New Zealand: Selected Issues
Author: International Monetary Fund
Publication Date:
27 Mar 2002
eISBN:
9781451830231
Language:
English
Keywords:
ISCR; CR; exchange rate; price; pass-through; price change; exchange rate volatility; volatility in New Zealand; pass-through to consumer prices; stage pass-through; Consumer prices; Australia and New Zealand; long-term exchange rate pass-through
Download PDF (1.1 MB)

This Selected Issues paper presents an empirical comparison of New Zealand’s growth performance with that of Australia during the post-reform period. The paper shows that most of the divergence in income per capita between the two countries has been the result of lower accumulation of capital per hour worked, and to a lesser extent, lower efficiency in utilizing resources in New Zealand. The paper also examines how migration has affected the income and welfare of New Zealand nationals.

Abstract

This Selected Issues paper presents an empirical comparison of New Zealand’s growth performance with that of Australia during the post-reform period. The paper shows that most of the divergence in income per capita between the two countries has been the result of lower accumulation of capital per hour worked, and to a lesser extent, lower efficiency in utilizing resources in New Zealand. The paper also examines how migration has affected the income and welfare of New Zealand nationals.

III. Exchange Rate Pass-Through and Inflation in New Zealand56

A. Introduction and Summary

96. The main question motivating this paper is the extent to which fluctuations in exchange rates affect inflation. This topic has been subject to considerable debate and remains a major concern in the conduct of monetary policy. Over the past decade, most of the advanced economies have achieved a remarkable stabilization of inflation at very low rates despite wide fluctuations in exchange rates and import prices. While many competing explanations have been put forward—from favorable price shocks (such as that associated with the Asian crisis) to structural changes (such as information technology and inventory management)—Taylor (2000) argues that, in a low-inflation environment, firms tend to reduce the degree to which they “pass through” cost and price increases due to exchange rate movements. This hypothesis would suggest that domestic inflation may be permanently rather than transitorily insulated from fluctuations in exchange rates and foreign prices, which may have important consequences for monetary policy.

97. In line with the experience in large advanced economies, inflation in New Zealand has been low and relatively stable during the past decade. However, in contrast to the experience in some of the large advanced economies (mainly the United States), where exchange rate shocks were primarily disinflationary, exchange rate shocks in New Zealand seem to have been largely of an inflationary nature since 1997. For a small open economy like New Zealand, it is somewhat puzzling that such large changes in the exchange rate have not been passed on fully to domestic prices.

98. Several factors may have contributed to change the price-setting behavior:

  • First, the commitment to keep inflation low, together with the high credibility of the inflation targeting framework, may have induced firms to change their price-setting behavior. Inflation targeting limits firms’ “pricing power” by reducing permanently their room to accommodate changes in costs or by inducing them to pass these changes less quickly onto prices. As noted by Taylor, a low inflation environment would lower the perceived persistence of changes in costs, and therefore, limit the pass-through of changes in costs associated with exchange rate fluctuations.

  • Second, New Zealand underwent significant structural reforms during the late 1980s and early 1990s which may have heightened the uncertainty about the equilibrium level and growth rate of potential output and the equilibrium exchange rate. Increased uncertainty on the underlying strength in domestic economic conditions and the business cycle may have influenced firms’ decisions on profit margins and the extent of pass-through, as they may have become less inclined to pass-through changes in costs—especially those caused by exchange rate changes—onto prices.

  • Third, the volatility of the exchange rate and the nature of the exchange rate shocks seem important, as firms facing highly volatile exchange rates would likely tend to smooth the adjustment on domestic prices.

  • Fourth, increased competitive pressures in sectors (such as retail) may also have limited the ability of firms to pass-through these fluctuations.

  • Finally, the lack of exchange rate pass-through may have only been a “mirage” in the sense that firms may have been able to absorb changes in costs induced by an exchange rate depreciation because of cost savings arising from the use of new technology and productivity gains.

99. This paper explores empirically the role of supply, demand, and exchange rate shocks in driving fluctuations in inflation in New Zealand. In particular, the estimation aims at assessing these effects, especially those associated with exchange rate pass-through, at different stages of the distribution chain: import prices and consumer prices. The exchange rate pass-through to import prices (the “first stage” pass-through) and from import prices to consumer inflation (the “second-stage” pass-through) are assessed using error correction models in the context of a “long-run PPP hypothesis” and a “mark-up model,” such as those of de Brouwer and Ericsson (1995) and Dwyer and Leong (2001), respectively. The paper assesses the second-stage pass-through following a somewhat different approach, by estimating explicitly the direct impact of exchange rate changes, rather than import price changes, on consumer price inflation. In addition, the paper presents additional evidence on whether these effects have differed across tradable and nontradable goods. The empirical evidence for New Zealand is compared with that for Australia and Canada.

100. The paper shows that the exchange rate pass-through to consumer prices has varied considerably during the 1990s in New Zealand. While the estimated models explain the fluctuations in inflation and the impact of the exchange rate reasonably well, this relationship appears to be far from clear across countries. Nevertheless, some important conclusions can be drawn:

  • The three economies experienced a small exchange rate pass-through to import and consumer prices during the 1990s. The first and second stage pass-through has fluctuated significantly during the 1990s, particularly in New Zealand.

  • The decline in the short-term exchange rate pass-through to import and consumer prices in New Zealand appears to have been largely transitory in nature and may be attributed to the impact of sequential and highly persistent shocks. Moreover, the recent rebound in the short-term exchange rate pass-through seems to be primarily a tradable-good phenomenon, and in the case of New Zealand, it does not appear to be related to changes in the conduct of monetary policy that were introduced in early 1999.

