Journal Issue
Share
Article

New Zealand: Selected Issues

Author(s):
International Monetary Fund. Asia and Pacific Dept
Published Date:
February 2016
Share
  • ShareShare
Show Summary Details

Prospects for Potential Growth in New Zealand1

New Zealand has world class institutions and strong policy framework, but income levels remain low relative to OECD peers. Boosting potential growth and income levels would require strengthening incentives to invest in productive assets, including through measures to raise domestic savings, enhance innovation and skills, boost infrastructure, and increase competition to mitigate the impact of distance on market size.

A. Introduction

1. New Zealand’s income level per capita remains lower than the average of OECD countries. Over 1980-2014, per capita income levels have remained about 20 percent below the OECD average. In Purchasing Power Parity (PPP) terms, the level is around 40 percent below the maximum OECD income level.

2. This reflects low labor productivity (LP) levels, linked to both low levels of total factor productivity (TFP) and of capital deepening. Assuming a standard production function, LP levels can be decomposed into TFP and capital deepening levels.2 OECD data show that the LP level is below average OECD levels, reflecting both TFP and capital deepening (measured as the ratio of capital to labor) being at or below the average.

3. Growth of LP (measured as output per worker), has been relatively low despite the income level gap. The relatively modest pace of LP growth has left the sizeable relative income gap broadly unchanged over time. The average growth rate of LP of 1.1 percent over the last 25+ years is below the OECD average of 1.3 percent. The difference in LP growth rates relative to high income OECD countries is quite large, with the third quartile growth rate at 1.7 percent. In terms of PPP income per capita, New Zealand’s growth over this period was the same as the OECD average at 4 percent, but little higher than the 3% percent for the highest income OECD countries.

Income Gap Relative to OECD Has Persisted

(%, PPP per capita)

Sources: IMF WEO database, and IMF staff calculations.

…And Average TFP Levels

(Index, 1994 = 100)

Sources: OECD Productivity database; and IMF staff calculations.

GDP Per Worker Has Remained At The Low End of OECD

(Index, 1984 = 100; 5-year moving average)

Sources: OECD Productivity database; and IMF staff calculations.

4. Moreover, LP growth has slowed since the GFC.

A historical decomposition of LP growth into its components reveals over 1997-2014, the measured sector3 recorded average LP growth of 1.5 percent. Compared to precrisis growth rate of 1.8 percent over 1997-2008, LP growth has declined to 1.2 percent over 2011-2014. This reflects a slowing in the contribution of capital deepening from 0.8 to 0.4 percent, and of TFP from 1.0 to 0.8 percent over the same time period.

5. The slowdown in capital deepening occurred even as labor input growth declined. Capital input growth has averaged 2 percent since 2011 compared to 3.5 percent precrisis (though the growth rate has been rising in recent years). Meanwhile labor input growth fell from 1.6 percent over 1997-2008 to 1 percent over 2011-2014.

6. Favorable terms of trade have supported incomes, but commodity prices have recently weakened considerably. Gross domestic income per capita grew at 2.2 percent on average over 1992-2014, compared to output per worker growth of around 1 percent. If commodity prices stabilize at lower than historical levels, or continue to decline over the medium term, this beneficial boost to income will not be sustained, necessitating higher GDP growth to sustain and increase growth rates of income.

Reflecting Moderate Capital Deepening…

(Index, 1994 = 100)

Sources: OECD Productivity database; and IMF staff calculations.

Decomposition of Output Per Unit of Labour Growth

(% change; March years)

Source: Statistics New Zealand.

Decom position of Output Growth

(% change; March years)

Source: Statistics New Zealand.

7. Measures to boost potential growth and incomes would require addressing the long-standing structural issues. These include addressing the savings-investment imbalance, which may be raising the cost of capital and discouraging a faster rate of capital growth, and addressing issues related to incentives for investment in directly productive assets as opposed to relatively high preference for housing. While New Zealand has high levels of competitiveness based on global rankings such as Doing Business when compared to other advanced economies, boosting productivity growth has yet proven challenging. Policy measures will need to help mitigate the limiting effects of distance from markets, boost innovation and competition, and address any infrastructure bottlenecks.

