VoxEU Column Exchange Rates

Exchange rate pass-through using highly disaggregated micro-data

Using aggregate data can bias estimates of exchange rate pass-through by ignoring heterogeneity in price adjustment across sectors. This column uses micro-level price data to estimate pass-through for South Africa, using actual CPI weights to reflect changes in consumption bundles. The result is a micro-based estimate of pass-through with aggregation consistent enough to interest monetary policymakers.

The pass-through of commodity prices and exchange rates to consumer prices seems to have fallen in industrialised countries since the late 1980s. Understanding the role of pass-through is of great interest to monetary policymakers – particularly in inflation targeting countries, where forecasting inflation plays a key role. Developing and emerging market (DEM) countries have been the biggest adopters of inflation targeting (or more credible monetary policy regimes) over the last two decades. Of the African countries, the key exemplar is South Africa.

Exchange rate pass-through (ERPT) refers to the degree to which a country's prices change in response to a change in its exchange rate. The original definition referred to the percentage change in import prices in response to a 1% change in the exchange rate (now called Stage 1 pass-through). This has been extended to address the effect of exchange rate movements on producer or consumer prices (overall pass-through). The effect of a change in import prices on producer or consumer prices is known as Stage 2 pass-through. Typically, the sensitivity to the exchange rate will decline down the price distribution chain, from import prices through producer prices to final consumer prices. ERPT to prices is incomplete if exchange rate changes elicit less than equi-proportionate changes in prices.

ERPT studies for consumer and producer prices have concentrated on aggregate price indices. Aggregate data offer two distinct advantages.

  • Underlying the aggregate indices are millions of micro-prices, and the heterogeneous weights applied to them – being derived from consumer price surveys (e.g. for the CPI) and regularly updated – reflect their economic relevance. By contrast, most disaggregated price data studies usually aggregate with equal weights for simplicity, and often for a subset of prices that may not be the most intensively used. The resulting ‘aggregate’ is not closely connected with the key concerns of monetary policymakers.
  • A second important advantage is that with aggregate price indices, a systems approach can help disentangle exogenous shocks from changes in the exchange rate, and hence address the monetary policy feedback effects in ERPT relationships (Aron et al. 2014a). This is missing in micro-analyses (see below).

Against these advantages, there is likely an ‘aggregation bias’ in ERPT. This means that aggregating up individual sub-sectors – while ignoring the consistent evidence for large and persistent heterogeneity in sectoral price adjustments – introduces measurement error in ERPT. For instance, it is known that service prices are stickier than goods prices, and raw goods prices are more flexible than processed goods (Klenow and Malin 2011). A disaggregated analysis thus captures more accurately the time dynamics involved in transmitting exchange rate changes to prices.

ERPT has been intensively studied using disaggregated trade prices at industry, firm and product levels, mainly for industrialised countries. Even in Menon’s (1995) early survey almost half the studies on import price pass-through used disaggregated industry data (with a handful at the product level). More recently this relative prevalence has been facilitated by standardised international definitions of traded goods1 (Goldberg and Knetter 2007).

A new, burgeoning strand of research examines pricing behaviour with highly-disaggregated consumer price, producer price and trade price data. Most of the small body of research pertains to the US and Euro-area (surveyed in Klenow and Malin 2011 and Melick and Galati 2006). Data sets using the millions of prices that underlie the official aggregate monthly consumer and producer price indices have become available, as have weekly ‘scanner’ (barcode) data and daily data from web-based retailers. In the US, micro-data underlying monthly import and export price indices are now provided by the Bureau of Labor Statistics (originating from surveys of firms). However, only a small subset of this research examines ERPT, e.g. using US micro-trade prices as summarised in Gopinath (2012), US retail and wholesale coffee prices (Nakamura 2008, Nakamura and Zerom 2010), or beer prices (Hellerstein 2008).