  • The small exchange rate pass-through to domestic inflation reflects a “decoupling” of non-tradable goods inflation from the exchange rate, which may be partly explained by the high credibility of the inflation targeting framework. This seems to be particularly the case in New Zealand, where permanent shocks to the exchange rate do not appear to have had major impact on non-tradable good inflation.

  • Positive supply shocks, particularly a subdued growth in unit labor costs, have helped to contain inflation by increasing the scope to absorb large exchange rate fluctuations; a common feature in the three economies.

101. This paper is organized as follows: the next section reviews the main stylized facts of the exchange rate pass-through to inflation in New Zealand, and compares it with those of Australia and Canada; section C describes the empirical methodology and the main estimation results; section D briefly discusses their implications for the conduct of monetary policy in New Zealand.

B. Stylized Facts on the Exchange Rate Pass-Through

102. Following the adoption of inflation targeting in New Zealand, measuring and forecasting the impact of changes in the exchange rate on domestic prices became very important for conducting monetary policy. As noted by the Reserve Bank of New Zealand (1999), the “direct effect” of the exchange rate on inflation declined markedly during the 1990s, a phenomenon that has also been observed in other advanced open economies, such as Australia and Canada. This has raised some questions about the underlying factors driving the lack of pass-through, its sustainability over time, and the implications for inflation.57

103. The first stage pass-through appears to have varied considerably in New Zealand and Canada, but less so in Australia (Figure III.1). In the case of New Zealand, the exchange rate experienced wide fluctuations since 1993, but import prices did not vary as much. The exchange rate appreciated markedly between 1994 and mid-1997, but import prices remained largely unchanged. In the wake of the Asian crisis, the sharp depreciation was not followed by a similar rise in import prices. However, since mid-1999, the 12-month percent change in the exchange rate and import prices has been very similar. In the case of Canada, there are some episodes in the past where the pass-through to import prices also appears to have been significantly less than expected. However, since early 2000, fluctuations in the exchange have been largely similar to those in import prices. As for Australia, there seems to be less of a divergence between exchange rate and import price fluctuations.

Figure III.1
Figure III.1

Exchange Rate and Import Prices

(12-month percent change)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

104. The second-stage pass-through has also varied substantially during the 1990s and seems to have been significantly smaller than for import prices across the three economies (Figure III.2). The second-stage pass-through in New Zealand appears to have been largely absent between mid-1991 and late 1999. Since early 2000, the correlation between inflation in import and consumer prices increased sharply. In Australia, the correlation between fluctuations in import and consumer prices has also risen markedly since the second half of 1999. In Canada, the pass-through to consumer prices has been muted, as the correlation between the 12-month change in import prices and core consumer prices (excluding energy and other volatile components) has remained low.

Figure III.2
Figure III.2

Inflation and Pass-Through

(12-month percent change)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

105. A closer examination of the second-stage pass-through to tradable and nontradable consumer prices in New Zealand shows that tradable goods inflation remained largely stable despite wide fluctuations in import prices throughout most of the 1990s, while inflation in nontradable goods seems to have been closely associated with the trend in the output gap (Figure III.3). Between mid-1991 and mid-1999, tradable good inflation stabilized at about 1 percent (12-month rate), only to accelerate to 5 percent by end-2000. In contrast, inflation in nontradable goods has been more volatile and seems to have been determined by “demand-pull” conditions, particularly since early-1993, as the correlation between the output gap and nontradable good inflation stood at 80 percent.

Figure III.3
Figure III.3

Import Prices and Tradable and Nontradable Good Prices

(12-month percent change)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

C. Some Casual Empiricism on the Underlying Factors

106. The decline in exchange rate pass-through might seem to be puzzling because these economies have experienced a marked increase in “import penetration” since the late 1980s. Between 1987 and 2000, the increase in import penetration has been broadly similar between New Zealand and Australia, although it was still higher in New Zealand by end-2000, Canada has experienced the fastest growth in import penetration during this period, largely reflecting the effects of trade liberalization under the FTA and NAFTA, with the average share of imports being almost twice as high as in Australia and roughly 30 percent higher than in New Zealand. The rise in the share of imports in domestic demand would suggest that the pass-through to domestic prices of fluctuations in exchange rates and import prices should have risen instead of declining in these economies.58

A Comparison of Imports to Domestic Demand

(In percent)

article image
Source: Staff estimates based on national authorities’ data.

Second quarter of 1987 to fourth quarter of 2000.

Second quarter of 1987 to fourth quarter of 1990.

First quarter of 1991 to fourth quarter of 1996.

First quarter of 1997 to fourth quarter of 2000.

107. An additional factor which tends to influence the extent of pass-through is the business cycle and the uncertainty in aggregate demand. Mann (1986) noted that firms may be reluctant to raise prices and prefer to adjust profit margins when the underlying strength in domestic economic conditions is uncertain. Therefore, the exchange rate pass-through would tend to diminish when aggregate demand is more volatile. Using the volatility of the output gap as a proxy for uncertainty in aggregate demand, it seems evident that the uncertainty rose between 1991 and 1998 in New Zealand (although it declined somewhat between 1996 and 1998), whereas it declined in Australia, and remained almost unchanged in Canada. Since 1999, the uncertainty in aggregate demand declined further in all countries and, as expected from the earlier discussion, there seems to be more evidence of higher pass-through, particularly in New Zealand. As noted by Orr et al (1998), firms may have been under less pressure to pass on the effects of an appreciating currency when domestic conditions were strong, and vice versa when the exchange rate depreciated.

A Comparison of Aggregate Demand Uncertainty

(Standard deviation of output gap, in percent)

article image
Source: staff estimates, OECD, national authorities’ data.

