A novel three-market view of price competitiveness

Abstract

A country’s price competitiveness is generally proxied by the real effective exchange rate of its currency. In this article we present a novel three-market breakdown (domestic, euro-area and non-euro area markets) of this measure, with an application to the four main euro-area countries. Price-competitiveness developments are indeed found to be heterogeneous across markets, thereby confirming the usefulness of this decomposition. In particular, we find that in the 1999–2018 period only Germany recorded comparable (large) gains in both euro and non-euro area markets. France posted a larger gain, and Spain marked a smaller loss, in euro-area markets; conversely, Italy’s mild improvement was greater in non-euro area markets. Finally, competitiveness in the domestic market and that in non-euro area markets are found to be the main, equally important, drivers of overall developments since 1999 in Italy and in Germany.

Keywords

Price-competitiveness indicators real effective exchange rates producer prices destination markets import competition

1. Introduction2

This study focuses on the measurement of price-competitiveness developments in the four main euro-area economies. We start from the standard definition of a country’s price-competitiveness indicator (PCI) – or real effective exchange rate (REER) for non-euro area countries according to the Eurosystem’s taxonomy – as a weighted geometric average of nominal exchange rates vis-à-vis its main trading partners or competitors (the nominal PCI or the nominal effective exchange rate, NEER, respectively), deflated by relative prices or costs. As in [1, 2], the weighting scheme encompasses both competitive pressures in the domestic market and export-based competition in foreign markets, such that the overall PCI can also be expressed as a weighted average of an import-based PCI and an export-based PCI. Moreover, the overall PCI can be centred on specific subgroups of competitors, in particular euro-area or non-euro area trading partners.

Many international institutions, such as the IMF [3] and the BIS [4], and central banks, such as the Bank of Italy [1], the ECB [2], the Bank of England [5] and the Federal Reserve Board [6], construct and publish overall PCIs, using a similar, standard methodology. Methodological differences (for example, in the number of competitors considered or in the updating frequency of the employed trade weights) are highlighted in Table A1 in Annex 1. To our knowledge, in addition to the Bank of Italy, only the ECB publishes PCIs disaggregated by subgroups of trading partners.

One significant difference across institutions concerns the choice of the price and/or cost index employed to deflate the nominal PCI or NEER. There is indeed consensus both in the trade literature and in the policy debate that no optimal deflator exists. Discussions on the various alternative cost and price indices employed to deflate the NEER are found, for example, in [7, 8], yet both theoretically and empirically, no index proves to be superior. In particular, the following deflators are commonly used, each of which has both general advantages and shortcomings as for the measurement of a country’s external competitiveness.

•
Consumer price indices (CPIs) are rapidly available on a monthly basis for most advanced and emerging economies. They are constructed with fairly homogeneous methodologies across countries and on the basis of comparable baskets. They include goods that are acquired by households for consumption, as well as services (many of which are non-tradable), excluding capital and intermediate goods, which are heavily traded. Moreover, they are subject to distortions owing to fiscal measures.
•
Producer price indices (PPIs) too are monthly indicators, yet they are compiled by fewer economies than CPIs and are often less timely. They refer to prices received by domestic producers for their output, generally with some sectoral disaggregation (typically, mining, manufacturing, utilities, construction and agriculture). They are less affected by taxes and subsidies than CPIs.3
•
Gross domestic product (GDP) deflators refer to all sectors and to all types of goods and services, but they are subject to significant composition effects between the public and private sectors. Moreover, they are only available on a quarterly basis with a significant delay relative to the reference period and may be subject to significant revisions.
•
Unit labour costs in manufacturing (ULCMs) refer solely to one sector, which however is the one tradable goods are usually identified with. They ignore additional components of production costs, so that their evolution may be affected by the possible substitution between material inputs, intermediate services, labour and capital. Furthermore, they are available on a quarterly basis, generally only for advanced economies, and are subject to strong backward revisions.
•
Unit labour costs in the total economy (ULCTs), on the one side, comprehensively include all sectors of the economy, so are less affected by input substitution; on the other side, they share the remaining drawbacks of ULCMs, in addition to the possible bias owing to potentially large sectoral composition effects.

In particular, the ECB publishes PCIs deflated by the CPI, the GDP deflator and the ULCT [2]. The Bank of Italy constructs PCIs deflated by the PPI of domestically sold manufactures [1]. The latter prices may be considered as a proxy for cost developments that encompasses all production cost pressures. Moreover, they are more likely to capture the whole sector of tradable goods, which is a broader concept than traded goods, since some tradables may turn out not to be actively traded internationally specifically because of competitiveness issues. In other terms, the basket of products represented in producer prices of manufactures sold on the internal market is more likely to capture the whole spectrum of potential supply to domestic and foreign markets.

Developments of alternately deflated PCIs can differ – and have differed – substantially for a number of countries, as discussed in [10, 11, 12], and as shown in Fig. 1A of Annex A for the four main euro-area countries.4 For example, according to the Bank of Italy’s PPI-based indicator, in the period 1999Q4-2009Q4 price competitiveness in Italy deteriorated by about six percentage points; the loss was much more pronounced (approximately 18 points) according to the ECB’s ULCT measure. [12] argues that this discrepancy was in part due to sharply falling labour shares in Italy’s main trading partner, Germany, which excessively penalised Italy’s relative ULCT, whereas it pushed up Italy’s relative PPI to a much lower extent, given that other cost components partly offset the labour dynamics. Another case-study is Spain. Since the end of 2009 Spain’s price competitiveness has improved by an impressive 20 percentage points if measured according to the ULCT-based PCI, yet only by around 3 according to the PPI-based measure. Labour shedding has indeed contributed to improve Spain’s overall price competitiveness, but the relative cost of other production factors and the evolution of mark-ups has dampened this effect. These examples imply that a comprehensive analysis of price and cost competitiveness should be based on a range of alternately deflated indicators (on this, see also [14, 15]). Yet this aim goes beyond the scope of this article.

Indeed, herein we examine PCIs from a different angle relative to the existing literature and which is independent from the choice of the appropriate PCI deflator. In particular, we consider subgroups of export markets, rather than subgroups of competitors, and derive measures of export-based competitiveness in specific markets (in particular, in euro-area and non-euro area markets). To our knowledge this is the first attempt to achieve this type of breakdown. This addition enables assessing price-competitiveness trends of a given economy in specific geographical areas, taking into account pressures from all countries that export to that region, instead of competitiveness developments in world markets against selected trading partners. A country’s price competitiveness performance may differ across markets for an array of heterogeneous factors;5 the analysis of these factors, however, goes beyond the scope of this study, which is centred solely on measurement issues.