B. Estimates of Potential Growth

8. A number of approaches are employed below to establish the range of plausible estimates. Estimates of potential growth are inherently subject to uncertainty, with different methods having different strengths and weaknesses. Methods like HP filtering and the Okun approach are simple to implement but are backward looking, and the HP filter is subject to known statistical problems. The production function approach is forward looking, but requires assumptions/projections of the evolution of inputs and productivity that may be subject to substantial uncertainty. Using a number of methods helps to establish a plausible range for potential growth.

9. Taking an Okun view, real GDP growth that appears to be consistent with stable unemployment is around 2.7 percent. This is based on a sample over 1987Q2:2015Q2. Over this period, unemployment growth rose in the early 1990s, and then declined steadily over the 2000s reaching a low of 3½ - 4 percent prior to the GFC. Since then, the unemployment rate has risen to stabilize around 6 percent. Over this period, GDP growth has fluctuated around the average rate of 2.7 percent.4

Okun Coefficient

Haver; and IMF staff calculations.

HP-Filtered GDP Growth

(Q/Q annualized percent change)

Haver; and IMF staff calculations.

10. A standard HP filter suggests that trend growth has hovered around 2.4 percent since 2013, recovering in the post-GFC period, but slower than rates observed in the mid-1990s to mid-2000s.

11. A production function approach suggests potential growth around 2.5 percent. Table 1 summarizes the baseline results using the production function approach. Potential growth is estimated to average 2.5 percent over the medium term.5 The contribution of TFP growth is assumed to rise gradually over the medium term to the long run average. Weakening WAP growth translates into a declining contribution from labor input growth, while the contribution of capital input is assumed to remain steady at around 1¼ percent.

Table 1.Potential Growth Estimates
20162017201820192020
Potential growth2.62.52.52.52.5
TFP0.50.50.60.70.7
Labor (contribution)1.00.80.70.60.6
Capital (contribution)1.21.21.21.21.2
Labor input growth1.71.41.11.01.0
Capital input growth3.03.03.03.03.0
WAP growth2.01.21.30.80.8
LFPR68.968.968.968.968.9
Unemployment5.75.65.55.45.2
Sources: Statistics New Zealand, and IMF staff estimates.
Sources: Statistics New Zealand, and IMF staff estimates.

12. The assumptions underlying the production function estimates are follows.

Multifactor Productivity Growth

(%, average)

Source: OECD Productivity Database.

Capital Input Growth

(%)

Source: Statistics New Zealand.

  • TFP. TFP growth has slowed over the last complete cycle – as in several other OECD economies – relative to higher growth rates in the past. Over 2003-07 (the last complete cycle), TFP growth averaged 0.4 percent, compared to around 1 percent over 1995-2003, and 0.4 percent over 2008-12. The long run average is 0.6 percent over 1995-2012. Compared with other OECD countries, New Zealand’s TFP growth performance compares favorably, particularly considering large negative growth rates recorded among several countries in the period encompassing the GFC. In our baseline we assume that TFP growth averages 0.6 percent, rising gradually from 0.5 percent (as recorded in 2014 by Statistics New Zealand) to 0.7 percent over the medium term.6

  • Capital input. Trend growth in capital input (HP filtered) averages around 3 percent in New Zealand, over 1997 - 2014. Capital input growth slowed after the GFC, but has since recovered to nearly 3 percent. For baseline estimates of potential growth we assume that capital stock growth averages around 3 percent over the medium term in line with its historical growth rate. The growth of capital stock over a longer period (1950 – 2011) has averaged 3.4 percent, but closer to 3 percent over 1970-2011.7

  • Labor input. Projected labor input growth may be decomposed into growth in working age population (WAP), labor force participation rates (LFPR), and the underlying rate of unemployment (NAIRU).

    • Average unemployment over 2000Q1-2015Q1 is 5¼ percent. The baseline estimates assume that unemployment declines from 5.7 percent (recorded over 2014) gradually to this average rate.

    • LFP rates have increased markedly since 2013Q1, reflecting a strong increase in net migration. Given the difficulties in projecting migration flows over the medium term, a simplifying assumption is made that LFPR stays constant at nearly 69 percent recorded in 2014 (though it has risen slightly above this figure in 2015).