Second quarter of 1987 to fourth quarter of 2000.

Second quarter of 1987 to fourth quarter of 1990.

First quarter of 1991 to fourth quarter of 1998.

First quarter of 1999 to fourth quarter of 2000.

108. Whether exchange rate volatility may have influenced the extent of first-stage pass-through differently across countries remains somewhat uncertain. In the early 1990s, New Zealand experienced a marked increase in the volatility of the exchange rate, possibly reflecting the effects of macroeconomic and structural reforms in the economy. As Governor Brash (1999) noted, in the years following the deregulation, both interest rates and the exchange rate were very volatile, and the Reserve Bank had “little sense of what ‘normal’ levels for these prices were….” However, exchange rate volatility in New Zealand was not higher on average in the 1990s than in the pre-reform period and was not significantly different than that observed in other industrial countries during the past decade. Nevertheless, exchange rate volatility increased markedly between 1991 and 1994 and subsequently, during the Asian crisis (Figure III.4).59 In contrast, Australia and Canada did not experience a significant change in their levels of exchange rate volatility during the 1990s. Australia’s exchange rate volatility has been high since the late 1980s, whereas in Canada it was relatively low.60 Therefore, the increased volatility in exchange rate shocks may have influenced for some time the extent of exchange rate pass-through in New Zealand, as these shocks may have raised the uncertainty about the underlying sources of exchange rate fluctuations, inducing firms to reduce the “first-stage” pass-through.

Figure III.4
Figure III.4

Exchange Rate Volatility

(Six-quarter rolling percent standard deviation of change in exchange rate)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

D. Estimation Framework and Econometric Evidence

109. This section describes the conceptual framework supporting the empirical analysis of exchange rate pass-through to inflation and also presents the results of its main determinants in New Zealand, Australia, and Canada. The evidence is based on two different methodologies. A vector autoregressive (VAR) model is used to decompose the variance of inflation rates at different stages of distribution in order to measure the explanatory power of exchange rate shocks in the fundamental determinants of inflation. Also, an error correction model is used to assess the exchange rate pass-through to inflation. In particular, the model is used to examine the exchange rate pass-through to import prices under the assumption that the law of one price holds, while the second-stage exchange rate pass-through is also examined by estimating a price mark-up error correction model (see de Brower and Ericsson, 1995, and Dwyer and Leong, 2001) for consumer inflation. These models help to assess the relative importance of the fundamental determinants of inflation, and the magnitude and stability of the exchange rate pass-through to inflation at different levels of the distribution chain (primarily import and consumer prices) over the short- and long-term.

Variance Decomposition

110. The main purpose of decomposing the variance of inflation rates is to assess the importance of shocks in the fundamental determinants of inflation at different stages of the distribution chain. In doing so, the variance decomposition may help to better assess the role of exchange rate shocks in driving fluctuations in inflation in the short-term and over the medium-term. To decompose the forecast variance of inflation, a VAR model is estimated for domestic import and consumer price inflation for each country.61

111. The variance decomposition for import prices for New Zealand, Australia, and Canada shows that exchange rate shocks account for a significant proportion of the forecast variance of domestic import prices in the short- and medium-term. In the case of New Zealand, the importance of exchange rate shocks appears to be significant and stable over time, as it accounts for about 50 percent of import price fluctuations within one year and three years, respectively. In the case of Australia, exchange rate shocks account for about 75 percent of import price fluctuations within a quarter, but their importance fades with the forecast horizon, accounting for about 65 percent of the CPI inflation variance beyond one year. In Canada, exchange rate shocks account for roughly 45 percent of fluctuations in domestic import prices.

Variance Decomposition of Prices

article image
Source: Staff estimates based on official sources.

Based on a VAR model using variables in first differences. In New Zealand, domestic import prices include oil.

112. The results for the variance decomposition for consumer price inflation supports the view that exchange rate shocks appear to have modest effects on inflation. These shocks account for only 30 percent of the CPI inflation forecast variance in New Zealand, and only 15 percent and 10 percent for Australia and Canada, respectively. A breakdown of consumer price inflation into tradable and nontradable goods reveals that exchange rate shocks account for a larger share of price fluctuations in New Zealand for tradable and nontradable goods than in Australia and Canada.

A Model for Import Prices

113. As explained in Dwyer and Leong (2001), estimating the first-stage pass-through is usually based on the assumption that the law of one price for tradable goods holds, which implies that the price of a traded good, when expressed in a common currency, should be the same in the domestic and foreign economies. In particular, the assumption in a log-linear form implies that:

p = p* + e

where p is the domestic price of imports, p* is the foreign price of imports, and e is the nominal effective exchange rate measured as domestic currency per unit of foreign currency. The first stage pass-through is represented by the elasticity of the domestic import price with respect to the exchange rate. For small open economies such as New Zealand, Australia, and Canada, the first stage pass-through should be complete over the long-run. An error correction representation would be as follows:

Δipt = a0 + a1Δert + a2Δfipt + a3Δogapt + α (ip - η⋅er - βfip)t-1 + a4ogapt-1 + εt

where linear price homogeneity (or PPP) would imply that η and β should each be equal to 1. The model also incorporates the output gap to allow for the mark-up on imports to vary with the business cycle. In estimating an import price equation for New Zealand, King and Steel (1998) noted that, under conditions of imperfect competition, foreign exporters may set their price above their costs of production depending on many factors such as demand conditions in the domestic as well as the foreign economies. Including the output gap aims at capturing some form of “pricing to the market” by exporters to New Zealand.

