Chapter 1 Introduction

Using comprehensive, shipment-level trade data, we examine the extent to which New Zealand exporters maintain stable New Zealand dollar prices by passing on exchange rate changes to foreign customers. That is, we consider exchange rate pass-through (ERPT) from the perspective of the exporter. While it is more common to consider ERPT from the importer's perspective, our aim is to inform debate on the role of exchange rate fluctuations in New Zealand exporter performance.

Over the period we consider, New Zealand experienced substantial fluctuations in bilateral exchange rates (figure 1). Bilateral currency swings of 20 to 30 percent were not unusual, motivating concerns about the sustainability of New Zealand-based export businesses at times of exchange rate appreciation. In large scale business surveys conducted in 2007 and 2011, exchange rate levels and volatility were the two most commonly cited challenges among firms with current overseas income (Statistics New Zealand 2012). Economic and political commentators have argued strongly for changes in the exchange rate regime to support export growth (Oram 2012; Tarrant 2012; NZMEA 2012). However, there is little New Zealand-based empirical evidence on the mechanisms through which exchange rate changes affect export (and exporter) performance. This paper takes a step towards filling this gap by estimating the extent to which exchange rate movements affect the New Zealand dollar (NZD) price received by exporters.[1]

Figure 1 - Exchange rate index, April 2004 - December 2012

Our approach builds directly on two recent microeconomic papers on ERPT. Using firm-level Customs and balance sheet data for French exporters, Berman, Martin & Mayer (2012) show that high-performing firms (those with relatively high productivity, export intensity, and with exports to a wide range of countries) tend to absorb exchange rate changes into their margins, while low-performing firms pass on a larger proportion of exchange rate fluctuations to their offshore customers which, in turn, impacts on sales volumes.

Meanwhile, Gopinath, Itskhoki & Rigobon (2010) consider the extent of ERPT to US import prices using longitudinal survey data. They examine differences in ERPT according to whether the currency of invoice is US dollars (USD) or not. They find that while both dollar and non-dollar invoiced goods show substantial nominal price stickiness in the short run, the degree of price adjustment in the long run differs across invoice currencies. Observed over the life of the good, goods priced in non-USD currencies continue to exhibit almost complete pass-through to USD prices - that is, the value received by the exporter remains almost entirely unaffected by exchange rate changes. In contrast, exporters invoicing in USD adjust their received value over time so that the change in the USD price reflects almost half the observed exchange rate movement. Gopinath et al. (2010) then use these findings to motivate a model of endogenous currency choice, in which exporters select their invoice currency in order to most closely reflect optimal prices during periods between price adjustments.

This paper draws a link between the two papers above by conditioning on both invoice currency and exporter performance, while extending the analysis of Gopinath et al. (2010) to distinguish between three invoicing options - producer currency (ie, NZD-denominated), local currency (denominated in the currency of the importing country), and vehicle currency (denominated in a third currency, predominantly the USD) - each of which is used by a substantial share of exporters. We first confirm that both short-run and long-run exchange rate pass-through differ by invoice currency. We then show that, within currency groups, there is little evidence for differential pass-through behaviour according to key firm or product characteristics. We relate this finding to the work of Berman et al. (2012) by showing that a negative relationship between pass-through and firms' export performance arises when invoice currencies are not controlled for. That is, the strong relationship found by Berman et al. (2012) may be driven by different invoicing strategies or opportunities across different types of firms, with lower value exporters being more likely to use producer currency pricing, which is in turn associated with greater pass-through to import prices.

The paper proceeds as follows. Section 2 reviews the relevant literature on ERPT. Section 3 outlines the specifics of our dataset, while sections 4 and 5 present descriptive and regression results respectively. Section 6 concludes.