In summary, we show that overall PCIs can be conveniently broken down according to a three-market perspective (domestic; euro-area; non-euro area). This allows gauging the role of each component in driving aggregate price-competitiveness trends of the four main euro-area countries, namely France, Germany, Italy and Spain. These four countries are an interesting case-study since they all display a non-negligible economic size and, despite being bound by a common currency (the euro), have recorded different export performances since 1999 (e.g. [18]); heterogeneous price-competitiveness developments may indeed be behind these dynamics.

The split between euro vs. non-euro area export outlets shows that in the overall 1999–2018 period only for Germany price-competitiveness developments were comparable in the two export markets, where significant gains were recorded. Instead, for France and Spain developments were relatively more favourable in the euro-area market; France recorded a larger gain and Spain posted a slightly smaller loss in the euro-area market than elsewhere. In contrast, Italy is the only economy that marked a better performance in non-euro area markets. Differences are much starker when focussing on the pre-global financial crisis period (1999–2007). Finally, price competitiveness in the domestic market and that in non-euro area markers is found to have contributed broadly equally to aggregate price-competitiveness dynamics in Italy and in Germany.

The structure of the article is the following. Section 2 first describes the standard methodology underlying the construction of the Bank of Italy’s PCIs and then explains the algebra underlying the new set of measures we here introduce. Section 3 discusses price-competitiveness developments in the four main euro-area countries since 1999 according to the standard aggregate (i.e. on all world markets) PCIs. Section 4 breaks the analysis down into euro and non-euro export markets, according to the novel set of indicators put forward in this article. Section 5 wraps up the analysis by applying a three-market decomposition of aggregate price-competitiveness developments, in order to gauge the main drivers of the latter in the four countries of interest. Section 6 concludes.
2. The methodology underlying the construction of the Bank of Italy’s price-competitiveness indicators

This section first recaps the methodology underlying Bank of Italy’s current PCIs, described more in detail in [1, 19, 20], which cover 61 countries, listed in Table B1 in the Annex, at a monthly frequency since January 1992. This section next describes the novel breakdown of export-based indicators in euro and non-euro markets, thus implementing the three-market decomposition of overall PCIs.

2.1 A recap of the current methodology

The starting point is the nominal effective exchange rate (NEER), or nominal PCI for euro-area countries, which is calculated as the weighted geometric average of bilateral nominal exchange rates. Omitting time subscripts for simplicity, the NEER of reporting country $i$ is defined as:

$\displaystyle\textit{NEER}_{i}=\prod^{N}_{j=1}{e^{w^{i}_{j}}_{ij}}$ (1)

where $N$ denotes the number of trading partners ( $N=$ 61 in Bank of Italy’s case, including the reporting country), $e_{ij}$ stands for the index of the nominal bilateral exchange rate between country $i$ and country $j$ (expressed in terms of $j$ ’s currency per unit of $i$ ’s currency, so that an increase indicates a loss of price competitiveness for $i$ ), and $w^{i}_{j}$ indicates the overall weight of competitor $j$ for the reference country $i$ .

The overall weight $w^{i}_{j}$ of each competitor $j$ in the group of $N$ trading partners is equal to the weighted average of an export weight ( ${}^{x}w^{i}_{j}$ ) and an import weight ( ${}^{m}w^{i}_{j}$ ):

$\displaystyle w^{i}_{j}=\alpha_{i}\ {}^{x}w^{i}_{j}+\left(1-{\alpha}_{i}\right% ){}^{m}w^{i}_{j}$ (2)

where ${\alpha}_{i}=\frac{X_{i}}{X_{i}+M_{i}}$ is the share of exports of reporting country $i$ on its total trade flows, computed in the reference period we detail in the Annex and reported in Table B2 therein for each of the four euro-area countries. Countries that are structurally net exporters (importers) are thus assigned a higher export (import) weight.

The import weight of competitor country $j$ is defined as its share in the reporting country $i$ ’s total imports (as accounted for by the set of N trading partners):

$\displaystyle{}^{m}w^{i}_{j}=m^{i}_{j}\left/\sum^{N}_{a=1}{m^{i}_{a}}\right.$ (3)

where $m^{i}_{j}$ denotes imports of reporting country $i$ from country $j$ and the denominator indicates total imports of reporting country $i$ , with $m^{i}_{i}=0$ and $\sum^{N}_{j=1}{}^{m}w^{i}_{j}=1$ . The higher the share of competitor $j$ in the reporting country’s total imports, the larger the weight of its exchange rate in the basket of currencies included in the NEER.

The export weight of competitor $j$ in the computation of the NEER for country $i$ is more articulate, as it is double-weighted in order to account for third-market effects:

${}^{x}w^{i}_{j}=\sum^{N+1}_{k=1,k\neq i}S^{j}_{k}x^{i}_{k},j=1,\ldots,j\neq i$ (4)

where $S^{j}_{k}$ denotes the share of competitor $j$ in market $k$ , $S^{i}_{k}=0$ by normalization, $x^{i}_{k}$ denotes the share of market $k$ in $i$ ’s exports and $N+1$ denotes the overall number of outlet markets considered, which is equal to $N$ plus the residual aggregate “rest of the world”. Each foreign market $k$ is defined, from the viewpoint of reporting country $i$ , as the sum of “locally-sold local production”, i.e. manufacturing gross output of country $k$ sold in its domestic market, and of manufacturing exports to country $k$ from all competitors $j$ ( $j\neq i$ ). Note that, due to data limitations, the residual aggregate “rest of the world” is treated as an additional outlet market while it is not included in the set of competitors for reporting country $i$ . In other terms, exports from the “rest of the world” to the other markets are disregarded and local production is not considered in the definition of the “rest of the world” market.6

The double-weighted export weight measures both the direct competition faced by reporting country $i$ in market $j$ from its local manufacturers ( $k=j$ ) and the indirect competition faced by reporting country $i$ from $j$ ’s exports in third markets $k$ ( $k\neq j$ ).

As is common practice in the literature, bilateral trade flows in some fixed reference period are employed to compute the weights ( $w^{i}_{j}$ , ${}^{x}w^{i}_{j}$ and ${}^{m}w^{i}_{j}$ ), as discussed more thoroughly in the Annex.