114. The estimated error correction models for import price inflation in New Zealand, Australia, and Canada are presented in Table III.1. In the case of New Zealand, two sets of regressions are estimated for import prices (including and excluding oil prices) to allow for a more precise estimation of the exchange rate pass-through. In both equations for New Zealand, linear price homogeneity or long-run PPP is imposed. The estimated models explain between 40 percent and 75 percent of the fluctuation in domestic import prices and suggest the following:

Table III.1.

An Error Correction Model for Import Prices

Dependent Variable: Change in Log Import Prices 1/

article image
Sources: Staff estimates.

Heteroskedastic-autoregressive corrected standard deviations in parenthesis.

Based on model by Dwyer and Leong (2001).

The dummy captures price undercutting by Asian exporters following the Asian crisis, as suggested by Dwyer and Leong (2001). The variable was not significant for New Zealand but was significant for Australia and Canada.

Linear homogeneity has been imposed in all the estimated equations.

As in Dwyer and Leong (2001), the trend is intended to capture a shift in imports towards lower-priced goods from non-G-7economies. In New Zealand and Canada, the trend was not significant.

The dummy intends to capture price effects following the FTA and NAFTA agreements in 1987 and 1994.

Prob-values.

Based on critical values for cointegration tests from McKinnon. The tests reject the null hypothesis of no cointegration for all equations.

  • The first stage short-term pass-through (a1) varies between 0.5 and 0.6 in New Zealand depending on whether import prices include or exclude oil prices (Figure III.5).62 The results seems to be consistent across economies, as the estimated elasticities for Australia and Canada are close to 0.6, respectively. While the short-term pass-through has been remarkably stable for Australia and Canada since 1994, it declined markedly in New Zealand in the early 1990s, and between 1994 and late 1999, it rebounded to levels more comparable to the early 1990s and to those estimated in other countries such as Australia.63

  • The speed of adjustment from “PPP-disequilibrium” (α) is very small, varying between 1 percent (Canada) and 25 percent (Australia) per quarter (Figure III.6). The estimated speed of adjustment for New Zealand import prices is less than 5 percent, somewhat below the estimation by King and Steel (1998), and does not vary significantly when oil prices are excluded. In addition, the speed of adjustment seemed to have declined markedly in all three economies during the early 1990s, following the adoption of inflation targeting. The speed of adjustment remained stable in New Zealand but has declined somewhat in Australia and Canada following the Asian crisis (1997–1999) and has stabilized at very low levels subsequently.

  • The decline in the first-stage pass-through in New Zealand possibly reflects the confluence of several temporary shocks which have been sequential and relatively persistent in nature.64 In particular, the fluctuations in the currency and the uncertainty associated with the underlying nature of its fluctuations, the uncertainty about the equilibrium exchange rate level following the wide range of macroeconomic and structural reforms in the first half of the 1990s, as evidenced by the significance of the output gap, and shocks associated with the Asian and Russian crises, may have triggered a temporary but long-lasting change in the pricing behavior of exporters. These factors appear to have played out differently, to some extent, in New Zealand than in Australia and Canada. In fact, as noted by Orr et al (1998), the recent direct effect of exchange rate changes on tradable good prices became more muted than in the early years of inflation targeting, possibly reflecting several factors, including the adaptation to large swings in the value of the domestic currency, the increased globalization of trade, which made firms more willing to smooth out the impact of exchange rate fluctuations on prices, and the underlying strength in domestic conditions, which may have influenced the extent of exchange rate pass-through.

Figure III.5
Figure III.5

First Stage Short-Term Pass-Through

(from exchange rate to import prices)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

Figure III.6
Figure III.6

Speed of Adjustment First Stage Pass-Through

(from exchange rate to import prices)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

A Mark-Up Model for Consumer Price Inflation

115. De Brouwer and Ericsson (2000) explain that the mark-up model of inflation has had a long-standing and continuing presence in efforts to explain the main determinants of inflation. The model is based on the idea that, in the long-run, the domestic general price level is basically determined as a mark-up over total costs, which are primarily represented by unit labor costs and import prices. Therefore, the long-run relationship of the price level to its fundamental determinants could be represented in a log-linear form as follows:

p = ln (m) + γ⋅ulc + δ⋅ip

where m is the mark-up over unit labor costs (ulc) and import prices (ip), with γ and δ representing the elasticities of the price level to unit labor costs and import prices, respectively. Linear homogeneity implies that γ + δ should sum up to 1. An interesting feature of error correction models is that they allow the estimation of not only the long-run relationship between prices and its fundamental determinants, but also the short-run dynamics of the inflationary process and its fundamentals and how disequilibrium in the long-run relationship feeds back onto inflation in the short-run. The framework presented in de Brouwer and Ericsson (1995) and in Dwyer and Leong (2001) is modified to explicitly assess the direct impact of the exchange rate on consumer price inflation by decomposing domestic import prices into the exchange rate and foreign import prices.65 The error-correction model for consumer price inflation was estimated as follows:66

Δpt = a0 + a1Δulct + a2Δfipt + a3Δneert + a4Δogapt + α (p - γulc - λfip - ηe)t-1 + a5ogapt-1 + vt

where α measures the speed of adjustment from the long-run disequilibrium (α<0 for dynamic stability to hold).

Econometric Results

116. This section presents the results of the estimated models of consumer price inflation for New Zealand, Australia, and Canada.67 The results (Table III.2) suggest the following:

Table III.2.

An Error Correction Model for Consumer Price inflation

Dependent Variable: Change in Log Consumer Prices 1/

article image
Sources: Staff estimates.

Consumer Price Index excluding credit services in New Zealand. Heteroskedasticity and autocorrelation consistent standard deviations in parenthesis.