    • WAP growth assumptions are derived from Statistics New Zealand National Population Projections (base 2014) data. Of the several population growth (by age cohort) scenarios in this database, the baseline scenario is based on median fertility and mortality projections, and cyclic migration that fluctuates over a 10-year cycle between -10,000 and 35,000 each year over the projection period.8

Unemployment Rate

(%)

Sources: Statistics New Zealand; and IMF staff calculations.

Rising net migration has boosted participation rates

(No. of persons, and in %)

Source: Statistics New Zealand.

13. There are significant risks to this baseline scenario. Given the known difficulties is projection TFP growth, and uncertainties around the labor input projections, particularly around estimates of NAIRU, difficulties in projecting migration, and hence in projection WAP and LFPR, the estimates are subject to uncertainty. Some risks could push potential growth above or below the 2.5 percent mark:

  • Upside risks may arise for instance due stronger than historical net migration into New Zealand. For example, Statistics New Zealand’s “very high migration” scenario for population growth would yield potential growth estimates higher by about ¼ of a percentage point than the baseline in the outer years of the projection.9

  • On the downside, absence of the gradual improvement in TFP growth would take off about ¼ percentage point from growth - a scenario which may be triggered by the general slowdown in productivity growth among advanced economies. Trend capital stock growth has been slowing; continued weaker capital stock growth of around 2-2¼ percent as estimated for recent periods would also subtract ¼ - ¼ percentage point from growth. LFPR rates are at historical highs. A 1 percentage point reduction in LFPR over the medium term relative to baseline would reduce potential growth to 2¼ percent. A combination of weaker TFP growth and weaker capital stock growth would imply potential growth around 2 percent over the medium term.

C. Boosting Living Standards: Productivity and Capital Deepening

14. Raising productivity levels in New Zealand is important to raise income levels and address the savings-investment imbalance. New Zealand has very high LFPR rates and is the second most intensive in terms of hours worked among major OECD economies, which has been declining over time like in the rest of the OECD. Future growth would have to rely on greater capital deepening and relatedly, on higher productivity, the ultimate driver of living standards.

15. New Zealand implemented far-reaching economic reforms in the late 80s and early 90s which boosted productivity. These reforms encompassed macroeconomic policy frameworks (adoption of a floating exchange rate and inflation targeting, implementation of the Fiscal Responsibility Act and comprehensive tax reforms generally moving to broad based and flat rates, and extensive privatization), trade liberalization, and labor market reform (moving to individual contracts).10 Evidence on TFP growth suggests that the post-reform decade over 1990-2000 saw a marked pick-up. Subsequently TFP growth has fallen off, particularly sharply after the GFC.

Actual Hours Worked Per Worker

(Annual average, 2014)

Source: OECD.Stat

Multifactor Productivity

(Index 1978 = 100, and growth in annual % change)

Source: Statistics New Zealand.

1/ Former market sector.

16. There are few if any low hanging fruit in terms of reforms. With generally high quality policy settings, globally competitive institutional features, there are few if any major economic reforms of the type implemented in the 1980s and 1990s left. The World Bank Doing Business indicators show New Zealand was the 2nd best country in overall ranking in 2015, and occupies the frontier among several dimensions.

New Zealand remains among the best places to do business…

(Distance from frontier; 100 = frontier 1/)

1/ Source: World Bank Doing Business database. Country sample includes Austria, Australia, Canada, Denmark, France, Germany, Italy, Japan, Netherlands, New Zealand, Norway, Spain, Sweden, U.K., and U.S.

17. However, there may be targeted areas for improvement. According to the World Economic Forum’s Global Competitiveness Report (2014-2015), though New Zealand ranks high at 17 in overall competitiveness (up from 25 in 2011-12), and the top spot in the subcomponent of institutions, it ranks much lower in terms of infrastructure, innovation and sophistication factors, and technological readiness. The top constraints cited by business in New Zealand include infrastructure, innovation capacity, access to finance, and workforce skills.

