RDP 2024-01: Do Monetary Policy and Economic Conditions Impact Innovation? Evidence from Australian Administrative Data Appendix B: Additional Regression Results

Omer Majeed, Jonathan Hambur and Robert Breunig

February 2024

Figure B1: Effect of Monetary Policy Shock on R&D Spending in VAR Model
100 basis point shock
Figure B1: Effect of Monetary Policy Shock on R&D Spending in VAR Model

Notes: Response of R&D spending to monetary policy shock in a VAR containing the log levels of the CPI, real GDP, real R&D spending, the TWI, the level of the cash rate, and the Beckers (2020) shock. Based on a Cholesky decomposition with the shock measure ordered first. Dashed lines show 90 per cent confidence intervals.

Sources: ABS; Authors' calculations.

Table B1: Effect of 100 Basis Point Contractionary Shock on Share of Firms Innovating
By firm size, cash rate change
  Year 0 Year 1 Year 2 Year 3
All firms
Effect −1.25*
(0.65)		 0.03
(0.92)		 0.38
(0.64)		 1.25
(1.20)
R2 0.20 0.14 0.11 0.09
No of observations 45,053 35,635 26,302 17,536
SMEs
Effect −1.28
(0.83)		 −0.29
(1.06)		 −0.25
(0.80)		 −0.64
(1.27)
R2 0.19 0.13 0.12 0.10
No of observations 29,551 22,185 14,616 7,291
Large firms
Effect −1.03
(0.91)		 0.48
(1.19)		 1.05
(0.85)		 2.50
(1.73)
R2 0.19 0.11 0.07 0.06
No of observations 15,502 13,450 11,686 10,245

Notes: Significance assessed using T ‐ distribution with t – n degrees of freedom as suggested by Cameron and Miller (2015) to account for small number of clusters, where t is sample length and n is number of coefficients. All regressions include controls for industry, (lag) GDP growth, (lag) inflation, (lag) growth in the exchange rate, (lag) turnover growth and (lag) employment, and lag of the shock and dependent variable. ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are shown in parentheses and are clustered at an annual level.
Table B2: Effect of 100 Basis Point Contractionary Shock on Share of Firms Innovating
By firm size, Beckers shock
  Year 0 Year 1 Year 2 Year 3
All firms
Effect −4.38**
(1.49)		 −3.52**
(1.55)		 −0.03
(2.06)		 5.51**
(2.26)
R2 0.05 0.05 0.05 0.04
No of observations 45,151 35,705 26,342 17,545
SMEs
Effect −5.60***
(1.44)		 −6.44***
(1.13)		 −3.87*
(1.84)		 −0.93
(1.67)
R2 0.05 0.05 0.05 0.05
No of observations 29,648 22,255 14,656 7,300
Large firms
Effect −1.37
(3.01)		 1.01
(2.65)		 3.84
(2.64)		 10.48**
(3.99)
R2 0.03 0.02 0.03 0.03
No of observations 15,503 13,450 11,686 10,245

Notes: Significance assessed using T ‐ distribution with t – n degrees of freedom as suggested by Cameron and Miller (2015) to account for small number of clusters, where t is sample length and n is number of coefficients. All regressions include controls for industry, (lag) GDP growth, (lag) inflation, (lag) growth in the exchange rate, (lag) turnover growth and (lag) employment, and lag of the shock and dependent variable. ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are shown in parentheses and are clustered at an annual level.
Table B3: Effect of 100 Basis Point Contractionary Shock on Share of Firms Innovating
By firm size, BT shock
  Year 0 Year 1 Year 2 Year 3
All firms
Effect −1.75
(1.09)		 −0.83
(1.40)		 0.19
(0.97)		 2.28
(1.69)
R2 0.05 0.05 0.05 0.04
No of observations 45,151 35,705 26,342 17,545
SMEs
Effect −2.51
(1.50)		 −2.16
(1.45)		 −1.51
(1.22)		 −2.10
(1.43)
R2 0.05 0.05 0.05 0.05
No of observations 29,648 22,255 14,656 7,300
Large firms
Effect −0.16
(1.18)		 0.47
(1.90)		 2.02*
(1.09)		 4.94*
(2.46)
R2 0.03 0.02 0.03 0.03
No of observations 15,503 13,450 11,686 10,245