The key findings from the micro-data studies for industrial countries on ERPT are that:2

  • Heterogeneous ERPT estimates are typically reported at the sectoral and goods levels, and ERPT is delayed and incomplete for imports, and for both retail and wholesale domestic prices.
  • Goods with frequently adjusting import and export trade prices have a far higher long-run ERPT than low-frequency adjusters (Gopinath and Itskhoki 2010b). Exporters to the US of homogeneous goods (e.g. raw products) mainly price in dollars and adjust prices more frequently than for product-differentiated goods, for which there is a higher proportion of non-dollar pricers (emphasising that choice of the invoicing currency is endogenous; see Gopinath and Itskhoki 2010a). The macro-level implications are clear if the trade composition should adjust towards differentiated goods.
  • The most important source of incomplete ERPT for destination country retail and wholesale prices is the combination of non-traded local costs in the destination market (e.g. costs of distribution and wholesale/retail such as transport, taxes, tariffs, storage, marketing, advertising, finance, insurance, real wages and rents) and imported inputs into the exporter’s good. Together they are estimated to explain in the region of 50-78% of the gap between actual and complete ERPT for industries as diverse as beer, coffee, and automobiles (Goldberg and Hellerstein 2008). Studies using alternative methodologies and more aggregated data reach similar conclusions (Burstein et al. 2003, Goldberg and Campa 2010). They suggest that the greater use of imported inputs in traded and non-traded goods across countries and industries is the key contributor to changing ERPT – not changes in distribution margins.
  • Mark-up adjustment accounts for most of the rest of the gap between actual and complete ERPT in Nakamura and Zerom’s (2010) study of the coffee industry and in Hellerstein’s (2008) study of the beer industry.
  • In explaining incomplete ERPT, the role of nominal rigidities (that cause unresponsiveness in prices in the short-run since firms paying a ‘menu cost’ will adjust prices less frequently) appears small in the longer-run but is important in the delayed response of prices to cost in the short-run (Nakamura and Zerom 2010). Rigidity seems to occur mainly at the wholesale and not retail level. 

A caveat is needed as restricted structural models were used for these studies. The ERPT may be biased for the usual reasons: the static single equation approach3 misses dynamic price-adjustment and may overestimate the role of local costs. More generally, assuming exogeneity of the exchange rate neglects monetary policy and other feedback effects.

With volatile food and energy prices so important to the inflation process in DEM countries, supplementing analysis at the aggregate and sector levels with micro-level investigations should advance the understanding of these and other underlying drivers of inflation. Some studies examine ERPT to the main sub-components of aggregate price indices such as the PPI or CPI, covering brief time periods (e.g. Soffer 2006). Parsley examines ERPT with partly disaggregated retail price data from the Economist Intelligence Unit (EIU) for emerging markets (e.g. Parsley 2012). Finally, a recent strand of the forecasting literature forecasts sectoral inflation for sub-components of the CPI, giving insights into ERPT.4

To our knowledge, no published studies for emerging market countries examine ERPT to micro-price data underlying the CPI. The first such emerging market study for ERPT is Aron et al. (2014a). This paper examines ERPT to consumer prices in South Africa using a micro data set with over one thousand product groups (and over 2.5 million price observations) dating from December 2001 to December 2007, and covers close to 63% (by index weight) of the prices in the CPI. A subset of services prices from the aggregate CPI is included. By using the actual weights of the CPI basket (from consumer expenditure surveys), the derived ‘aggregate’ pass-through measure has macro-policy content.

In common with the above findings, considerable heterogeneity was found in pass-through for the different product groups using differenced single equation methods. Non-food products were divided into 10 groups, with the same set of regressors within each group, but with the regressors potentially different between the groups.  Food (with a large weight) was further divided into 10 sub-groups and the same empirical procedure applied, with the further innovation that the role of switches between import and export parity pricing5 for maize – an important ingredient in cereals and an input into the production of meat, milk, cheese and eggs – was investigated. Weighting the estimates by index weights, pass-through after two years for the CPI components covered by the study was estimated at 30%, near the upper end of earlier studies on aggregate CPI data for South Africa.  After six months, pass-through is about 6%, 18% after 12 months and 26% after 18 months. For food as a whole, pass-through is estimated at 46% after two years. Interestingly enough, when heterogeneity between different food groups was ignored and a common set of variables was forced on each food sub-group, both the estimate of overall food pass-through and overall pass-through for the ‘CPI’ fell significantly. This finding underlines the importance of recognising heterogeneity in modelling and in forecasting inflation. It suggests that in future work some of the other broad groups with significant weights – particularly transport goods – should be decomposed into sub-components.

To conclude, studies at the level of industries, firms, products or retail goods potentially lend insight into underlying structural price adjustments and the sources of incomplete and changing ERPT. As indicated by Bailliu and Murray (2010), this promising research area could assist in forecasting future patterns of ERPT of interest to monetary policymakers – and it should certainly include emerging markets. 