18. New Zealand is handicapped by its long distance from markets and its small market size, limiting incomes and productivity. Based on structural policy settings, New Zealand’s per capita income should be 20 percent above the OECD average, not 20 percent below.11 Distance from markets can impact on productivity through limits on the potential to exploit scale economies and agglomeration effects, higher transport and distribution costs, etc. Large distance to markets may be suppressing New Zealand’s trade intensity, may explain a sizeable portion of the productivity gap (more than half), and has been estimated to cost around 12 percent of GDP relative to other developed countries.12 Other narratives have ascribed weak productivity to slowing pace of economic reforms, and to the nexus of low savings and high relative interest rates, and consequently low investment, productivity and exports.13

Low Market Access May Be Suppressing Trade Intensity

Sources: Baulhol and de Serres (2010) for index of market access; IMF WEO database; and IMF staff calculations.Trade intensity is definedas sum of exports and imports over nominal GDP in percent.

19. Increasing New Zealand’s international exposure is a major aspect of productivity oriented reforms. While physical distance is immutable, its impact can be lessened with policies to encourage a greater competitive and outward-oriented focus for New Zealand firms. Some of the key drivers of productivity growth that are considered important areas to address are shown in Table 2, as detailed in a series of Treasury papers investigating productivity growth options.

Table 2.Policy Considerations for Boosting Key Productivity Drivers
EnterprisePromote competition, including with greater integration of New Zealand markets with international goods and factor markets. Domestic policy settings may have to keep an international perspective; given few obvious barriers to trade, capital flows, etc.
InnovationIncrease R&D expenditure, which still lags behind OECD levels (0.5 percent of GDP compared to 1.5 percent OECD average in 2006-7). Promote international linkage to widen beyond domestically sourced innovation to increase absorption. Leverage public sector innovation with stronger market-facing focus and responsiveness to firms' needs.
SkillsImproving educational outcomes of disadvantaged groups, increasing completion rates in secondary and tertiary levels, and enhancing market-facing skills to match firms' needs.
InvestmentEnhance integration with global markets to increase opportunities for innovation.
Natural resourcesBeing a natural resource dependent economy, ensuring sustainable and productive use of natural resources balancing environmental considerations with economic benefits. In this regard, clarifying the role and relationships of central and regional governments.
Sources: Various Treasury Productivity Papers; 2008.
Sources: Various Treasury Productivity Papers; 2008.
  • A key aspect of the policy considerations to promote drivers of productivity growth is greater integration with and exposure to global markets. This would help enhance competition, and provide opportunities for more investment by increasing the size of markets.

  • A second aspect is policies to directly enhance innovation through greater expenditure on R&D, which is appears on the low side compared to the OECD (though the data only refer to 2006-07), and increasing the market interface of public research. Innovation would in turn help to promote investment.

  • A third aspect is the emphasis on increasing labor productivity through education, particularly in raising the attainments of hitherto disadvantaged groups of the population, though gains have been made in education in recent times.

20. Addressing infrastructure bottlenecks would help boost competitiveness. Measures include (i) implementing infrastructure demand management strategies to reduce urban road congestion (ii) diversifying infrastructure funding, including by expanding user charges; (iii) more frequently update immigration targets and skill shortage categories to reduce labor market bottlenecks.14

21. There is space for TFP to catch up to the frontier in several sectors. Based on Groningen Growth and Development Centre (GGDC) data on sectoral TFP levels (which does not include New Zealand and only extends to 2007), and NZStat data on sectoral TFP in New Zealand (which extends up to 2013), it is possible to compare TFP levels relative to the 2007 frontier, and the extent to which the gap has closed or widened. While somewhat dated (the frontier may have moved since 2007), this may still be useful to identify potential sectors for further targeted reforms.

  • In 2007, TFP levels in New Zealand were well distant from the frontier in a number of sectors, including mining, manufacturing, hotels and restaurants, utilities, and construction. By contrast it was at or very near the frontier in financial intermediation and business services, distribution and communication, and was the foremost in agriculture.

  • By 2013, the gap from the frontier in 2007 narrowed somewhat in construction, distribution and communication, and finance and business services. Levels remained broadly unchanged relative to the frontier in other sectors. Overall, sizeable TFP level gaps remain across most sectors barring agriculture, and distribution and communication, in 2013.15

  • Bridging these productivity level gaps can have a sizeable impact on the level of income. Closing the sectoral TFP level gaps would imply an increase in the level of aggregate value added (GDP) by 19 percent.16 This would serve to bridge a large part of the income gap relative to higher income OECD countries.