The PCI of a country $i$ is defined as the weighted geometric average of its relative prices or costs, where all prices/costs are measured in a common currency. Alternatively but equivalently, again omitting time subscripts, the indicator can be computed as the product of the NEER and a weighted geometric average of relative prices or costs:

$\displaystyle\textit{PCI}_{i}=\prod^{N}_{j=1}{{\left(P_{ij}e_{ij}\right)}^{w^{% i}_{j}}}=\textit{NEER}_{i}\prod^{N}_{j=1}{{\left(P_{ij}\right)}^{w^{i}_{j}}}$ (5)

where the last equal sign follows from Eq. (1) and $P_{ij}\equiv P_{i}/P_{j}$ , where $P_{i}$ and $P_{j}$ are the price indices (in our case, PPIs of domestically sold manufactures) for countries $i$ and $j$ , respectively. Given the way it is constructed, an increase in the indicator implies a loss in price competitiveness for the reporting country $i$ .

Finally, Eqs (2)–(5) can be combined to obtain an alternative interpretation of the “overall” PCI, expressed as a geometric weighted average of an export-based competitiveness indicator ( $\textit{PCI}^{X}_{i}$ ) and an import-based indicator ( $\textit{PCI}^{M}_{i}$ ), the weights being $\alpha_{i}$ and ( $1-\alpha_{i}$ ):

$\displaystyle\textit{PCI}_{i}=\left[\prod^{N}_{j=1}\left(P_{ij}e_{ij}\right)^{% x_{w^{i}_{j}}}\right]^{\alpha_{i}}\cdot\left[\prod^{N}_{j=1}\left(P_{ij}e_{ij}% \right)^{m_{w^{i}_{j}}}\right]^{1-\alpha_{i}}=\left[\textit{PCI}^{X}_{i}\right% ]^{\alpha_{i}}\cdot\left[\textit{PCI}^{M}_{i}\right]^{1-\alpha_{i}}$ (6)

2.2 The new, market-based view of export-based price competitiveness

In order to construct market-restricted indicators, the starting point is the export-based PCI for country $i$ ( $\textit{PCI}^{X}_{i}$ ), referred to in Eq, (6). By taking logs of this equation and by replacing the export weights with Eq. (4), one obtains:

$\displaystyle\ln\left(\textit{PCI}^{X}_{i}\right)=\sum^{N}_{j=1}\left[\left(% \sum^{N+1}_{k=1,k\neq i}S^{j}_{k}x^{i}_{k}\right)\ \ln(e_{ij}P_{ij})\right],$ (7)

where the inner summation also considers competitive pressures of local producers by competitor $k$ in the competitor’s own market $k$ (namely the case $j=k$ ) is also considered.

By inverting the two summations and using the definition $P_{ij}\equiv P_{i}/P_{j}$ , Eq. (7) becomes:

$\displaystyle\ln\left(\textit{PCI}^{X}_{i}\right)=\sum^{N+1}_{k=1,k\neq i}x^{i% }_{k}\left[\sum^{N}_{j=1}S^{j}_{k}\ \ln\left(e_{ij}\frac{P_{i}}{P_{j}}\right)% \right],$ (8)

where recall that $S^{i}_{k}=0$ by normalization, so that $\Sigma^{N}_{j=1}{S^{j}_{k}=1}$ for any $k\neq i$ . By applying the non-arbitrage condition such that $e_{ij}=e_{ik}/e_{jk}$ , where $e_{ik}$ is the bilateral nominal exchange rate between country $i$ and outlet market $k$ ( $k$ ’s currency for one unit of $i$ ’s currency) and $e_{jk}$ is the bilateral nominal exchange rate between competitor $j$ and outlet market $k$ ( $k$ ’s currency for one unit of $j$ ’s currency), Eq. (8) can be recognized as the market-based version of the export-based PCI. Indeed, the second summation is, for a given outlet market $k$ , a weighted average of the (log) relative prices, all measured in market $k$ ’s currency, which reporting country $i$ faces in market $k$ ; the weights are given by the share of each competitor $j$ in market $k$ .7 This term thus captures the competitiveness pressures which reporting country $i$ faces in a given market $k$ (including local producers). The first summation then sums up the competitiveness pressures stemming from all markets $k\neq i$ , weighting them with the share of each outlet market $k$ in country $i$ ’s exports.

So, for example, the export-based PCI of country $i$ in euro-area markets can be computed, by appropriately restricting the outer summation, as:

$\displaystyle\ln\ \left(\textit{PCI}^{X_{\textit{EA}}}_{i}\right)=\sum_{% \begin{subarray}{c}k\epsilon\ \textit{EA}\\ k\neq i\end{subarray}}x^{i}_{k}\left[\sum^{N}_{j=1}{S^{j}_{k}}\ \ln\left(\frac% {{e_{ij}P}_{i}}{P_{j}}\right)\right].$ (9)

Here $x^{i}_{k}$ is the weight of market $k$ in $i$ ’s overall exports, but Eq. (9) can be equivalently re-written in terms of the weight of market $k$ in $i$ ’s exports to the euro area ( $x^{i,\textit{EA}}_{k}$ ) as follows:

$\displaystyle\ln\ \left(\textit{PCI}^{X_{\textit{EA}}}_{i}\right)={\beta}_{i}% \sum_{\begin{subarray}{c}k\ \epsilon\ \textit{EA}\\ k\neq i\end{subarray}}x^{i,\textit{EA}}_{k}\left[\sum^{N}_{j=1}S^{j}_{k}\ \ln% \left(\frac{{e_{ij}P}_{i}}{P_{j}}\right)\right],$ (10)

where $\beta_{i}$ is the weight of euro-area exports of country $i$ on its total exports.

The aggregate (i.e. on all markets) export-based PCI of country $i$ can thus be expressed as a weighted average of the two restricted (euro-area and non-euro area) measures, where the weights are $\beta_{i}$ and $(1-{\beta}_{i})$ , respectively:8

$\displaystyle\ln\left(\textit{PCI}^{X}_{i}\right)={\beta}_{i}\ \ln\ \left(% \textit{PCI}^{X_{\textit{EA}}}_{i}\right)+\left(1-{\beta}_{i}\right)\ \ln\ % \left(\textit{PCI}^{X_{\textit{NEA}}}_{i}\right),$ (11)

where $\textit{PCI}^{X_{\textit{NEA}}}_{i}$ is defined along the lines of Eq. (10) by letting $k$ index non-euro area markets.