Aron, J, G Farrell, J Muellbauer, and P Sinclair (2014a), "Exchange Rate Pass-through to Import Prices and Monetary Policy in South Africa", Journal of Development Studies, vol. 50(1), pages 144-164, January. 

Aron, J, R MacDonald, and J Muellbauer (2014b), “Exchange Rate Pass-through in Developing and Emerging Markets: a Survey of Conceptual and Policy Issues, and Empirical Findings”, Special Section on Exchange Rate Pass-through, Journal of Development Studies 50 (1): 101-143.

Aron, J, K Creamer, J Muellbauer, and N Rankin (2014c), “Exchange Rate Pass-Through to Consumer Prices in South Africa: Evidence from Micro-Data”, Journal of Development Studies 50 (1): 165-185. 

Aron, J, and J Muellbauer (2012), “Improving forecast accuracy in an emerging economy, SA, by means of changing trends, long run restrictions and disaggregation”, International Journal of Forecasting 28, 2: 456-76.

Bailliu, J, and J Murray (2010), “Has ERPT really declined? Some recent insights from the literature”, Bank of Canada Review, Autumn, 1-8.

Burstein, A, J Neves, and S Rebelo (2003), “Distribution Costs and Real Exchange Rate Dynamics During Exchange-Rate-Based-Stabilizations”, Journal of Monetary Economics, 50, 1189-1214.

Goldberg, L, and J Campa (2010), "The Sensitivity of the CPI to Exchange Rates: Distribution Margins, Imported Inputs, and Trade Exposure", The Review of Economics and Statistics 92(2): 392-407.

Goldberg, P, and M Knetter (1997), “Goods prices and exchange rates: What have we learned?”, Journal of Economic Literature 35:1243–1292.

Goldberg, P, and R Hellerstein (2008), “A Structural Approach to Explaining Incomplete Exchange-Rate Pass Through and Pricing-to-Market”, American Economic Review 98 (2): 423–29.

Gopinath, G, O Itskhoki, and R Rigobon (2010a), “Currency Choice and ERPT”, American Economic Review 100:304-336.

Gopinath, G, and O Itskhoki (2010b) “Frequency of Price Adjustment and Pass-Through”, Quarterly Journal of Economics 125:675-727.

Gopinath, G (2012), “International Prices and Exchange rates”, NBER Reporter 2012 no 2 Research Summary.

Hellerstein, R (2008), "Who bears the cost of a change in the exchange rate? Pass-through accounting for the case of beer", Journal of International Economics, Elsevier, vol. 76(1), pages 14-32, September.

Klenow, P J, and B A Malin (2011), "Microeconomic Evidence on Price-Setting", in: B M Friedman and M Woodford (ed.), Handbook of Monetary Economics, edition 1, vol 3, chapter 6, pp.231-284, Elsevier.

Melick, W R,  and G Galati (2006), "The evolving inflation process: an overview", BIS Working Paper No.196, Bank for International Settlements.

Menon, J (1995), " Exchange Rate Pass-Through", Journal of Economic Surveys, Wiley Blackwell, vol. 9(2), pages 197-231, June.

Nakamura, E, and D Zerom (2010), "Accounting for Incomplete Pass-Through", Review of Economic Studies, Wiley Blackwell, vol. 77(3), pages 1192-1230, 07.

Nakamura, E (2008), "Pass-Through in Retail and Wholesale", American Economic Review, American Economic Association, vol. 98(2), pages 430-37, May.

Parsley, D (2012), “Exchange Rate Pass-through in South Africa: Panel Evidence from Individual Goods and Services”, Journal of Development Studies 48(7): 832-846.

Soffer, Y (2006), “ERPT to the consumer price index: a micro approach”, Bank of Israel Discussion Paper no.2.


1 The Standard International Trade Classification (SITC) and Harmonized System (HS) are different trade classifications, enabling trade price comparisons amongst different countries and years from the late 1980s. The UN-maintained SITC categorises by the materials used in production/processing stage to 5 digits, while the HS allows disaggregation by product up to at least 6 digits.

2 This draws on the empirical ERPT survey in Aron et al. (2014b).

3 The micro analyses for the beer and automobile industries use static equations; Nakamura and Zerom (2010) explore the coffee market with a dynamic model. However, all these models, and hence the role attributed to mark-up, rely heavily on the functional form assumptions employed for estimated demand.

4 Aron and Muellbauer (2012) survey the international literature, also reporting results for South Africa.

5 The import/export parity price is the world price plus/minus transport costs and switches to the higher import parity price occur when local harvests fall short.