The Treasury

Global Navigation

Personal tools

Treasury
Publication

Past, Present and Future Developments in New Zealand's Terms of Trade - WP 06/09

Appendix 2 – Decomposing New Zealand’s economic growth

The technique below, developed by Fox, Kohli and Warren (2003), is used to decompose the sources of New Zealand’s economic growth. Using a modified Diewert-Morrison decomposition, the contributions of total factor productivity (TFP) growth, labour and capital utilisation, the terms of trade, and the trade balance to New Zealand’s GDP growth for the period of 1983 to 2005 are examined.

The technique, originally proposed by Kohli (2003), allows for the fact that the terms of trade effect is not homogeneous of degree zero in prices. This is because a proportional change in export and import prices will change the impact of the terms of trade unless the trade is balanced.

Using the same notation and definitions as in Fox et al (2003), nominal GDP growth, real value added and real GDP are decomposed as follows:

.

Real value added is

.

and real GDP is

.
.

Where . is the growth in nominal GDP between periods t-1 and t. Rt-1,t, Gt-1,t, Ht-1,t, . and . are the contributions to nominal GDP growth from changes in TFP, the terms of trade, the trade balance, labour utilisation and capital utilisation respectively. . is the growth rate in domestic prices and . is the contribution to nominal GDP growth from prices.[23] Appendix Table 2 presents the results for 1991 to 2005 as well as geometric means for the entire period and the periods from 1991 to 1999 and 2000 to 2005.

The analysis uses annual Statistics New Zealand National Accounts data for 1990 to 2005. All variables are as defined in Fox, Kohli and Warren (2003) with domestic expenditure equalling the sum of public and private consumption and private investment. To calculate the input of labour, Statistics New Zealand data on the total number of people employed and the average weekly paid hours were multiplied together. Capital stock data was taken directly from Statistics New Zealand which differs from the methodology used by Fox et al as they use a method comparable with that of the OECD. Compensation of employees was taken as the value of labour with the share of labour defined as the value of labour divided by nominal GDP. The share of capital is defined as the remainder over nominal GDP.

For the entire period, nominal GDP grew at an average annual rate of 5.0%.[24] However for the period of 2000 to 2005, this annual growth rate in nominal GDP average 6.1%. The difference between ‘real value added’ and ‘real GDP’ is due to changes in the terms of trade and the trade balance. Over the period, the contribution from changes in the trade balance average out to be zero. As a result, the difference between the growth in real value added and the growth in real GDP is due exclusively to improvements in the terms of trade. Although the terms of trade contributed only 0.2% to the average growth in real GDP since 1991, this increased to 0.6% for the period of 2000 to 2005. In fact, for 2004 and 2005, this contribution was significantly higher. For these years the terms of trade accounted for 1.8% and 1.4% respectively (approximately two-fifths) of real GDP growth.

Appendix Table 2– Decomposition of GDP Growth using Fox et al (2003) Alternative Methodology
Year NGDP Domestic Prices (PS) Real VA Real GDP Terms of Trade (G) Balance of Trade (H) Labour (XL) Capital (XK) TFP(R)
1991 1.040 1.040 1.000 1.015 0.985 1.000 0.991 1.021 1.003
1992 0.997 1.013 0.984 0.992 0.992 1.000 0.985 1.008 0.999
1993 1.022 1.018 1.004 1.000 1.002 1.001 1.000 1.013 0.987
1994 1.071 1.007 1.063 1.056 1.007 1.000 1.013 1.029 1.013
1995 1.075 1.015 1.059 1.055 1.005 0.999 1.026 1.037 0.991
1996 1.066 1.022 1.043 1.042 1.002 0.999 1.015 1.031 0.995
1997 1.054 1.014 1.039 1.036 1.003 1.000 1.011 1.022 1.003
1998 1.037 1.011 1.026 1.032 0.995 1.000 1.006 1.014 1.011
1999 1.024 1.013 1.011 1.014 0.996 1.000 1.000 1.016 0.998
2000 1.048 1.006 1.042 1.042 1.000 1.000 1.003 1.020 1.018
2001 1.059 1.026 1.032 1.029 1.002 1.001 1.012 1.029 0.988
2002 1.074 1.024 1.049 1.038 1.010 1.000 1.015 1.026 0.997
2003 1.054 1.014 1.039 1.049 0.993 0.998 1.018 1.032 0.998
2004 1.064 1.002 1.062 1.044 1.018 0.999 1.011 1.042 0.991
2005 1.069 1.019 1.049 1.034 1.014 1.000 1.016 1.061 0.960
Geometric Means                  
1991-2005 1.050 1.016 1.033 1.032 1.002 1.000 1.008 1.027 0.997
1991-1999 1.042 1.017 1.025 1.027 0.999 1.000 1.005 1.021 1.000
2000-2005 1.061 1.015 1.045 1.039 1.006 1.000 1.012 1.035 0.992

Notes

  • [23]To see how these equations are calculated please see Fox, Kohli and Warren (2003).
  • [24]To calculate percentage growth figures from the index numbers from Appendix Table 2 above, subtract one and multiply by one hundred.
Page top