Lastly, the log-version of Eqs (6) and (11) underpin a three-market view of PCIs based on the following decomposition:

$\displaystyle\ln\ \left(\textit{PCI}_{i}\right)=\left[{\alpha}_{i}{\beta}_{i}% \ \ln\ \left(\textit{PCI}^{X_{\textit{EA}}}_{i}\right)\right]+\left[{\alpha}_{% i}\left(1-{\beta}_{i}\right)\ \ln\left(\textit{PCI}^{X_{\textit{NEA}}}_{i}% \right)\right]+\left[{\left(1-{\alpha}_{i}\right)\ \ln\ \left(\textit{PCI}^{M}% _{i}\right)}\right]$ (12)

As mentioned in Section 1, this decomposition is independent of the choice of the deflator. The PCIs discussed hereon are all deflated by the PPI of manufactures sold domestically. In principle, one could explore the possibility of employing PPIs of manufactures sold in foreign (i.e. euro-area and non-euro area) markets, in order to capture price discrimination across markets, for the first and second terms in Eq. (12) and import prices for the third term. As discussed in [19], there are several arguments against this choice. First, these price indices refer by definition to traded goods, as opposed to tradables, so that in a sense they are inherently unfit for a forward-looking analysis of sustainable price developments. Second, they are available only for a limited number of countries, namely EU members and in some cases for short time spans (the same applies to overall-sales producer price indices, which are a weighted average of domestic-sales and foreign-sales producer prices). Third, these indices would still not appropriately measure pricing to market due to different countries’ invoicing currency choices. Finally, the alternative option of resorting to unit values is unpalatable since they are known to be biased deflators. In particular, changes in unit value indices may be due to variations in composition: the bias in these indices is thus tied to changes in the mix of goods exported and to the poor quality of data on quantities (e.g. [21]).

3. PCI developments in the four largest euro-area countries since 1999

In this section we appraise price-competitiveness developments in the four main euro-area countries between 1999 and 2018 (Table 1; Fig. 1, panel a) by looking at overall worldwide PCIs that take into account both import and export-based competitive pressures on all markets. We restrict the analysis in this article to post-1999 developments, although our series are available since 1992, in order to focus solely on the homogeneous period following the adoption of the euro.

Since the inception of the Economic Monetary Union France, Germany, and to a much lesser extent, Italy have recorded a gain in price competitiveness, according to the Bank of Italy’s PCIs; conversely, Spain has marked a substantial loss. Spain’s overall performance deserves a few remarks. Since 1999 this country has undergone a process of significant catching-up, reflected in its stellar goods export performance (e.g. [18]). The disconnect between the rise in its foreign sales and its unfavourable price-competitiveness developments, when measured on a PPI basis, have been marked as the “Spanish paradox”. [22] shed light on this disconnect by showing that non-price competitiveness factors, such as company size, R&D spending, product diversification, product and process innovation were behind Spain’s high export growth since 1999.

Price-competitiveness developments are, however, found to be heterogeneous across sub-periods. In the years prior to the global financial crisis, all countries except Germany lost price competitiveness. Germany is a clear outlier relative to its peers, and its price-competitiveness gain in this period has been object of much discussion. The governance structure of German labour market institutions, underpinned by flexible contracts and mutual agreements between trade unions, employer associations and work councils, which led to the decentralization of wage bargaining to the firm level, has been highlighted as a key explanation of Germany’s wage, and therefore price, moderation in this period (e.g. [23]); other explanations refer to a very intense offshoring of its production processes relative to its main trading partners (e.g. [12]). The recessionary years then featured price-competitiveness gains in all four countries, due to both cyclical factors and to structural reforms that were pushed through in these, and the following, years (e.g. [14]). The post-2010 years were characterised by further gains for all countries bar Spain, which went back to experiencing a slight loss.

Table 1
Price-competitiveness developments in the four largest euro-area countries by sub-period (percentage changes of annual averages)

	France	Germany	Italy	Spain
1999–2007	1.1	$-$ 3.8	3.9	9.1
2007–2010	$-$ 6.0	$-$ 5.7	$-$ 4.6	$-$ 1.8
2010–2018	$-$ 3.4	$-$ 0.9	$-$ 1.2	2.0
1999–2018	$-$ 8.2	$-$ 10.1	$-$ 2.1	9.3

Source: Authors’ calculations. Notes: A positive (negative) change indicates a loss (gain) in price competitiveness.

Figure 1.

Price competitiveness in the four largest euro-area countries. a. Developments (indices 1999 $=$ 100; quarterly averages); b. Differentials with respect to Germany (differences between indices 1999 $=$ 100; quarterly averages). Source: Authors’ calculations. Notes: An increase (a decrease) in the PCI indicates a loss (gain) in price competitiveness in the left hand-side panel. A positive (negative) differential indicates that since 1999 the country has incurred a cumulative loss (gain) in price competitiveness relative to Germany in the right hand-side panel.

Given these dynamics, vis-à-vis Germany the three other euro-area countries progressively lost ground until 2008 (Fig. 1, panel b). The gaps temporarily closed during the Great Recession, but then widened again until 2012–2013. Since then, the differentials relative to Germany have decreased, thereby confirming that after the sovereign debt crisis across Europe a recovery in competitiveness vis-à-vis Germany was determined by relative-price adjustments. Focusing on Italy and Spain, between 1999 and the third quarter of 2012 their PCIs had cumulated a competitive loss of 10.1 and 21.1 percentage points, respectively, relative to Germany. The shortfall reached a minimum in the first quarter of 2016 to end up at 7.9 and 19.1 percentage points, respectively, in the fourth quarter of 2018.9

4. Moving from a competitor-based view to a market-based view of export-based price competitiveness

In addition to aggregate indicators computed vis-à-vis all 60 trading partners, the Bank of Italy, similarly to the ECB, provides indicators vis-à-vis subsets of competitors, for instance the euro-area members. Only the set of trading partners is restricted; they are still assumed to compete, as is the case for the aggregate measure, on all markets.

In Section 2 we have provided the algebra to develop a novel set of indicators. Instead of being competitor-based, these measures are market-based: they capture competitiveness dynamics of a given reporting country vis-à-vis all competitors in selected countries/areas (“markets”). These indicators refer only to export-based competitiveness (that is, to PCIs that are constructed using solely export weights and not also import weights, as is the case of the overall PCIs described in the previous section) since a market-based disaggregation is not meaningful for import competition, which by definition only takes place in the domestic market. The following example helps clarifying the difference between the competitor-based and the market-based disaggregations. The competitor-based measure for Italy vis-à-vis euro-area partners simply relates Italy’s PPIs to those of euro-area competitors, as the corresponding nominal exchange rates are fixed to unity, and weighs the resulting relative prices across all markets. Conversely, the market-based measure for Italy in euro-area markets compares Italy’s PPIs to those of all 60 competitors, taking into account the corresponding exchange rates, yet the weighting only refers to euro-area markets.