Sectoral TFP Levels Relative to Frontier

Sources: Groningen Growth and DevelopmentCentre Productivity Level Database (1997 benchmark); Statistics New Zealand; and author's calculations.

Note: Frontier refers to the 2007 TFP level in the leading country.

Sectoral TFP Level Gaps Relative to Frontier

Note: Gaps measured off 2013 TFP level in New Zealand and 2007 levels in frontier countries.

Sources: Groningen Growth and Development Centre Productivity Level Database (1997 benchmark); Statistics New Zealand; and author's calculations.

22. Targeted reforms in the services sector could be usefully implemented. The Productivity Commission has identified enhancing competition in services, and boosting investment in ICT as areas that could yield sizeable productivity gains. The degree of competition is relatively weak in finance and insurance, rental, hiring and real estate, retail, and in professional, scientific, and technical services. ICT investment per capita is also lower than in other advanced economy peers. Measures to increase competition include further lowering barriers to trade, reducing search and switching costs for consumers, and improving the current competition law. ICT adoption is hindered among other factors by restrictions of market size and distance, but also due to shortage of IT-skilled workers. Greater emphasis on job-ready tertiary education, and improved workforce management practices would help to increase the take-up of ICT technology by firms.

Nominal rates and S-I balance

Notes: 1. Sample includes Australia, Austria, Canada, Denmark, France, Germany, Italy, Japan, Netherlands, New Zealand, Norway, Sweden, Spain, U.K., and U.S. 2. Average over 1970 - 2014.

Sources: IMF WEO Database, and Haver.

23. Increasing capital deepening – also relatively low by OECD standards – will also help boost living standards. A high cost of capital would reduce incentives to increase investment, other things equal, and lead to lower capital intensity. Across developed countries, long term interest rates appear to be related to the size of the S-I balance – wider gaps tend to be associated with higher interest rates. Compared to other OECD countries, New Zealand has experienced relatively low rates of savings, while investment rates are not particularly high. Partly as a result of the wide S-I imbalance, interest rates in New Zealand have prevailed higher. The higher cost of capital in New Zealand may be driven not only by the low savings rate, but also by factors including the large existing external debt stock (requiring a higher premium for additional borrowing), and a banking dominated financial system with relatively shallow capital markets, combining to depress levels of capital intensity.17

Nominal rates and saving/GDP

Notes: 1. Sample includes Australia, Austria, Canada, Denmark, France, Germany, Italy, Japan, Netherlands, New Zealand, Norway, Sweden, Spain, U.K., and U.S. 2. Average over 1970 - 2014.

Sources: IMF WEO Database, and Haver.

Real rates and investment/GDP

Notes: 1. Sample includes Australia, Austria, Canada, Denmark, France, Germany, Italy, Japan, Netherlands, NewZealand, Norway, Sweden, Spain, U.K., and U.S. 2. Average over 1970 - 2014.

Sources: IMF WEO Database, and Haver.

24. Maintaining a strong pace of investment without running up ever larger external debt at high interest rates requires an increase in domestic savings. The envisaged path of fiscal consolidation is a helpful policy to raise national savings, reversing the deterioration in fiscal balances following the GFC. Household savings have also increased over the past few years after declining steadily from 1998. At around 2% of GDP, however, the rate of savings is much lower than that of other leading OECD countries, which average closer to 6 percent (with some including France, Germany, and Sweden saving closer to 10 percent of household disposable income). Even in Australia, household savings rate has risen markedly following the GFC, prevailing around 8 percent.

25. Tax policies may also have an impact on the form of savings and investment. New Zealand households appear to show a greater preference for saving in housing rather in other assets that may fund directly productive capital.18 New Zealand’s tax system may favor investment in housing assets. The returns on owner-occupied housing are not taxed at all (though interest payments on mortgages are not tax-deductible), while returns on rental housing are effectively taxed more lightly than those on financial assets, including shares, debt instruments, and bank accounts. Boosting savings may require increasing uptake of Kiwisaver including through incentives or through compulsory participation, and other reforms to the tax system more widely that may help boost domestic savings and reduce incentives to invest in housing.19

26. In conclusion, policy efforts to boost living standards have to address multiple fronts.

With no obvious liberalization policies at hand, efforts would have to be made to exploit opportunities for greater international integration in order to boost competitiveness and overcome the disadvantage of distance. Lifting productivity levels would also help boost incentives to invest; but to do so sustainably without exacerbating the high external debt level requires efforts to increase domestic savings. New Zealand’s world class institutions and strong macroeconomic track record have supported high debt levels, but the New Zealand economy is not immune to adverse events globally. Thus, pursuing sustainable improvements in living standards would also help reduce vulnerabilities and increase the economy’s resilience to future adverse events.

