 85/86 to 96/97 60.6 57.6 58.6 64.0 65.2 61.9
P. 5% 62.5 60.2 60.5 66.4 68.6 64.3
P. 95% 58.4 54.7 56.5 61.7 61.5 59.3
Annual Implicit Bias from 85/86 to 96/97 8.11 7.51 7.71 8.88 9.16 8.40
P. 5% 8.53 8.04 8.10 9.44 9.98 8.95
P. 95% 7.67 6.95 7.28 8.34 8.31 7.86
Cumulative Bias in CPI from 85/86 to 04/05 61.0 63.5 58.7 64.4 69.0 62.3
P. 5% 63.0 66.3 61.0 67.2 72.4 65.0
P. 95% 58.3 60.2 56.0 60.5 64.5 58.5
Annual Implicit Bias from 85/86 to 04/05 4.59 4.92 4.33 5.03 5.68 4.76
P. 5% 4.85 5.30 4.60 5.42 6.23 5.11
P. 95% 4.28 4.50 4.02 4.54 5.04 4.30
Cumulative Bias in CPI from 96/97 to 04/05 0.95 13.90 0.27 1.07 10.80 1.04
P. 5% 7.26 20.00 6.11 8.73 19.80 8.14
P. 95% − 5.70 7.12 − 5.84 − 8.10 0.44 − 7.09
Annual Implicit Bias from 96/97 to 04/05 0.11 1.65 0.03 0.12 1.26 0.12
P. 5% 0.83 2.44 0.70 1.01 2.42 0.94
P. 95% − 0.62 0.82 − 0.63 − 0.87 0.05 − 0.76
  1. Bold numbers indicate bias values estimated
  2. Notes P. 5 and P. 95% correspond to percentile 5 and percentile 95 of 90 percent bootstrap confidence interval. Small set of control variables includes percentage of members ages 0–4, percentage of members ages 5–9, percentage of members ages 10–15, percentage of members ages 15–19, dummies for Capital Federal, male head, spouse present, head has a job, spouse has a job, head and spouse have both a job, owner occupied and free housing occupied. Extended set of control variables includes also percentage of members ages 20–35, percentage of members ages 35–60, number of income recipients, dummies for head self-employed, head employer, household has a last one car, head is married, head is single, head unmarried with spouse, educational levels of heads, and head’s job sectors. Robust standard errors in parentheses. * significant at 10%; ** significant at 5%; *** significant at 1%
  3. Source Own elaboration based on expenditure surveys