To our knowledge, no other institution produces market-based indicators, yet they are essential to shed light on the drivers of export growth in selected markets. Indeed, they have evolved differently at business-cycle frequencies compared to competitor-based measures (Fig. 2). On the basis of this newly developed set of indicators, it is possible to compare export-based competitiveness dynamics of Italy, France, Germany and Spain in euro-area markets, as opposed to those in non-euro area markets (Table 2; Fig. 3).

Table 2
Export-based price-competitiveness developments in euro-area and non-euro area markets by sub-period (percentage changes on annual averages)

	Euro-area markets				Non-euro area markets
	France	Germany	Italy	Spain	France	Germany	Italy	Spain
1999–2007	$-$ 1.5	$-$ 5.6	2.4	7.0	5.4	$-$ 0.7	7.1	13.1
2007–2010	$-$ 5.0	$-$ 3.7	$-$ 2.5	$-$ 0.5	$-$ 9.2	$-$ 8.0	$-$ 7.3	$-$ 4.9
2010–2018	$-$ 3.2	$-$ 0.5	$-$ 0.9	2.4	$-$ 3.4	$-$ 1.2	$-$ 1.3	1.9
1999–2018	$-$ 9.5	$-$ 9.6	$-$ 1.1	8.9	$-$ 7.6	$-$ 9.8	$-$ 2.0	9.6

Source: Authors’ calculations. Notes: A positive (negative) change indicates a loss (gain) in price competitiveness.

Figure 2.

Export-based price-competitiveness developments: market-based vs. competitor-based disaggregation a. Euro area markets/competitors (indices 1999 $=$ 100; quarterly averages); b. Non-euro area markets/competitors (indices 1999 $=$ 100; quarterly averages). Source: Authors’ calculations. Notes: An increase (decrease) in the PCI signals a loss (gain) in price competitiveness

Figure 3.

Export-based price-competitiveness developments in euro-area and non-euro area markets. a. Euro-area markets (indices 1999 $=$ 100; quarterly averages); b. Non-euro area markets (indices 1999 $=$ 100; quarterly averages). Source: Authors’ calculations. Notes: An increase (decrease) in the PCI indicates a loss (gain) in price competitiveness.

First, the export-based price-competiveness performance in the two sets of markets is found to be very heterogeneous across the four countries. Italy’s overall gain since 1999 was driven by a modest improvement in both euro and non-euro area markets, yet to a larger extent in the latter (2 percentage points). Spain’s performance was poor in both sets of destinations, yet more so in non-euro area markets (where it lost nearly 10 percentage points). France’s gain was higher in euro-area markets (9.5, against 7.6 in non-euro area markets), whereas Germany’s was very balanced between the two sets of markets, gaining almost 10 percentage points from each.

Seen from a different angle, in the overall 1999–2018 period in both euro and non-euro area markets France, Germany and Italy achieved a price-competitiveness gain (although to a much lower extent for Italy) against a sharp loss for Spain. These results would suggest that the non-price factors found to drive Spain’s favourable export dynamics since 1999, mentioned in Section 3, were relevant both for sales to the euro area and to those outside the euro area.

Second, developments again differ by sub-period. In the years prior to the eruption of the global financial crisis, France, Italy and Spain lost export competitiveness in non-euro area markets, in connection with the appreciation of the nominal effective exchange rate of the euro (as sourced by the ECB), whereas Germany recorded a broad stability. In euro-area markets competitiveness developments mirrored those of relative prices among the four economies; France and, more starkly, Germany ended up improving their competitiveness, whereas Italy and, more significantly, Spain reported a deterioration. During the Great Recession all four countries recorded competitiveness gains in both sets of export markets. After 2010 the substantial improvement continued in non-euro area markets, where the weight of non-euro area competitors is larger and where therefore the benefit of the nominal depreciation of the euro was greater, except for Spain. France stood out as marking the largest improvements in both sets of markets. This has led to a “French puzzle”, given the country’s recent modest export performance relative to its main euro-area peers despite its favourable price-competitiveness developments. [24] argue that one plausible explanation is the shift abroad of many French enterprises, which is not captured by standard macroeconomic PCIs. In the period 2010–2018 Germany and Italy’s gains were comparable in the non-euro area market, whereas Italy’s was relatively larger in the euro-area market.

Finally, in euro-area markets the export-based competitiveness gaps of France, Italy and Spain relative to Germany rose until 2008, to a varying and increasing extent across the three countries (Fig. 4, panel a). The differential continued to widen in Spain until 2013, broadly stabilising thereafter. Conversely, after peaking in the course of 2008, the gap levelled out in Italy and decreased in France. On average in 2018 Spain’s gap vis-à-vis Germany was of nearly 19 percentage points, Italy’s stood at 8.5 points, whereas France even recorded a tiny advantage over Germany.

Figure 4.

Export-based price-competitiveness differentials with respect to Germany. a. Euro-area markets (differences in indices 1999 $=$ 100; quarterly averages); b. Non-euro area markets (differences in indices 1999 $=$ 100; quarterly averages). Source: Authors’ calculations. Notes: A positive (negative) differential indicates that since 1999 the country has incurred a cumulative loss (gain) in price competitiveness relative to Germany.

Concerning non-euro area markets, developments vis-à-vis Germany were similar, for each country, to the corresponding dynamics observed in euro-area markets, although the size of the gap was different in Italy and in France (Fig. 4, panel b). In particular, on average in 2018 Italy’s export-based competitiveness differential was only slightly smaller than that reported for euro-area markets, standing at 7.8 percentage points, whereas France’s gap was negative, albeit contained (over 2 points).

Figure 5.

A decomposition of overall price competitiveness by market. a. France; b. Germany; c. Italy; d. Spain. (percentage-point contributions; averages by period). Source: Authors’ calculations. Notes: Positive (negative) bars indicate a loss (gain) in price competitiveness. Each bar is obtained by adding up the annual log changes of the corresponding sub-indicator, thereby disregarding compounding.