Notes: Significance assessed using T ‐ distribution with t – n degrees of freedom as suggested by Cameron and Miller (2015) to account for small number of clusters, where t is sample length and n is number of coefficients. All regressions include controls for industry, (lag) GDP growth, (lag) inflation, (lag) growth in the exchange rate, (lag) turnover growth and (lag) employment, and lag of the shock and dependent variable. ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are shown in parentheses and are clustered at an annual level.
Table B4: Effect of 100 Basis Point Contractionary Shock on Share of Firms Innovating
By firm size, levels shock
  Year 0 Year 1 Year 2 Year 3
All firms
Effect 0.12
(0.26)		 −0.18
(0.42)		 −0.21
(0.25)		 −0.39
(0.44)
R2 0.05 0.05 0.05 0.04
No of observations 45,151 35,705 26,342 17,545
SMEs
Effect 0.22
(0.38)		 0.05
(0.43)		 0.19
(0.35)		 0.81
(0.49)
R2 0.05 0.05 0.05 0.05
No of observations 29,648 22,255 14,656 7,300
Large firms
Effect −0.11
(0.29)		 −0.37
(0.53)		 −0.59*
(0.27)		 −1.02
(0.63)
R2 0.03 0.02 0.03 0.03
No of observations 15,503 13,450 11,686 10,245

Notes: Significance assessed using T ‐ distribution with t – n degrees of freedom as suggested by Cameron and Miller (2015) to account for small number of clusters, where t is sample length and n is number of coefficients. All regressions include controls for industry, (lag) GDP growth, (lag) inflation, (lag) growth in the exchange rate, (lag) turnover growth and (lag) employment, and lag of the shock and dependent variable. ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are shown in parentheses and are clustered at an annual level.
Table B5: Effect of 100 Basis Point Contractionary Shock on Share of Firms Innovating
By exporter status and firm size
  Year 0 Year 1 Year 2 Year 3
SMEs
Exporters
Effect −3.84
(1.98)		 −3.29
(1.94)		 −0.16
(3.92)		 4.07
(3.26)
R2 0.19 0.14 0.13 0.10
No of observations 7,733 5,788 3,847 2,009
Non-exporters
Effect −2.24
(1.60)		 −6.90***
(1.21)		 −2.85
(1.88)		 0.99
(1.34)
R2 0.19 0.13 0.10 0.09
No of observations 21,818 16,397 10,769 5,282
Difference by exporter status significant at: na na na na
Large firms
Exporters
Effect 3.47
(2.76)		 4.76
(3.06)		 5.91*
(2.62)		 9.73*
(4.92)
R2 0.20 0.11 0.08 0.06
No of observations 10,241 8,946 7,825 6,895
Non-exporters
Effect 3.65
(2.98)		 −1.49
(3.30)		 3.12
(2.63)		 9.79*
(4.52)
R2 0.18 0.09 0.07 0.06
No of observations 5,261 4,504 3,861 3,350
Difference by exporter status significant at: na 10 per cent level na na

Notes: Significance assessed using T ‐ distribution with t – n degrees of freedom as suggested by Cameron and Miller (2015) to account for small number of clusters, where t is sample length and n is number of coefficients. All regressions include controls for industry, (lag) GDP growth, (lag) inflation, (lag) growth in the exchange rate, (lag) turnover growth and (lag) employment, and lag of the shock and dependent variable. ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are shown in parentheses and are clustered at an annual level.
Table B6: Effect of 100 Basis Point US Contractionary Shock on Share of Firms Innovating
By exporter status and firm size
  Year 0 Year 1 Year 2 Year 3
SMEs
All firms
Effect −7.00
(4.52)		 −10.33**
(3.80)		 −9.85*
(5.22)		 −21.19
(16.05)
R2 0.19 0.14 0.12 0.10
No of observations 29,551 22,185 14,616 7,291
Exporters
Effect −11.44*
(5.61)		 −11.76
(6.85)		 −18.81**
(6.92)		 −16.62
(19.25)
R2 0.19 0.14 0.13 0.10
No of observations 7,733 5,788 3,847 2,009
Non-exporters
Effect −5.48
(4.59)		 −9.59**
(4.58)		 −6.93
(5.41)		 −25.36
(18.18)
R2 0.19 0.12 0.10 0.09
No of observations 21,818 16,397 10,769 5,282
Difference by size significant at: na na na na
Large firms
All firms
Effect −3.66
(9.19)		 −11.49*
(6.05)		 −5.59
(5.71)		 −9.62
(8.97)
R2 0.19 0.11 0.07 0.06
No of observations 15,502 13,450 11,686 10,245
Exporters
Effect −5.36
(9.22)		 −10.66
(6.44)		 −5.90
(7.54)		 −12.94
(10.87)
R2 0.20 0.11 0.08 0.06
No of observations 10,241 8,946 7,825 6,895
Non-exporters
Effect −0.14
(10.20)		 −14.17
(7.93)		 −5.39
(4.51)		 −2.67
(7.54)
R2 0.18 0.09 0.07 0.06
No of observations 5,261 4,504 3,861 3,350
Difference by size significant at: na na na na

Notes: Significance assessed using T ‐ distribution with t – n degrees of freedom as suggested by Cameron and Miller (2015) to account for small number of clusters, where t is sample length and n is number of coefficients. All regressions include controls for industry, (lag) GDP growth, (lag) inflation, (lag) growth in the exchange rate, (lag) turnover growth and (lag) employment, and lag of the shock and dependent variable. ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are shown in parentheses and are clustered at an annual level.