RDP 2018-11: Consumer Credit Card Choice: Costs, Benefits and Behavioural Biases Appendix D: Heckman Sample Selection Models

Table D1: Regression Results – Probability of Perceived Net Monetary Benefit among Loss-making Respondents
Heckman sample selection probit marginal effects, 2016
  Motivation only Motivation plus controls
(1) (2)
Household income quartile (omitted category = lowest quartile)
2nd   −0.02
    (0.04)
3rd   −0.01
    (0.04)
4th   −0.02
    (0.05)
Age (years) (omitted category = 18–29)
30–39   −0.04
    (0.05)
40–49   −0.07
    (0.06)
50–64   −0.14**
    (0.06)
65+   −0.03
    (0.06)
Employment status (omitted category = employed)
Not employed   −0.00
    (0.05)
Retired   −0.01
    (0.05)
Education (omitted category = did not complete year 12)
Year 12   0.08
    (0.06)
Certificate/diploma   0.13**
    (0.06)
Bachelor or higher   0.16**
    (0.07)
Typical value of bank deposits (omitted category = less than $500)
$500–1,000   0.04
    (0.06)
$1,000–7,500   0.08
    (0.05)
$7,500–30,000   0.11**
    (0.05)
$30,000+   0.17***
    (0.05)
Motivation for holding credit card (omitted category = other)
Rewards points 0.37 0.06*
  (0.61) (0.03)
Borrowing 0.11 0.05
  (0.15) (0.04)
Payment 0.14 0.04
  (0.16) (0.03)
Interest-free period and insurances 0.24 0.04
  (0.29) (0.03)
Observations 1,388 1,388

Notes: Heckman sample selection model controls are: income, age, perceptions of the cost, convenience and usefulness for managing finances of credit cards, typical value of bank deposits, internet access, employment status, and education; ***, ** and * represent statistical significance at the 1, 5 and 10 per cent level, respectively; standard errors are in parentheses

Source: Author's calculations, based on data from Ipsos and RBA

Table D2: Regression Results – Probability of Considering Switching
Heckman sample selection probit marginal effects on censored observations, 2016
  Basic controls Full controls Full controls with income × perceived benefit interaction
Net monetary benefit (omitted category = neutral)
Gain −0.03 −0.08* −0.07
  (0.04) (0.05) (0.05)
Loss 0.10** 0.04 0.04
  (0.04) (0.05) (0.05)
Household income quartile (omitted category = lowest quartile)
2nd 0.02 −0.05 −0.04
  (0.04) (0.05) (0.05)
3rd 0.17*** 0.07 0.08
  (0.04) (0.05) (0.05)
4th 0.13*** −0.00 0.02
  (0.05) (0.06) (0.06)
Perceived net benefit (omitted category = neutral)
Better off   −0.04 −0.04
    (0.04) (0.04)
Worse off   0.20*** 0.23***
    (0.06) (0.06)
Education (omitted category = did not finish year 12)
Year 12   0.08 0.07
    (0.06) (0.06)
Certificate/diploma   0.09* 0.08
    (0.05) (0.05)
Bachelor or higher   0.17*** 0.17***
    (0.05) (0.05)
Motivation for holding credit card (omitted category = other)
Rewards points   −0.03 −0.04
    (0.06) (0.06)
Borrowing   −0.01 −0.01
    (0.06) (0.06)
Payment   −0.12** −0.12**
    (0.05) (0.05)
Interest-free period and insurances   −0.15*** −0.15***
    (0.05) (0.05)
Credit card interest rate   −0.01 −0.01
    (0.01) (0.01)
Number of bank branches in home postcode   −0.00 −0.01
    (0.00) (0.00)
Card type (omitted category = standard)
Low-rate   −0.02 −0.01
    (0.08) (0.09)
Rewards   0.09** 0.10**
    (0.04) (0.04)
Observations 813 813 813

Notes: ***, ** and * represent statistical significance at the 1, 5 and 10 per cent level, respectively; standard errors are in parentheses

Source: Author's calculations, based on data from Ipsos and RBA