5. A three-market view of overall PCIs

The overall PCI of a given reporting country can be decomposed into three components (the three square brackets in Eq. (12) in Section 2), each referring to a different market: a) the domestic market, in which domestic producers compete with foreign exporters; b) euro-area export markets, in which all countries compete against each other; c) non-euro area markets, in which again all countries compete against each other. Figure 5 depicts the contribution of each component by sub-period in the four main euro-area countries.

Several facts stand out. First, all three components generally display the same sign (within country and sub-period), so that export-based competitiveness developments (i.e. the sum of the two differently-shaded grey bars in Fig. 5) are magnified by those in the domestic market. Viewed from a different angle, this also means that differences across countries in overall competitiveness developments are larger than differences in export-based competitiveness alone. Second, the contribution of export-based competitiveness in euro-area markets tends to be the smallest in absolute value across the four countries. This is plausibly due to the fact that in euro-area markets the weight of euro-area competitors (including local producers), against which the nominal exchange rate is fixed, is very high, and therefore the resulting indicators are more stable over time than the non-euro area export-based measure, as seen in Fig. 3. Third, in each sub-period non-euro area export markets typically turn out to be the main driver of overall price-competitiveness developments in Italy, France and Germany. Turning to the overall 1999–2018 period, instead, competitiveness in the domestic market and export-based competitiveness in extra-euro area markets are the main drivers of, and contribute roughly equally to, aggregate price-competitiveness developments in Italy and in Germany.

6. Conclusions

This article develops a novel decomposition of price-competitiveness indicators (PCIs), which enables to correctly disentangle export-based competitiveness developments in euro-area vs. non-euro area markets, as well as those in the domestic market. We implement this decomposition within the set of PCIs computed by the Bank of Italy, which uses domestic-sales manufacturing producer prices as the relevant deflator. According to these new measures and focusing on the four main euro-area countries, whereas Germany’s substantial gain in export-based price competitiveness in the 1999–2018 period is found to be quite balanced between euro and non-euro markets, in France it was relatively larger in euro-area markets. In contrast, Italy’s mild improvement was larger in the non-euro area market, whilst Spain’s deterioration was smaller in euro-area markets. These findings point to the usefulness of considering market breakdowns when analysing price-competitiveness developments, since they can differ significantly across export outlets. Finally, in the overall 1999–2018 period competitiveness in the domestic market and export-based competitiveness in non-euro area markets equally shaped the aggregate competitiveness developments observed in Italy and in Germany.

A note of caution to the analysis here presented is warranted. Bank of Italy’s PCIs are based on one particular price index, yet, as discussed and shown in Fig. A1 in Annex A, it is now common knowledge that competitiveness developments vary significantly according to the price or cost index employed to deflate NEERs or nominal PCIs. In general, therefore, a complete assessment of competitiveness trends should be based on the examination of a set of differently deflated PCIs.

Footnotes

We are grateful to the editor Charles Renfro and to an anonymous referee, as well as to Andrea Brandolini, Justas Dainauskas, Silvia Fabiani, Martin Schmitz, Roberto Tedeschi, Francesco Zollino and all participants of the EER-HCI workshop held at the European Central Bank in January 2019 for comments to previous versions of this article. Any error is the responsibility of the authors. The views here presented are those of the authors, and not of the institution represented. This article is based on the 8th July 2019 Bank of Italy data vintage and on R version 3.3.3 software.

See [] for a specific comparison between the CPI and the PPI.

As for the ability of alternative PCIs to explain export dynamics in the four main euro-area countries, see [12]. More recent research has further pointed out that export performance is better understood when considering relative PPIs (or relative unit labour costs) in conjunction with profit margins [], so that PCIs alone, however deflated, are insufficient to explain export developments.

On the one hand, enterprises choose ex ante which outlays to penetrate, how much to sell and at what price, according to their pricing strategies and to the strength of demand in the various destination markets (see, for example [16]). On the other hand, not all firms export to all markets, and especially small and less innovative firms typically export only to the closest markets, so as to contain trading costs. For some stylised facts concerning external trade and productivity across EU countries based on micro-aggregated data, see [].

This is of course a simplification of reality, but it is anyhow standard in the construction of price-competitiveness measures (see, for example []).

It is noteworthy that, as in [], country $i$ ’s weight in market $k$ is not considered, in that the weights $S^{j}_{k}$ are defined as the share of competitor $j$ in market $k$ net of country $i$ . In other terms, when $j=i$ in the second summation, the addendum is zero.

See Annex B for details on the computation of the weights $\beta_{i}$ , which for the four main euro-area countries range between 0.43 (Germany) and 0.62 (Spain) in 1999–2001 and between 0.40 (Germany) and 0.58 (Spain) in 2009–2011.

On average in 2018 the competitiveness gap vis-à-vis Germany stood at 19.4 points in Spain, 8.1 in Italy and 1.9 in France.

Annex A. A comparison of different methods used by selected organizations to calculate PCIs

Updating a similar table in [1], in Table A1 we provide a summary of the different methodologies underlying the computation of overall PCIs across international institutions, as well as details on any available sub-indicators.

Figure A1 provides developments of PCIs of the four largest euro-area countries based on different price and cost indices since 1999.

Table A1

A comparison of different methods to calculate price-competitiveness indicators

Institution	Trade Basis	Maximum number of trading partners	Weighting methods	Update of weights	Indicators provided	Deflators	References
Bank of Italy	Manufactured goods (SITC 5–8)	60	Weighted average of import and double export weights (including weighted third market effects)	Average weight fixed over three years, updated every ten years	Overall PCIs; import-based and export-based PCIs; euro and non-euro area trade partner-weighted PCIs; euro and non-euro area market export-based PCIs	PPI	This paper and Felettigh et al. (2016), and indicators available at: https://www.banc aditalia.it/statistiche/tematiche/ rapporti-estero/competitivita/ index.html?com.dotmarketing. htmlpage.language=1
European Central Bank	Manufactured goods (SITC 5–8)	58 (CPI only) 38	Weighted average of import and double export weights (including weighted third market effects)	Average weight fixed over three years, updated every three years	Overall PCIs; import-based and export-based PCIs; euro and non-euro area trade partner-weighted PCIs	CPI, PPI, GDP deflator, ULCM, ULCT (PPI and ULCM not publicly available)	Schmitz et al. (2012)

