RDP 2023-03: Doing Less, with Less: Capital Misallocation, Investment and the Productivity Slowdown in Australia Appendix B: Additional Results

Jonathan Hambur and Dan Andrews

March 2023

Table B1: Robustness for Other Capital Measures
  Non-current assets PIM Gross investment
Productivity 0.036***
(0.007)		 0.060***
(0.007)		 0.085***
(0.005)
GFC*Productivity −0.001
(0.008)		 0.0148*
(0.008)		 0.017***
(0.005)
Post-GFC*Productivity −0.024***
(0.008)		 −0.035***
(0.009)		 −0.025***
(0.006)
Controls
Industry*Year Y Y Y
State*Year Y Y Y
State unemp*Productivity Y Y Y
Age Y Y Y
Size Y Y Y
Sales growth Y Y Y
Observations 2,048,427 1,254,739 2,048,427
Adj R-squared 0.031 0.020 0.024

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17. Gross investment expressed relative to lagged non-current assets; winsorised at 99th percentile.
Table B2: Results with No Primary Sectors
Growth in non-current assets
  All non-primary Most financially dependent quartile Other financial dependence
Productivity 0.030***
(0.007)		 0.051***
(0.016)		 0.026***
(0.009)
GFC*Productivity 0.008
(0.008)		 −0.012
(0.017)		 0.006
(0.009)
Post-GFC*Productivity −0.022***
(0.008)		 −0.043***
(0.014)		 −0.008
(0.009)
Controls
Industry*Year Y Y Y
State*Year Y Y Y
State unemp*Productivity Y Y Y
Age Y Y Y
Size Y Y Y
Sales growth Y Y Y
Observations 1,900,014 550,463 1,187,647
Adj R-squared 0.030 0.032 0.032

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17. Agriculture and mining sectors removed.
Figure B1: Change in Industry Mark-ups
Unweighted average, relative to 05/06
Figure B1: Change in Industry Mark-ups

Notes: Regression of industry average mark-ups on time dummies and industry dummies. Primary sector removed. Quartiles defined based on Rajan Zingales financial dependence metric.

Sources: Authors' calculations; Demmou, Stefanescu and Arquie (2019)

Table B3: Robustness for Dynamics
Growth in non-current assets
  Baseline With lagged stock With lagged growth
Productivity 0.036***
(0.007)		 0.031***
(0.008)		 0.036***
(0.007)
GFC*Productivity −0.001
(0.008)		 0.002
(0.007)		 0.000
(0.008)
Post-GFC*Productivity −0.024***
(0.008)		 −0.024***
(0.007)		 −0.024***
(0.008)
Lagged stock/growth −0.057***
(0.0016)		 −0.098***
(0.003)
GFC*Lagged stock/growth 0.003***
(0.001)		 −0.008**
(0.003)
Post-GFC*Lagged stock/growth 0.017***
(0.001)		 −0.011***
(0.003)
Controls
Industry*Year Y Y Y
State*Year Y Y Y
State unemp*Productivity Y Y Y
Age Y Y Y
Size Y Y Y
Sales growth Y Y Y
Observations 2,048,427 2,048,427 2,048,427
Adj R-squared 0.031 0.055 0.043

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17.
Table B4: Productivity Persistence
Regression of MFP on lagged MFP
  Pre-GFC GFC Post-GFC
Lagged productivity 0.831***
(0.034)		 0.777***
(0.022)		 0.813***
(0.020)
Observations 501,012 645,395 1,112,118
Adj R-squared 0.644 0.635 0.620
Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and robust to heterogeneity and autocorrelation. Pre-GFC sample is 2004/05 to 2006/07, GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17.
Table B5: Robustness for Size and Controls
Growth in non-current assets
  Full sample No firms < $2 million sales Allow firm controls to vary
Productivity 0.036***
(0.007)		 0.036**
(0.014)		 0.037***
(0.007)
GFC*Productivity −0.001
(0.008)		 −0.002
(0.014)		 0.001
(0.007)
Post-GFC*Productivity −0.024***
(0.008)		 −0.040**
(0.017)		 −0.022***
(0.007)
Controls
Industry*Year Y Y Y
State*Year Y Y Y
State unemp*Productivity Y Y Y
Age Y Y Y
Size Y Y Y
Sales growth Y Y Y
Observations 2,048,427 609,735 2,048,427
Adj R-squared 0.031 0.045 0.031

