5  Discussion and Summary

The evidence that external economic links are imperfect substitutes for internal economic links raises a variety of questions for a small economy like New Zealand, for it is likely that New Zealand residents obtain fewer of the benefits of both specialisation and diversity that most people gain from internal trade. These effects of home bias are compounded by New Zealand’s remoteness, for international trade is an expensive alternative. Indeed, New Zealand has relatively little trade for a small country; consequently, compared to other geographic regions, New Zealand is rather insular.

The imperfect substitutability of external and internal links has spurred many small countries to adopt policies aimed explicitly at enhancing international linkages. New Zealand is no stranger to this process, as is evident from the Closer Economic Relations agreement with Australia and a more general strategy of reducing protection. Nonetheless, it is probable that more needs to be done if New Zealanders consider that there are benefits from more intensive trade in goods, services, and financial products. Such steps could include the pursuit of less restrictive multilateral migration regulations, widening the CER agreement to include other countries, taking further steps to harmonise business law, or forming a monetary union with another country. The Maastricht Treaty (1992) and the subsequent European Monetary Union are highly visible statements of a group of countries adopting several of these policy options in order to intensify economic and social ties.

While the evidence is clear that economic integration is much greater within countries than it is between countries, the extent to which adopting another country’s currency will enhance economic integration is much less obvious. The evidence does indicate that pricing in a single currency reduces price differences across space. For instance, from 1966 to 1996 the standard deviation of the real exchange rate movements between New Zealand and Australia was a factor of ten higher than the average standard deviation of real exchange rate movements between different Australian states. Distinct currency zones may also have a more subtle effect on prices, by demarcating the areas where different marketing operations prevail. This effect is harder to disentangle from straight border effects, as currency zones and national borders ordinarily coincide. Nonetheless, if firms have different marketing and distribution strategies in different regions because of different currency zones (which is plausible if firms find that exchange rate volatility drives a wedge between actual prices and desired prices in different regions), the use of a single currency will enhance goods market integration.

While it is becoming increasing clear from trade volume data that goods markets within countries are considerably more integrated than goods markets between countries, the effect of currency zones on trade volumes is not so apparent. To date the literature examining the effect of exchange rate volatility on inter-country trade flows has been inconclusive, although by and large it has not addressed the most relevant question which is whether long term exchange rate uncertainty has a detrimental effect on trade. The extent of this home trade bias is sufficiently large that even if only a small fraction of home bias is caused by exchange rate issues, the costs for a small country like New Zealand could be non-trivial. To date, however, it has not proved easy to make estimates of these benefits.

There is somewhat stronger evidence that capital market integration is affected by the choice of exchange rate regime as well as national borders. In most countries, internal capital markets are much more integrated than external capital markets; capital markets also appear to have been more integrated under the pre-1914 Gold Standard than they are now. This makes sense if agents are risk averse, for they will want to hold a large quantity of their debt instruments in local currency since most of their financial contracts will be denominated in local currencies. It is also true that financial intermediaries offer lower cost contracts denominated in local currencies, reinforcing the bias towards capital market integration within a currency zone.

The greater extent of capital market integration within currency zones rather than between currency zones means that maintaining a separate currency may have quite high welfare costs, as it limits the extent to which risk sharing occurs across regions. These costs are likely to be higher for small countries than for large countries because of the low degree of risk diversification available in a small country. As related phenomena, investment and saving decisions are also more highly correlated within currency zones than between currency zones.

There will also be a cost for a small net debt country if lenders impose a currency risk premium rather than a geographic risk premium as part of the cost of loans. This is quite plausible, as geographic risk premiums (within currency zones) seem to be smaller than currency risk premiums. New Zealand dollar interest rates are usually higher than those prevailing in other currencies, and if part of this premium is due to currency risk rather than country risk, forming a monetary union with another country could lead to a substantial reduction in the cost of capital.

These benefits of forming a monetary union have become better quantified in the last decade, and will become more so in the next as the effects of European countries forming a monetary union are observed. Nonetheless, there are potential costs to forming a monetary union, and these need to be taken into account before a decision to form a monetary union is made. The relative size of the costs and benefits of New Zealand forming a monetary union with another country is an empirical matter, with the calculus depending on the choice of the other country. The traditional literature analysing the optimal currency area provides some guidance on how some elements of the equation should be calculated, although there still remains considerable debate. Previous authors considering the costs and benefits of New Zealand forming a monetary union with Australia decided that the costs were too high. Recent developments in theory suggest that these costs have been overstated, however.

There are two potential advantages of New Zealand maintaining a separate currency and independent monetary policy. First, it allows New Zealand politicians and officials to choose their own inflation rate. This is most likely to be an advantage when the inflation rate implicit in choosing the currency of another country is considered to be too high for internal objectives. Recent developments in central bank technology have resulted in low inflation around the world, however, with many central banks increasing their political independence and adopting inflation targets, similar to the changes undertaken at the Reserve Bank of New Zealand in 1990. Inflation has fallen in most such countries since this time, and consequently inflation outcomes in New Zealand have been comparable to those in the rest of the world (including Australia). Average annual consumer price inflation in New Zealand was 2.1 per cent from 1990 – 1997, compared with 3.1 per cent in the United States, 2.4 per cent in Australia, 1.4 per cent in Japan, 2.9 per cent in Germany, and 3.8 per cent in the United Kingdom. In addition, most of these countries have a better long term inflation record than New Zealand.[41]

A second potential cause of high inflation would occur if productivity levels in New Zealand increased much faster than in the rest of a monetary union. This has been the Hong Kong experience since 1983 due to exceptional productivity growth. Where relative productivity increase has been gradual, however, the higher rate of price increase has not been particularly large, as demonstrated by the Irish experience. In this situation it is not the case that inflation is a problem because it causes low income growth. Rather, inflation is a problem because of its effect on income distribution, in particular disadvantaging those who like to hold their wealth in debt instruments. If this is the prime reason for choosing an independent monetary policy, the Government should ensure that other aspects of its fiscal and monetary policies are consistent with this distribution objective.