Table A1. continued
Institution	Trade Basis	Maximum number of trading partners	Weighting methods	Update of weights	Indicators provided	Deflators	References
European Commission DG ECFIN	Total goods	41	Double export weights (including weighted third market effects)	Annual weights, updated yearly	Overall PCIs	CPI, GDP deflator, Export prices, ULCM, ULCT	European Commission (2014)
Bank of International Settlements	Manufactured goods (SITC 5–8)	59	Weighted average of import and double export weights (including weighted third market effects)	Average weight fixed over three years, updated every three years	Overall PCIs	CPI	Klau and Fung (2006) and subsequent update available at http://www. bis.org/statistics/eer/
International Monetary Fund	Manufactured goods (SITC 5–8)
Manufactured goods (SITC 5–8), commodities (overall weight in global markets), and services (same weights as manufacturing except for countries where tourism is important)	26 163	Weighted average of import and double export weights (including weighted third market effects)
Weighted average of import and double export weights (including weighted third market effects) for manufacturing, simple weights for other categories	Average weight fixed over three years, updated at irregular intervals	Overall PCIs	ULC CPI	Bayoumi et al. (2005)

Table A1. continued
Institution	Trade Basis	Maximum number of trading partners	Weighting methods	Update of weights	Indicators provided	Deflators	References
Federal Reserve Board	Total goods (excluding gold and military items from exports, oil from imports when possible)	25	Fixed weighted average of import and double export weights (including third market effects)	Annual weights, updated yearly	Overall PCIs	CPI	Loretan (2005)
Bank of England	Manufactured goods and services	42	Fixed weighted average of import and double export weights (including third market effects for manufacturing)	Annual weights, updated yearly	Overall PCIs	CPI	Lynch and Whitaker (2004) and subsequent update available at http:// www.bankofengland.co.uk/ statistics/Pages/eri/neweri. aspx

Figure A1.

PCI developments in the four main euro-area countries according to different deflators (indices 1999Q1 $=$ 100; quarterly averages for PPI and CPI-based PCIs). a. France; b. Germany; c. Italy; d. Spain.

Annex B. Additional information on the Bank of Italy’s PCIs

Concerning the price index employed to deflate Bank of Italy’s PCIs, the following hierarchy of sources for producer price indices (PPIs) is employed: Eurostat/ECB, OECD, IMF (International Financial Statistics) and Datastream. For all OECD countries, and for the most recent years at least, PPIs of manufactured goods sold in domestic markets are used; for all remaining countries and/or previous periods manufacturing, industrial or total PPIs or wholesale price indices (WPIs) are employed,10

¹⁰

Wholesale price indices have a less broad coverage of products and industries than PPIs, and are less conceptually aligned with the system of national accounts [9].

whereas consumer price indices (CPIs) are adopted only as a last resort for Algeria, Nigeria and Saudi Arabia (see all details in Table B1).

Table B1

The price indices and sources underlying the Bank of Italy’s 61 PCIs

	Country	Deflator	Source
Euro-area	Austria	PPI, domestic sales of manufactures	ECB
countries	Belgium	PPI, domestic sales of manufactures	Eurostat
	Cyprus	PPI/PPI, domestic sales of manufactures	IMF/ Eurostat
	Estonia	CPI/PPI/PPI, domestic sales of manufactures	IMF/IMF/ Eurostat
	Finland	PPI, domestic sales of manufactures	ECB
	France	PPI, domestic sales of manufactures	ECB
	Germany	PPI, domestic sales of manufactures	ECB
	Greece	PPI, domestic sales of manufactures	ECB
	Ireland	PPI, domestic sales of manufactures	ECB
	Italy	PPI, domestic sales of manufactures	Istat
	Latvia	CPI/PPI/PPI, domestic sales of manufactures	IMF/IMF/ Eurostat
	Lithuania	CPI/PPI/PPI, domestic sales of manufactures	IMF/IMF/ Eurostat
	Malta	CPI/PPI, domestic sales of manufactures	IMF/ Eurostat
	Netherlands	PPI, domestic sales of manufactures	ECB
	Portugal	PPI, domestic sales of manufactures	Eurostat
	Slovakia	PPI, dom. sales of ind. goods /PPI/PPI, dom. sales of manuf.	ECB/Eurostat
	Slovenia	PPI, domestic sales of manufactures	ECB
	Spain	PPI, domestic sales of manufactures	ECB
Non Euro-area	Bulgaria	CPI/PPI/PPI, domestic sales of manufactures	IMF/IMF/ECB
EU countries	Croatia	CPI/PPI/PPI, domestic sales of manufactures	IMF/IMF/ECB
	Czech Republic	PPI, domestic sales of manufactures/	ECB
	Denmark	PPI, domestic sales of manufactures	OECD/ECB
	Hungary	PPI/PPI, domestic sales of manufactures	IMF/ECB
	Poland	PPI/PPI, tot. sales of manuf./PPI, dom. sales of manuf.	IMF/OECD/ECB
	Romania	PPI/PPI, domestic sales of manufactures	IMF/ECB
	Sweden	PPI/PPI, tot. sales of manuf./PPI, dom. sales of manuf.	IMF/OECD/ECB
	United Kingdom	PPI/PPI, domestic sales of manufactures	IMF/OECD
Non EU countries	Algeria	CPI	IMF
	Argentina (*)	CPI/PPI/PPI	IMF/IMF/Datastream
	Australia (*)	PPI/PPI, total sales of manufactures	IMF/OECD
	Brazil	PPI, industry	Central Bank of Brasil


Country	Deflator	Source
Canada	PPI, total sales of manufactures	OECD
Chile	WPI/PPI, total sales of manufactures	IMF/OECD
China	CPI/PPI/PPI	IMF/National Bureau of Statistics China/IMF
Colombia	PPI/PPI, domestic sales of manufactures	IMF/OECD
Ecuador	WPI/CPI/PPI	IMF/IMF/IMF
Hong Kong SAR (*)	CPI/PPI	IMF/IMF
India	WPI	IMF
Indonesia	WPI/ WPI of total manufactures	IMF/OECD
Israel	WPI/PPI, domestic sales of manufactures	IMF/OECD
Japan	PPI/PPI, domestic sales of manufactures	IMF/OECD
Kuwait	WPI	IMF
Malaysia	CPI/PPI	IMF/IMF
Mexico	PPI/PPI, domestic sales of manufactures	IMF/OECD
Morocco	WPI/CPI/PPI	IMF/IMF/IMF
New Zealand (*)	PPI, domestic sales of manufactures	OECD
Nigeria	CPI	IMF
Norway	PPI, domestic sales of manufactures	OECD
Pakistan	WPI	IMF
Peru	WPI	IMF
Philippines	CPI/PPI	IMF/IMF
Russia	PPI/PPI, domestic sales of manufactures	IMF/OECD
Saudi Arabia	CPI	IMF
Singapore	WPI	IMF
South Africa	PPI, domestic sales of manufactures	OECD
South Korea	PPI, domestic sales of manufactures	OECD
Switzerland	PPI/PPI, domestic sales of manufactures	IMF/OECD
Taiwan	WPI	Datastream
Thailand	PPI	IMF
Turkey	PPI, dom. sales of ind. goods/ PPI, dom. sales of manuf.	OECD/OECD
United States	PPI, total sales of manufactures	OECD

Notes: PPI $=$ producer price index; CPI $=$ consumer price index; WPI $=$ wholesale price index. The indices and relative sources are listed in chronological order, such that the last index and source refers to the data employed for the most recent (country-varying) period. (*) Quarterly data are converted into monthly series for certain periods.