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17. Column 3 allows age, size and sales growth variables' coefficients to change across the periods.
Table B6: Cobb-Douglas Production Function
Growth in non-current assets
  Full sample By period
Productivity 0.021***
(0.003)		 0.030***
(0.007)
GFC*Productivity   0.000
(0.008)
Post-GFC*Productivity   −0.017**
(0.008)
Controls
Industry*Year Y Y
State*Year Y Y
State unemp*Productivity Y Y
Age Y Y
Size Y Y
Sales growth Y Y
Observations 1,983,297 1,983,297
Adj R-squared 0.013 0.013

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17.
Table B7: Results by Age
Growth in non-current assets, MFP
  Full sample Young Old
Productivity 0.036***
(0.007)		 0.046**
(0.022)		 0.033***
(0.007)
GFC*Productivity −0.001
(0.008)		 −0.009
(0.028)		 −0.002
(0.008)
Post-GFC*Productivity −0.024***
(0.008)		 0.008
(0.027)		 −0.018**
(0.008)
Controls
Industry*Year Y Y Y
State*Year Y Y Y
State unemp*Productivity Y Y Y
Age Y Y Y
Size Y Y Y
Sales growth Y Y Y
Observations 2,048,427 230,972 1,817,455
Adj R-squared 0.031 0.043 0.027

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17. ‘Young’ is firms 0–5 years of age, ‘Old’ over 5 years of age.
Table B8: Capital Reallocation by Other Financial Dependence Regression
Growth in non-current assets
  Most financially dependent quartile (with R&D) Other financial dependence (with R&D) Most financially dependent quartile (with intangibles) Other financial dependence (with intangibles)
Productivity 0.053***
(0.016)		 0.025***
(0.008)		 0.054***
(0.017)		 0.025***
(0.008)
GFC*Productivity −0.019
(0.017)		 0.007
(0.009)		 −0.021
(0.018)		 −0.008
(0.008)
Post-GFC*Productivity −0.041***
(0.015)		 −0.012
(0.009)		 −0.045***
(0.015)		 −0.010
(0.009)
Controls
Industry*Year Y Y Y Y
State*Year Y Y Y Y
State unemp*Productivity Y Y Y Y
Size Y Y Y Y
Age Y Y Y Y
Sales growth Y Y Y Y
Observations 481,708 1,405,591 468,587 1,418,712
Adj R-squared 0.035 0.033 0.033 0.031

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17.
Table B9: Capital Reallocation by Selected ANZSIC Divisions
Growth in non-current assets
  Manufacturing Construction Wholesale trade Retail trade Accommodation Professional services Other services
Productivity 0.039**
(0.019)		 0.058**
(0.023)		 0.017
(0.022)		 0.029
(0.028)		 0.041**
(0.015)		 0.002
(0.010)		 0.057***
(0.018)
Post-GFC* Productivity −0.012
(0.024)		 −0.038*
(0.021)		 −0.001
(0.025)		 −0.022
(0.036)		 −0.028
(0.034)		 −0.017
(0.012)		 −0.042
(0.031)
Controls
Industry*Year Y Y Y Y Y Y Y
State*Year Y Y Y Y Y Y Y
State unemp*
Productivity		 Y Y Y Y Y Y Y
Size Y Y Y Y Y Y Y
Age Y Y Y Y Y Y Y
Sales growth Y Y Y Y Y Y Y
Observations 204,401 350,881 170,718 262,083 137,995 285,431 138,830
Adj R-squared 0.038 0.032 0.030 0.028 0.029 0.032 0.029

Notes: ***, ** and * denote statistical significance at the 1, 5 and 10 per cent levels, respectively. Standard errors are in parentheses and clustered at an industry level. GFC sample is 2007/08 to 2010/11, post-GFC sample is 2011/12 to 2016/17, overall sample is 2004/05 to 2016/17.