Prob-values. One and two asterisks denote rejection at the 5 and 1 percent level of significance, respectively.

Based on critical values for cointegration tests from McKinnon. The tests reject the null hypothesis of no cointegration for all equations.

  • The short-term exchange rate pass-through to consumer price inflation, as well as the speed of adjustment in the long-term disequilibrium, have been small and relatively similar in New Zealand and Australia. The estimated short-term pass-through from changes in the exchange rate was about 10 percent in New Zealand and Australia, and significantly higher than in Canada. The long-term exchange rate pass-through has been higher in Canada (69 percent) than in New Zealand (27 percent), and slightly higher in New Zealand than in Australia (17 percent), supporting the view that the pass-through tends to be higher in more open economies.68

  • The short-term exchange rate pass-through to consumer price inflation has fluctuated markedly during the 1990s in all these economies (Figure III.7). In New Zealand, the short-term pass-through has declined markedly since the early 1990s, appears to have bottomed out by mid-1999, and has risen since then to 1993 levels, much in line with the behavior of the first stage pass-through. To some extent, the rebound in the pass-through reflects the large real depreciation that took place in 2000. In Australia and Canada, the short-term pass-through has also declined somewhat from the levels in the early 1990s, and has rebounded modestly in Australia and Canada since late 1998.69

  • The degree of response of consumer price inflation to a permanent exchange rate shock has also been remarkably low since 1994 but has risen somewhat in the past few years in New Zealand and Canada. The stability of the impulse response function to a permanent 1 percent shock in the exchange rate (Figure III.8)70 indicates that such response over six quarters has been less than 10 percent in these economies since 1994. In the case of New Zealand, the impulse response of consumer price inflation to exchange rate shocks has declined between 1994 and mid-1999, suggesting that there has been an increased tendency by price-setters to perceive exchange rate shocks as relatively transitory. However, more recent exchange rate shocks appear to be perceived as more permanent, as reflected by the modest rebound of the impulse response function. A relatively similar trajectory is estimated for Canada, where the exchange rate shocks between 1997 and early 1999 appear to have been perceived as largely transitory. In the same vein, the impulse response function has risen since 1999, suggesting that these shocks may be now perceived as more permanent by price setters. As noted by Dwyer and Leong (2001), the impulse response function in Australia has been largely stable since 1994, with a modest decline since mid-1997.

  • Favorable shocks to unit labor costs have been an important influence in stabilizing consumer price inflation during the 1990s; however, the deterioration in productivity since 2000 may have contributed to increase the exchange rate pass-through in all these economies. The models do not find a significant impact of changes in unit labor costs on inflation in the short-term but the long-term elasticity appears to be quite high across countries. For New Zealand and Australia, the estimated elasticities of inflation to unit labor costs were roughly 40 percent, while that for Canada exceeds one.71 Productivity improvements between 1993 and 1999 in New Zealand, Australia, and Canada have allowed price setters to absorb the trend depreciation of the exchange rate and contributed markedly to the stabilization of the inflation processes (Figure III.9).

Figure III.7
Figure III.7

Short-Term Second Stage Pass-Through

(from exchange rate to consumer prices)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

Figure III.8
Figure III.8

Impulse Response to a Permanent Shock: Stability Test

(from exchange rate to consumer prices)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

Figure III.9
Figure III.9

Unit Labor Costs

(year on year percent growth rate)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

Is the Exchange Rate Pass-Through Different Across Goods?

117. Burstein, Eichenbaum, and Rebelo (2002) examined empirically the behavior of inflation and the real exchange rate after large devaluations for several countries.72 Their main conclusions indicated that while CPI inflation tends to be very low relative to exchange rate depreciation, there were important differences in the behavior of the pass-through between tradable and non-tradable goods. In particular, they noted that the relative price of nontradable goods tends to fall after large devaluations and that the rise in tradable good prices is significantly lower than it would be expected from a PPP relationship. This is largely due to the presence of distribution services for tradable goods and by the substitution in consumption away from imports to local goods, which are usually classified as tradable goods but which are produced mainly for domestic consumption and are inferior substitutes for imports.73

118. The mark-up error correction model is estimated for the tradable and non-tradable components of the consumer price index for New Zealand and Australia.74 The results (Tables III.3 and III.4) suggest the following:

Table III.3.

An Error Correction Model for Tradable Goods Inflation

Dependent Variable: Change in Log Consumer Good Prices 1/

article image
Sources: Staff estimates.

Consumer Price Index excluding credit services in New Zealand. Heteroskedasticity and autocorrelation consistent standard deviations in parenthesis.

Prob-values. One and two asterisks denote rejection at the 5 and 1 percent level of significance, respectively.

Based on critical values in McKinnon. The tests reject the null hypothesis of no cointegration.

Table III.4.

An Error Correction Model for Non-tradable Goods Inflation

Dependent Variable: Change in Log Consumer Service Prices 1/

article image
Sources: Staff estimates.

Consumer Price Index excluding credit services in New Zealand. Heteroskedasticity and autocorrelation consistent standard deviations in parenthesis.

Prob-values. One and two asterisks denote rejection at the 5 and 1 percent level of significance, respectively.

Based on critical values in McKinnon. The tests reject the null hypothesis of no cointegration.

  • Much of the increased short-term exchange rate pass-through to inflation has primarily been a tradable-good phenomenon in New Zealand and Australia. The pass-through has been small and relatively similar in both economies during the 1990s. Since 1999, the estimated coefficients have risen to about 10 percent.

  • The short-term pass-through to non-tradable good inflation is very small and has declined steadily in New Zealand during the 1990s. The estimated pass-through was about 15 percent in 1992 and has declined to about 5 percent in 2001 (Figure III.7).