Households’ Net Savings

(% household disposable income)

Sample includes Australia, Austria, Canada, Denmark, France, Germany, Italy, Japan, Netherlands, Norway, Sweden, Spain, U.K., and U.S.

Sources: OECD Householdaccountsdatabase;and Haver.

References

Prepared by Adil Mohommad.

This is a useful way to organize a discussion on per capita income in New Zealand into factors that have bearing on TFP and those that bear on capital deepening, though these factors are interconnected, as productivity both benefits from, and incentivizes, investment in capital. See “An Empirical Study of Sectoral Level Capital Investments in New Zealand,” Treasury WP 14/04.

The measured sector excludes property and business services, health and community services, cultural and recreational services, and personal and other community services, providing a coverage of about three-quarters of the economy.

This is in line with a recent Treasury estimate; see Treasury Working Paper 14/01.

These results are in line with Treasury estimates of potential growth (see 2012 Half Yearly Economic and Fiscal Update, and steady growth assumptions in Treasury Working Paper 13/18).

The data for cross-country comparison use OECD statistics for the sake of comparability. Statistics New Zealand’s estimates produce a similar long run average TFP growth of 0.7 percent (over 1997-2012), and indicate that the slightly different cycle over 2003-08 produced average TFP growth of 0.6 percent, and 0 percent over 2009-12 (an incomplete cycle).

Federal Reserve of Saint Louis data (sourced from UC Davis and University of Groningen).

This assumption appears plausible when compared to historical annual net migration, except in 2014 when net migration exceeded 51,000. While net migration appears to have peaked since then, this is subject to uncertainty.

In the high migration scenario, population growth in the 15-39 age group is 1.3 percent per year over a 5-year period (2018-2023), compared to 0.4 percent per year in the baseline case, and slightly higher than the baseline in the 40-64 cohort (with no difference in 65+ cohort). Based on the share of these cohorts in the projected WAP size, and assuming a labor coefficient of 0.6, the additional growth amounts to about ¼ percent.

See “New Zealand’s Radical Reforms,” OECD 1997.

See “An International Perspective on New Zealand’s Productivity Paradox,” New Zealand Productivity Commission Working Paper 2014/01.

See “Have developed countries escaped the curse of distance?,” Journal of Economic Geography 2010 (10).

See “Holding on and Letting Go,” Treasury 2014.

OECD 2015 Economic Review – New Zealand.

Qualitatively similar results are found by Steenkamp (2015); average TFP levels over 2000-10 lag U.S. levels substantially in most sectors, including mining and agriculture, albeit with methodological differences in the measurement of capital stock.

The estimate varies depending on whether the TFP levels in a given sector and country are weighted by its share in value added in that country, or not. Measuring the gap without weighting in the above manner produces a smaller estimated impact on aggregate value added (11 percent).

According to recent estimates, capital intensity in New Zealand is around 60 percent that of Australia (in 2009; see Mason 2013), and 50 percent the levels in the U.S. (in 2005; see “Investment, Productivity and Cost of Capital: Understanding New Zealand’s “Capital Shallowness”,” Treasury Productivity Paper 08/03). A good description of the impact of New Zealand’s savings-investment imbalance on interest rates and investment (and the exchange rate) and can be found in Reddell (2013).

It has been estimated that almost 90 percent of household wealth is held in the form of housing in New Zealand, compared to between half and three-quarters of total household wealth among other OECD countries (2007 figures). See “Savings in New Zealand – Issues and Options” (Treasury September 2010).

See Chapter on “New Zealand: Options for Tax Policy Reform”.

Other Resources Citing This Publication