Table D.7

Model estimates, alternative specification: Different bounds for V distribution moments |$ \underline{V},\overline{V}$|

$5,000$50,000
Bunching0.40750.167
(0.0981)(0.075)
Elasticity0.050.0349
(0.0119)(0.015)
Nonoptimizer value0.84350.8512
(0.0185)(0.0272)
Optimal compensation0.0270.019
(0.0063)(0.008)
Fraction of nonoptimizers0.432
Nonoptimizer participation sensitivity–0.002
$5,000$50,000
Bunching0.40750.167
(0.0981)(0.075)
Elasticity0.050.0349
(0.0119)(0.015)
Nonoptimizer value0.84350.8512
(0.0185)(0.0272)
Optimal compensation0.0270.019
(0.0063)(0.008)
Fraction of nonoptimizers0.432
Nonoptimizer participation sensitivity–0.002

Starting with the baseline parameters in Table 3, I change the parameters |$ \underline{V}$| and |$ \overline{V}$|⁠. I change |$ \underline{V}$| from $3,500 ($35,000) to $3,600 ($36,000) for the estimation using data near the kink $5,000 ($50,000). I change |$ \overline{S}$| from $4,000 ($40,000) to $3,900 ($39,000) for the estimation using data near the kink $5,000 ($50,000). I then reestimate (1) bunching; (2) the elasticity parameter e; (3) the nonoptimizer value δ; and (4) the optimal commission. This table presents the resultant estimates. The first column lists the statistics. The second column displays my estimate of each statistic when I use data near the kink K = $5,000. The third column displays my estimate of each statistic when I use data near the kink K = $50,000. I bootstrap the estimation procedure 500 times and present bootstrapped standard errors in parentheses. See Table 4 for details.

Table D.7

Model estimates, alternative specification: Different bounds for V distribution moments |$ \underline{V},\overline{V}$|

$5,000$50,000
Bunching0.40750.167
(0.0981)(0.075)
Elasticity0.050.0349
(0.0119)(0.015)
Nonoptimizer value0.84350.8512
(0.0185)(0.0272)
Optimal compensation0.0270.019
(0.0063)(0.008)
Fraction of nonoptimizers0.432
Nonoptimizer participation sensitivity–0.002
$5,000$50,000
Bunching0.40750.167
(0.0981)(0.075)
Elasticity0.050.0349
(0.0119)(0.015)
Nonoptimizer value0.84350.8512
(0.0185)(0.0272)
Optimal compensation0.0270.019
(0.0063)(0.008)
Fraction of nonoptimizers0.432
Nonoptimizer participation sensitivity–0.002

Starting with the baseline parameters in Table 3, I change the parameters |$ \underline{V}$| and |$ \overline{V}$|⁠. I change |$ \underline{V}$| from $3,500 ($35,000) to $3,600 ($36,000) for the estimation using data near the kink $5,000 ($50,000). I change |$ \overline{S}$| from $4,000 ($40,000) to $3,900 ($39,000) for the estimation using data near the kink $5,000 ($50,000). I then reestimate (1) bunching; (2) the elasticity parameter e; (3) the nonoptimizer value δ; and (4) the optimal commission. This table presents the resultant estimates. The first column lists the statistics. The second column displays my estimate of each statistic when I use data near the kink K = $5,000. The third column displays my estimate of each statistic when I use data near the kink K = $50,000. I bootstrap the estimation procedure 500 times and present bootstrapped standard errors in parentheses. See Table 4 for details.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close