  • New Zealand has experienced a remarkable change in the responsiveness of non-tradable goods inflation to exchange rate shocks, which may be explained by the increased credibility of the inflation targeting framework. The impulse response of non-tradable goods inflation to exchange rate shocks has declined sharply since 1994 (Figure III.10). This decline may not only reflect the impact on inflation of the response from monetary and fiscal policies but also the increased credibility of the inflation targeting framework, as the impact of exchange rate shocks on non-tradable sectors has been increasingly perceived as transitory. This may be primarily attributed to the increased credibility of the inflation targeting regime, as price setters realized that the monetary framework in place increasingly left less scope to accommodate the direct effects of exchange rate shocks on non-tradable prices. The estimated impulse response functions for Australia differ from that for New Zealand; however, they still show a very limited pass-through to non-tradable prices, and are consistent with a high credibility of the inflation targeting framework adopted.

Figure III.10
Figure III.10

Impulse Response to a Permanent Shock: Stability Test

(from exchange rate to non-tradable consumer prices)

Citation: IMF Staff Country Reports 2002, 072; 10.5089/9781451830231.002.A003

E. Implications for Monetary Policy

119. In conducting monetary policy, New Zealand policymaker’s views about the relationship between fluctuations in the exchange rate and inflation has evolved over time, mainly as a result of their increased experience with the inflation targeting regime that was adopted in 1990. As noted by Orr et al (1998), in the early stages of the inflation targeting framework, the Reserve Bank tended to rely heavily on those aspects of the transmission mechanism which it had traditionally identified with relatively good precision, specifically, the direct effect of the exchange rate on prices. This was done primarily to bolster the credibility and ensure the success of the inflation targeting framework.

120. This focus has since evolved, and as experience with the new framework was gained, the Reserve Bank introduced several important changes to its conduct of monetary policy which, in principle, may have increased the scope for accommodating the direct impact of exchange rate fluctuations. The Reserve Bank lengthened the horizon over which the inflation is targeted from about 6–18 months to about 18–24 months and also modified the Policy Target Agreement (PTA) by widening the target range from 0–2 percent to 0–3 percent in 1996. The econometric evidence presented in this paper does not show that these changes had any significant impact on the exchange rate pass-through to domestic inflation.

121. However, the decision by the Reserve Bank to use a monetary conditions index (MCI) in implementing monetary policy between mid-1997 and March 1999 may have influenced the exchange rate pass-through. As explained by Svensson (2001), during this period, changes in interest rates were automatically triggered by fluctuations in the exchange rate, irrespective of the underlying nature of the exchange rate shocks. Svensson notes that this policy may have contributed to interest rate variability and may have caused undue output volatility. This policy was abandoned in March 1999 and was replaced by the official cash rate as the main instrument for monetary policy.

122. A question arises as to whether the decision to implement and subsequently abandon the MCI target may have had some influence on the exchange rate pass-through in New Zealand, as the exchange rate pass-through should be endogenous to the existing monetary policy framework. One could argue that the use of an MCI may have contributed to reduce the exchange rate pass-through in the short-term, as price setters realized that exchange rate depreciations would tend to be offset by higher interest rates. Likewise, abandoning the MCI could have been perceived as increasing the scope for passing-through exchange rate shocks onto prices in the short-term.

123. While the empirical results in this paper show that the short-term pass-through has risen somewhat since 1999, there is no compelling evidence to link such a rise to the Reserve Bank’s decision to abandon the MCI targeting, as several sequential and highly persistent shocks make this identification difficult. The analysis suggests that while the relationship between the exchange rate and the inflation rate may have changed during the 1990s, it is likely to have been primarily the result of the increased credibility of the inflation targeting framework. The increased credibility has made, on balance, the conduct of monetary policy easier for the RBNZ. Moreover, the change in the observed relationship between exchange rate changes and inflation should have no major implications for the way the RBNZ conducts monetary policy in the current framework.

ANNEX III.1: Data Description

Common Variables

Foreign inflation: proxied by the total trade weighted consumer price index for trading partners, as estimated by INS. Foreign import prices for New Zealand include an average effective tariff rate on imports. In addition, in the case of the import price model which excludes oil, foreign inflation was proxied by the U.S. producer price index excluding energy.

New Zealand

Import prices: implicit price deflator for all imports, and for all imports excluding petrol prices. Source: Statistics New Zealand.

Output gap: based on calculations by Reserve Bank of New Zealand.

Terms of trade: ratio of ANZ’s commodity price index to import price of manufacturing goods.

Exchange rate: the reciprocal of the nominal effective exchange rate weighted by import and export shares of major trading partners. An increase in the index denotes a depreciation of the domestic currency. Source: IMF.

Consumer prices: price index excluding housing and credit services, seasonally adjusted. Source: Statistics New Zealand.

Unit labor costs: total economy, seasonally adjusted, based on calculations by Reserve Bank of New Zealand.

Australia

Import prices: price deflator for merchandise imports. Base 1989/90 = 100.

Output gap: natural logarithm of real GDP minus the trend component based on the Hodrick-Prescott filter.

Terms of trade: Calculated as the ratio of the Reserve Bank of Australia commodity price index in Australian dollars to the domestic import price of manufacturing goods.

Exchange rate: the reciprocal of the nominal effective exchange rate weighted by import shares of major trading partners. An increase in the index denotes a depreciation of the domestic currency. Source: IMF.

Consumer prices: price index for all groups, and goods and services components, seasonally adjusted, 1989/90 = 100.