The second reason for choosing an independent monetary policy is that it is a useful tool for macroeconomic stabilisation. This is clearly the focus of most of the theoretical and empirical literature, and would appear to be the reason offered by many of the world’s policy makers. Two comments can be made on the basis of the recent literature.

First, recent theoretical developments, discussed in section 4, show that is not necessarily the case that maintaining a separate currency will help stabilise an economy. An independent exchange rate assists the process of stabilisation in response to real economic shocks if local wages and price are sticky, for then exchange rate changes can assist with the rapid adjustment of prices and wages relative to those in other countries. However, if exchange rates are excessively volatile, in that their value deviates from fundamentals in the short and medium term, the exchange rate will be a cause of shocks which destabilises the economy. Whether or not a flexible exchange rate is stabilising is an empirical matter; there should be no presumption that it is stabilising given the international evidence that exchange rates (and other asset prices) seem to be more volatile than justified by movements in macroeconomic fundamentals.

Secondly, the need for macroeconomic stabilisation is an empirical matter that depends on the nature of the shocks affecting a geographic region. For New Zealand not to join a currency union on this basis would require that New Zealand is buffeted by quite different shocks than regions within the other country. This is an empirical matter – but as Frankel and Rose (1998) make clear, one subject to the Lucas critique, as the shocks are likely to be endogenous to the exchange regime. Joining a monetary union would most likely cause an expansion of trade with that currency bloc; if so, the nature of the real shocks affecting New Zealand will change, as will the nominal shocks affecting New Zealand. In short, forming a monetary union with another country may reduce the need to have a separate monetary policy.

New Zealand is the smallest OECD country to have a fully independent monetary policy. To continue to justify this stance on an economic basis, there should be evidence that monetary independence has been beneficial - or, at least, of little cost. Over the last eight years, New Zealand has enjoyed very low rates of inflation, although its record is not dissimilar to that of most other OECD countries. Despite this low inflation, however, real short term interest rates have been higher than in our main trading partners; the New Zealand economy has not been noticeably more stable than other OECD economies; and trade volumes have grown only slowly despite having a trade share substantially below that of almost all other small OECD economies. While the counterfactual is not available, there should be no presumption from these outcomes that monetary independence should be the natural option for the New Zealand economy.

Other Issues

A smorgasbord of issues remain to be discussed. First, it is quite apparent that many of the recent advances in the literature stem from an analysis of intra-regional economics, not international economics. There is a growing literature examining the behaviour of regional economies within countries, and regional economies within larger regional groupings. This research examines amongst other things the behaviour of labour markets and goods markets within and between regions, inter-regional migration, the different types of shocks buffeting different regions, and the extent of trade between regions. To date there is little knowledge of the performance of the different sub-regions of the New Zealand economy, or how New Zealand-Australia interactions compare to the interactions of different Australian states with each other.[42] More research on this topic is likely to be useful in understanding the performance of the New Zealand economy less as an independent entity and more as a part of a wider regional group; it also may shed some light on the effects of one region having a different currency from the other regions.

Secondly, even if there are few good reasons for New Zealand having an independent currency, there is still likely to be a demand for New Zealand specific financial products that help New Zealanders save, invest, and risk-share. If New Zealand were to join a currency union, the Reserve Bank could consider introducing financial instruments or even currencies whose payment structure would be tied to the performance of various New Zealand financial statistics, but which would not be used as a unit of account or medium of account for every day transactions.[43] This would provide an additional avenue for New Zealanders to diversify their risk given a world of incomplete financial markets.

Thirdly, this paper has not discussed the difference between forming a monetary union with another country, and simply adopting the currency of another country. If the former path were adopted, New Zealand economic conditions would be taken into account when setting monetary conditions across the union; if another currency were adopted, the monetary policy of the partner country would be determined by its domestic considerations alone. In practice, New Zealand is such a small economy that the difference is likely to be of little economic consequence, because New Zealand considerations would have little weight in a monetary union. There could be a substantially different treatment of seignorage revenue, however, and this would be of economic significance.

Fourthly, this paper has not discussed transition issues. Joining a monetary union involves various costs, and it could be the case that these transitional costs outweigh any positive effects once the transition is over. These costs not only involve the direct costs involved with switching monetary units, but involve the risk that a bad shock hits the country while economic integration is only partial.

Finally, if New Zealand were to seriously consider adopting the currency of another country and giving up its monetary policy, the most obvious question concerns the choice of country. The choice is vital, as the benefits of greater goods market and capital market integration would primarily stem from greater integration with that country. The two most obvious choices are Australia and the United States, for these countries are relatively close and impose relatively few border or linguistic barriers to an expansion of trade. The former is New Zealand’s closest neighbour and main trading partner, particularly for manufactured goods and for services; given that New Zealand also has open capital and labour markets with Australia it would be the leading choice. There are already very close links between many New Zealand and Australian firms, particularly in the financial sector, reducing the transaction costs of any link. The United States has the advantage of being the world’s largest economy, and thus a considerably larger economy from which to gain the benefits of greater integration than Australia, as well as the currency partner of other countries such as Hong Kong. While goods market integration and labour market access are substantially less than with Australia, the benefits from greater capital market integration would be much greater. Unless Australia were to simultaneously adopt the U.S. dollar with New Zealand, it is a choice that will not be easy to make.