Understanding the sources of incomplete exchange rate pass-through and the role of nominal rigidities in price adjustment, as well as the extent of expenditure switching by households and the allocative implications of invoicing currency in trade, are classical themes in international economics (e.g. Obstfeld and Rogoff 2001, Burstein and Gopinath 2014).1 In recent work (Auer et al. 2021), we shed light on these issues, using as a case study the appreciation of the Swiss franc (CHF) on 15 January 2015. On that date, the Swiss National Bank (SNB) discontinued the minimum exchange rate of CHF1.20 per euro that it had enforced for more than three years (Figure 1, top panel).
The subsequent appreciation episode is unique in several ways. First, it was unexpected, following a period of remarkable exchange rate stability. This makes it unlikely that the price dynamics we examine reflect any anticipation of the shock or adjustment lags due to prior exchange rate movements.2 Second, the exchange rate movement – with the Swiss franc temporarily appreciating by over 20% against the euro – was large in magnitude relative to standard short-term exchange rate fluctuations in advanced economies, which have been a main focus of the literature.3 Third, the appreciation occurred against the backdrop of a stable Swiss economy and reflected a policy response to foreign events.4
Figure 1 The exchange rate, aggregate prices, and border prices by currency of invoicing
Notes: The top panel shows the monthly EUR/CHF nominal exchange rate, core import price index, and consumer price index for imports and for domestic goods and services, all relative to December 2014. The bottom panel presents the EUR/CHF exchange rate and border price changes compared with the fourth quarter of 2014, displaying the coefficient of time fixed effects by quarter and invoicing currency at the border, in a sample restricted to non-zero price changes.
Source: Auer et al. (2021).
To investigate the event’s impact, we examine border data on prices and invoicing, as well as household-level data on prices and expenditure on non-durable consumer goods.5 This lets us link the currency of invoicing to border prices, retail prices, and expenditure allocations at the consumer level.
We first document, for each invoicing currency, large differences in the pass-through of border prices in the first year after the appreciation.6 Border prices fell much further for euro-invoiced goods than for Swiss franc-invoiced goods, even conditioning on non-zero price changes (Figure 1, bottom panel). This is consistent with findings in Gopinath et al. (2010) for border prices in the US. Qualitatively consistent with models of endogenous invoicing, estimated differences in conditional price changes by invoicing currency weaken over time and become statistically insignificant about one year after the Swiss franc’s appreciation.7
Second, we examine how the appreciation affected consumer prices and allocations. The transaction-level information on non-durable retail prices and expenditures is from the Swiss AC Nielsen ‘homescan’ data, which we augment with data on the origin of the purchased goods. We document a decline in the retail price of imports relative to Swiss-produced goods, and we show that variation across goods in the invoicing currency at the border has a sizeable impact on retail price changes (Figure 2, top panel).
We estimate the sensitivity of import prices at the retail level with respect to changes in border prices, leveraging heterogeneity in border price changes induced by variation in pre-appreciation euro-invoicing shares. These estimates imply that, after two quarters, a one percentage point reduction in import prices at the border resulted in a price cut of roughly 0.55 percentage points for imported products at the retail level.8
Figure 2 Border invoicing and retail prices
Notes: The top panel examines the impact of invoicing at the border for the retail prices of imported goods. It presents estimates of the interaction of time fixed effects (per quarter) with the pre-appreciation invoicing intensity, with the independent variable being the cumulative average price change of import prices relative to the fourth quarter of 2014. The bottom panel examines the impact of invoicing and import market share on the price response of domestic goods. It presents estimates of the interaction of time fixed effects with the pre-appreciation invoicing intensity and import market share, with the independent variable being the cumulative average price change of domestic prices relative to the fourth quarter of 2014. Whiskers indicate the bounds of a 95% confidence interval, calculated clustering at the level of retail product class.
Source: Auer et al. (2021).
Even though the response of retail prices of Swiss-produced goods was on average very muted, we show that prices fell more for border product categories invoiced in euros than they did for those invoiced in Swiss francs, conditioning on the expenditure share of competing imported goods in the same product category. This observation points to the presence of pricing complementarities whereby domestic producers react to changes in prices of competing imported retail products (Figure 2, bottom panel).
We also examine how the extensive margin of adjustment of retail prices responded. The average decline in retail import prices in 2015 was partly accounted for by an increase in the fraction of nominal price changes (Figure 3, top panel). This can, in turn, be decomposed into a large increase in the frequency of price reductions (Figure 3, bottom panel) and a smaller decline in the fraction of price increases. Differences in border price changes associated with the currency of invoicing carry over to consumer prices, not only for average changes but also for the extensive margin of price adjustment. Specifically, the increase in the frequency of price reductions was larger for imported products with a larger share of euro invoicing and with larger price reductions at the border.
Figure 3 Fraction of price changes and price decreases compared to December of the previous year
Notes: The top panel displays the weighted average fraction of good-specific price changes for imported goods relative to December of the previous year, and for 1–12 month horizons. The bottom panel displays the same statistic but counting only price decreases. For the construction of the underlying price series from micro data, see Auer et al. (2021).
Finally, we examine how far consumers switched their expenditure from one category of goods to another in response to the appreciation. Import shares rose substantially even at short horizons after the appreciation (Figure 4). Leveraging cross-sectional variation along the invoicing dimension, we show that expenditure shares on imported goods increased by more in product categories in which imports are invoiced in euro than they did for categories invoiced in Swiss francs. Hence, differences in invoicing currency at the border also matter for consumer allocations.9 To estimate the sensitivity of import expenditure shares with respect to changes in relative prices, we instrument import price changes across product categories using euro-invoicing shares at the border.
Estimated price elasticities of import shares are close to one based on border-level measures of import prices. They are much higher (ranging between two and five) based on retail-level measures of import prices, but also less tightly estimated given large idiosyncratic movements in consumer prices.
Figure 4 Aggregate import share in total expenditures
主题