New Zealand's international competitiveness challenges and the Woody Allen syndrome

Abstract

The purpose of this paper is to investigate the (relative) roles of two sources of reduced international competitiveness in New Zealand during the last few years: the appreciation of the trade-weighted real exchange rate, and the (relative) inefficiency of the local service sector. According to the exercises reported in this paper, reducing services' inefficiencies is a more effective way of improving competitiveness, broadly defined, than engineering (or hoping for) a real exchange rate depreciation.

Keywords:

• real exchange rates
• New Zealand
• competitiveness
• services sector
• overvaluation
• policy mix

Acknowledgements

This is a revised version of a paper prepared for the policy forum ‘New Zealand's Macroeconomic Imbalances: Causes and Remedies', 23--24 June 2011, Wellington, NZ. I am indebted to Mark Blackmore, Anne-Marie Brook, Jean-Pierre Andre, Grant Scobie, Tim Hampton, John Janssen, Paul Dyer, Stephen Toplis, Arthur Grimes, Murray Sherwin, Kerry McDonald, Geoff Bertram, Bill Hampton, John Morris, Matthew Haigh, Colin Lynch, Bill Rosenberg, Anella Munro, Christie Smith, Shamubeel Eaqub, Kirdan Lees, Daan Steenkamp, Phil Briggs, Bob Buckle, David Hargreaves, Bernard Hodgetts, Norman Gemmell, Simon McLoughlin, Brian Easton, Bill Moran, Dennis Rose, Gary Blick, John McDermott, and Michael Reddell for very helpful discussions. I thank Viv Hall and two reviewers for helpful and detailed comments and suggestions.

Notes

 1. This paper takes a short-term, forward-looking approach. A number of recent works, including the two reports by the 2025 Taskforce (2009, 2010) take a longer view perspective.

 2. See Ministry of Economic Development, The Treasury, and Statistics New Zealand (2011, p. 108).

 3. For comparisons with Australia see the two reports published to date by the 2025 Taskforce (2009, 2010). Also, see the joint report by Statistics New Zealand and the Treasury (2010).

 4. See Edwards (2011b) for a discussion of what I have called the ‘unofficial–official’ story.

 5. The ‘absorption’ of the large transfer of funds to finance reconstruction is likely to pose challenges to macroeconomic policy. In particular, it is likely to put upward pressure on the NZD. It is precisely for this reason that after the major 2010 earthquake, Chile implemented macro policies aimed at avoiding such absorption issues.

 6. In many analyses the terms ‘nontradables’ and ‘services’ are used interchangeably. This is not completely correct in the case of New Zealand, as a major service sector, tourism, has many of the characteristics of tradables.

 7. On the persistently large current account deficits in New Zealand see, for instance, Edwards (2006), and Munro and Sethi (2006).

 8. Interest rate differentials are computed as nominal interest rate differentials relative to the US. An interesting question is what is the steady-state level of interest rate differentials.

 9. On ‘collateral damage’, currency wars and the commodity countries, see Edwards (2011a).

10. Studies that use lower frequency data tend to include other ‘fundamentals’ that change rather slowly through time, including openness, productivity growth differentials, and the share of government expenditure. These data, however, are generally not available on a monthly frequency, and are not included in this paper. One way of dealing with their possible role is by including time trends and ‘year’ dummies. See the discussion below.

11. Of course, this is a reduced form, and as such it is subject to the usual caveats.

12. Cointegration tests also show that the log of the RER is cointegrated. The panel V-statistic is 4.35.

13. The role of commodity prices as drivers of RER is well documented. See, for example, Edwards (1990). This point has been made forcefully in recent work by Kenneth Rogoff and his associates, e.g. Chen and Rogoff (2003), and Chen, Rogoff, and Rossi (2010).

14. The carry trade directly affects nominal exchange rates. Movements in nominal rates are, precisely, the main cause of RER movements in the short and medium run.

15. In the discussion that follows I am excluding Canada, since the results in Table 1 indicate that its RER behaviour does not respond to the same forces as in the other nations.

16. OECD (2011).

17. Of course, one could think of a number of scenarios where a reduction in interest rates stemming from a Central Bank action would have undesirable effects on other important variables such as inflation. A general equilibrium model would, in principle, allow analysing the ramifications of such a policy.

18. The relative price ratio is calculated from sub-components of the Consumer Price Index published by Statistics NZ. Regression results suggest that, while the relative price of nontradables is negatively correlated (as expected) with the share of tradables in GDP, the RER is not. That is, while the relative price does behave like the slope of the production possibilities frontier in the Australian, tradable–nontradable model, the RER (in either version) does not.

19. Of course, as in the effective rates of protection literature, there is no need for the a_ij's to be fixed. They should be interpreted as the ‘optimal’ input–output coefficient. I have written the import tariff as being specific, purely for convenience.

20. Remember that, for now, I am ignoring currency issues; I am implicitly assuming a unitary exchange rate. Of course, if instead of specific import tariffs I use ad-valorem ones, the results are almost identical.

21. This is, of course, the general concept of the effective rate of protection. As will be seen below, the actual operational formula is different, as the ECI emphasizes

22. It should be noted that some authors have argued that the ERP concept lacks full general equilibrium properties.

23. On the treatment of nontradables in effective protection computations see the essays in Corden (1985).

24. In contrast with the tradables sector, where domestic prices will be related to international prices – and in some cases quite rigidly so – in the nontradable case production inefficiencies will be translated into higher prices.

25. Yang and Stephenson (2011) calculate labour productivity levels by sector in both New Zealand and Australia. Of course, b_ik is not exactly the inverse of labour productivity; indeed, it is a broader measure of productivity. However, the assumption that the gaps in productivity levels are similar for labour and nontradable productivity seems very reasonable, and will be used in the rest of this paper.

26. This assumes, without loss of generality, that prices of all goods are normalized to one. This assumption is usually made in the ERP literature.

27. It is important to emphasize that, while providing ‘protection’, import tariffs will result in misallocation of resources and economic costs. This, of course, is the point extensively made in the ERP literature.

28. Statistics New Zealand publishes fairly detailed data on productivity growth. See, http://www.stats.govt.nz/browse_for_stats/economic_indicators/productivity/productivitystatistics_mr7810.aspx