Unit labor costs: nominal for the non-farm sector, calculated as the ratio of nominal hourly labor costs (non-farm compensation of employees plus payroll tax and fringe benefits tax less employment subsidies, per hour worked by non-farm wage and salary earners) to average hourly productivity (real gross non-farm product per hour worked by all employed persons). Base for index: 1998–99 = 100. Source: Department of the Treasury.

Canada

Import prices: Paasche price deflator for imports of all goods, seasonally adjusted, index 1992 = 100. Source: Statistics Canada.

Output gap: natural logarithm of real GDP minus the trend component based on the Hodrick-Prescott filter.

Terms of trade: calculated as the ratio of the Bank of Canada commodity price index excluding energy to the import price for manufactured goods. Source: staff calculations based on official data.

Exchange rate: the reciprocal of the nominal effective exchange rate weighted by import shares of major trading partners. An increase in the index denotes a depreciation of the domestic currency. Source: IMF.

Consumer prices: price index excluding food, energy, and the effects of indirect taxes, seasonally adjusted, 1992 = 100.

Unit labor costs: for the business sector. Source: Statistics Canada.

References

  • Bank of Canada, Monetary Policy Report, November 2000.

  • Brash, Donald T., “The New Zealand Dollar and the Recent Business Cycle,” address to the Canterbury Employers’ Chamber of Commerce, January 29, Reserve Bank of New Zealand Bulletin, 1999, Vol. 62, No. 1.

    • Search Google Scholar
    • Export Citation

  • Burstein, A. Eichenbaum, M. and Rebelo, S., 2002, “Why are Rates of Inflation so Low After Large Devaluations?, unpublished manuscript.

    • Search Google Scholar
    • Export Citation

  • Canada: Selected Issues, 2001, IMF Staff Country Report No. 01/157, September.

  • Cashin, Paul and John McDermott, 2001, “Intertemporal Substitution and Terms of Trade Shocks,” unpublished manuscript.

  • Dwyer, Jacqueline and Leong, Kenneth, 2001, “Changes in the Determinants of Inflation in Australia,” Research Discussion Paper 02, Reserve Bank of Australia, May.

    • Search Google Scholar
    • Export Citation

  • De Brouwer, Gordon and Ericsson, Neil, 1995, “Modeling Inflation in Australia,” International Finance Discussion Papers, No. 530, Board of Governors of the Federal Reserve System, November.

    • Search Google Scholar
    • Export Citation

  • King, Alan and Douglas Steel, 1998, “Import Prices and Exchange Rate Pass-Through in New Zealand” Economic Discussion Papers, No. 9805, University of Otago, New Zealand.

    • Search Google Scholar
    • Export Citation

  • Mann, Catherine, 1986, “Prices, Profit Margins, and Exchange Rates,” Federal Reserve Bulletin, No. 72, 366–79.

  • McCarthy, Jonathan, 2000, “Pass-Through of Exchange Rates and Import Prices to Domestic Inflation in Some Industrialized Economies” Federal Reserve Bank of New York, Working Paper.

    • Search Google Scholar
    • Export Citation

  • Orr. A., Alasdair Scott, and Bruce White, 1998, “The Exchange Rate and Inflation Targeting,” Reserve Bank of New Zealand, Bulletin, September, Vol. 61 No. 3, pp. 183–191.

    • Search Google Scholar
    • Export Citation

  • Reserve Bank of New Zealand, Monetary Policy Statement, May 2001.

  • Reserve Bank of New Zealand, Bulletin, 1999, Vol. 62, No. 1 and No. 3.

  • Svensson, L., 2001, “Independent Review of the Operation of Monetary Policy in New Zealand: Report to the Minister of Finance,” Reserve Bank of New Zealand, February.

    • Search Google Scholar
    • Export Citation

  • Taylor, J., 2000, “Low Inflation, Pass-Through, and the Pricing Power of Firms,” European Economic Review, forthcoming.

56

Prepared by Martin Cerisola (x38314) who is available to answer questions.

57

For example, the Reserve Bank of New Zealand (RBNZ, 2000) notes that “…while the first stage of pass-through (from the exchange rate to import prices) was very much dampened between 1994 and 1999, it seems to have returned in the third and fourth quarter of 2000…At the second stage of pass-through (from import prices to consumer prices), the pass-through relationship is even less clear….”

58

The sharp rise in import penetration experienced in New Zealand, Australia, and Canada, would suggest that the direct effect of the changes in the exchange rate to consumer prices should have increased proportionately with the share of imported goods in the consumer price index. For example, see the Bank of Canada, Monetary Policy Report, Technical Box 2, November 2000. However, increased import penetration may result in more competition vis-à-vis domestic producers, limiting their ability to pass higher costs onto prices.

59

Alternatively, proxying exchange rate volatility as deviations in the NEER from a trend estimated by using the Hodrick-Prescott filter shows that volatility rose sharply after 1991, and even rose further after 1994, only to decline from 1999 on. However, these deviations should be interpreted with caution as they do not appear to be stationary.

60

Arora and Jeanne (2000) note that the Canadian dollar has fluctuated by less against the U.S. dollar than have several other floating currencies, such as those of Australia and New Zealand. The low relative volatility seems to be consistent with the behavior of official interest rates and foreign reserves (an indicator of market intervention), both of which have fluctuated more in Canada.

61

Most of the variables are nonstationary, and the Augmented Dickey Fuller tests for stationarity suggest that most of them are difference stationary, with the exception of unit labor costs. Nevertheless, the VAR models were specified and estimated in first differences given that the models in levels did not meet the stability conditions specified by Lutkepohl (1991). While this specification may entail a potential significant loss of information on the long run relationships, it may preserve the efficiency of the estimates. However, differencing the data would imply imposing the restriction that all shocks tend to be permanent, which may not necessarily be the case for some of the variables used in the VAR. Moreover, the reduced form residuals from the VAR were orthogonalized using a Choleski decomposition to identify the shocks. Since the order of the variables affect the results, the order chosen was as follows: output gap, unit labor costs, foreign inflation, exchange rate, and prices. The variance decomposition results were relatively robust to different orderings in these variables. See Annex III.I for a detailed description of data.

62

The results also suggest that the estimated first stage short-term pass-through to-import price inflation does not differ when long-run PPP is not imposed in the model.

63

In addition, the increased reliance by firms on foreign exchange risk hedging may have also contributed to reduce the degree of first-stage exchange rate pass-through.

64

The estimated pass-through coefficient declined from about 0.65 (0.76 excluding oil) in early 1991 to 0.5 (0.43) in early 1999, and has risen to about 0.6 (0.52) in the third quarter of 2001.

65

In doing so, the estimated coefficient for the second-stage pass-through is not strictly comparable to more traditional estimates of the pass-through, which are based on the impact of import prices on consumer price inflation.

66

This model was extended and estimated to assess the long-term impact on consumer prices of terms of trade and of the upward trend in import penetration, and whether these variables affected the estimated exchange rate pass-through. Terms of trade shocks influence the intratemporal substitution of consumption between tradable and nontradable goods, as well as private savings and the current account. Cashin and McDermott (2001) find that terms of trade shocks induce large and significant intratemporal and intertemporal substitution effects in Australia, New Zealand, Canada, United Kingdom, and the United States. In addition, a higher share of imported goods in the CPI would tend to increase the direct effects of exchange rate changes on the CPI. Including proxies for both variables in the model support the hypothesis that the increased share of imports in the basket of goods comprising the consumer price index has positively affected the exchange rate pass-through in the long-term, but these proxies do not appear to significantly affect the short-term exchange rate pass-through. When including these proxies, the estimation is subject to multicollinearity with other long-term determinants, such as unit labor costs, making it difficult to interpret the long-run relationship between these determinants and the price level.

67

The estimation for Australia is based on the consumer price index and differs from that in Dwyer and Leong (2001), since theirs is based on the price index which excludes volatile components.

68

In the case of New Zealand, the estimated speed of adjustment does not appear to be statistically significant, suggesting the lack of a long-term relationship between these variables. However, this result does not appear to be robust to a decomposition of consumer prices into tradable and nontradable good prices, which suggests the existence of a long-term relationship.

69

In the case of Australia, when excluding the volatile components of the consumer price index, the exchange rate pass-through was significantly more stable during the 1990s.

70

The impulse response function traces the extent to which adjustment to a 1 percent shock to the exchange rate is estimated to have occurred by the period shown. For example, each point in time traces the effects of a shock that took place in the past four and six quarters.

71

The sensitivity of the long-term elasticity to unit labor costs may be influenced by the inclusion of the output gap. Also, the specification based on core inflation results in a more reasonable long-run elasticity for unit labor costs in Canada.

72

The countries included were Finland, Sweden, Mexico, Brazil, Korea, Indonesia, Philippines, Thailand, and Malaysia.

73

Therefore, the prices for these local goods tend to be determined mainly by domestic factors and may be largely immune to changes in the exchange rate.

74

The results for Canada are not presented as models for tradable (goods) and non-tradable (services) consumer price inflation were highly rejected.

Same Series

Other IMF Content

Other Publishers

Cover IMF Staff Country Reports
New Zealand: Selected Issues
Author: International Monetary Fund
Volume/Issue:
Volume 2002: Issue 072
Publisher:
International Monetary Fund
ISBN:
9781451830231
ISSN:
1934-7685
Pages:
98
DOI:
https://doi.org/10.5089/9781451830231.002
  • View in gallery

    Exchange Rate and Import Prices

    (12-month percent change)

  • View in gallery

    Inflation and Pass-Through

    (12-month percent change)

  • View in gallery

    Import Prices and Tradable and Nontradable Good Prices

    (12-month percent change)

  • View in gallery

    Exchange Rate Volatility

    (Six-quarter rolling percent standard deviation of change in exchange rate)

  • View in gallery

    First Stage Short-Term Pass-Through

    (from exchange rate to import prices)

  • View in gallery

    Speed of Adjustment First Stage Pass-Through

    (from exchange rate to import prices)

  • View in gallery

    Short-Term Second Stage Pass-Through

    (from exchange rate to consumer prices)

  • View in gallery

    Impulse Response to a Permanent Shock: Stability Test

    (from exchange rate to consumer prices)

  • View in gallery

    Unit Labor Costs

    (year on year percent growth rate)

  • View in gallery

    Impulse Response to a Permanent Shock: Stability Test

    (from exchange rate to non-tradable consumer prices)

Exchange Rate and Import Prices

(12-month percent change)

Inflation and Pass-Through

(12-month percent change)

Import Prices and Tradable and Nontradable Good Prices

(12-month percent change)

Exchange Rate Volatility

(Six-quarter rolling percent standard deviation of change in exchange rate)

First Stage Short-Term Pass-Through

(from exchange rate to import prices)

Speed of Adjustment First Stage Pass-Through

(from exchange rate to import prices)

Short-Term Second Stage Pass-Through

(from exchange rate to consumer prices)

Impulse Response to a Permanent Shock: Stability Test

(from exchange rate to consumer prices)

Unit Labor Costs

(year on year percent growth rate)

Impulse Response to a Permanent Shock: Stability Test

(from exchange rate to non-tradable consumer prices)