Bank of Italy’s trade weights are based on flows of manufactured goods only, which are less subject to non-market practices, typical of agricultural goods, and to large price volatility, as is the case for raw commodities. Two matrices of fixed weights are used: the first, based on 1999–2001 bilateral trade data for 61 countries and their exports to the residual aggregate “rest of the world”, is employed to calculate PCIs for the January 1992 – December 2004 period; the second matrix, based on 2009–2011 data, is employed for indicators as of January 2005. The two series are then chain-linked in January 2005.11

¹¹

As a result, whenever the period of interest includes January 2005, PCIs change also due to the switch between the two sets of weights.

The same approach holds for the computation of the weights

{\alpha}_{i}

and

{\beta}_{i}

in equations [2, 11] of the main text, respectively.

The trade weights, as well as ${\alpha}_{i}$ and ${\beta}_{i}$ , are based on the following sources: Eurostat bilateral export data for EU countries; United Nations Comtrade (UN) export data for non-EU countries; CEPII-BACI data for Taiwan and for other missing observations. In all cases the 60-by-60 matrix of bilateral exports of a given country, together with its overall exports, was retrieved so that exports to the 61st partner “rest of the world” were computed residually. As in [2], local producers’ sales in their domestic market are approximated by the difference of the country’s manufacturing value added (VA) and its net manufacturing imports. This method is based on the assumption that manufacturing goods imports are a reasonable approximation of the total value of intermediate inputs of foreign origin, in turn reflecting the high and growing degree of internationalization of manufacturing production. Current-price manufacturing VA data for all 61 countries, except Taiwan and China, are retrieved from United Nations Statistics. Data for Taiwan are sourced from WIOD for 1999–2001 and from the Taiwan Statistical Bureau for 2009–2011; data for China for 1999–2001 are taken from World Bank data. As for manufacturing imports, the hierarchy of sources used for (bilateral) exports is confirmed.

Table B2

The weight of selected countries in the four largest euro-area economies’ trade baskets

A. 1999–2001
Country	Italy			France			Germany			Spain
	Import weights	Double export weights	Overall weights	Import weights	Double export weights	Overall weights	Import weights	Double export weights	Overall weights	Import weights	Double export weights	Overall weights
Bulgaria	0.4	0.2	0.3	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
Brazil	0.7	1.1	0.9	0.3	1.0	0.6	0.3	1.0	0.7	0.3	1.2	0.7
Canada	0.3	1.2	0.8	0.3	1.2	0.8	0.3	1.2	0.8	0.2	0.7	0.4
Switzerland	3.7	2.6	3.1	2.7	2.7	2.7	4.4	3.1	3.7	1.9	1.7	1.8
Colombia	0.0	0.1	0.1	0.0	0.1	0.1	0.0	0.1	0.1	0.0	0.2	0.1
Denmark	0.5	0.9	0.7	0.6	1.0	0.8	1.7	1.3	1.5	0.6	0.8	0.7
United Kingdom	7.2	7.3	7.3	9.0	9.0	9.0	7.6	8.4	8.1	8.3	8.4	8.3
Hong Kong SAR	1.3	1.3	1.3	1.3	1.4	1.4	1.8	1.4	1.6	1.2	1.1	1.2
Croatia	0.5	0.4	0.4	0.0	0.1	0.1	0.2	0.2	0.2	0.0	0.1	0.1
Hungary	0.9	0.9	0.9	0.6	0.7	0.6	2.5	1.2	1.8	0.4	0.6	0.5
Japan	3.3	4.7	4.1	2.7	4.8	3.7	4.5	5.2	4.9	2.6	3.4	3.0
Morocco	0.2	0.3	0.2	0.8	0.7	0.7	0.1	0.2	0.1	0.4	0.8	0.6
Mexico	0.1	1.0	0.6	0.1	0.9	0.5	0.4	1.2	0.8	0.2	1.2	0.6
Poland	1.1	1.4	1.3	0.6	1.1	0.9	2.5	2.0	2.2	0.4	1.0	0.6
Russia	0.4	1.1	0.8	0.1	0.8	0.5	0.4	1.3	0.9	0.1	0.6	0.3
United States	5.5	12.3	9.3	7.5	12.1	9.8	6.8	13.5	10.6	3.9	8.6	6.0
South Africa	0.3	0.5	0.4	0.1	0.5	0.3	0.5	0.6	0.6	0.2	0.4	0.3
Sum of these countries	26.3	37.3	32.5	27.0	38.1	32.6	34.0	42.2	38.7	20.9	30.8	25.3
Remaining extra-euro area countries	12.6	17.0	15.1	8.9	16.8	12.9	15.3	17.9	16.8	9.5	14.1	11.5
Euro-area countries ${}^{(1)}$	61.1	45.7	52.5	64.2	45.1	54.5	50.8	39.9	44.5	69.6	55.1	63.2
Total	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0
Weight of the export-based PCI in the overall PCI ( $\alpha$ )	56.3			57.2			50.7			44.1

Table B2, continued
B. 2009–2011
Country	Italy			France			Germany			Spain
	Import weights	Double export weights	Overall weights	Import weights	Double export weights	Overall weights	Import weights	Double export weights	Overall weights	Import weights	Double export weights	Overall weights
Bulgaria	0.6	0.3	0.4	0.2	0.2	0.2	0.3	0.2	0.2	0.1	0.2	0.2
Brazil	0.5	1.3	0.9	0.2	1.1	0.7	0.4	1.2	0.8	0.5	1.3	0.9
Canada	0.2	0.8	0.6	0.4	0.9	0.6	0.3	1.0	0.7	0.2	0.6	0.4
Switzerland	4.1	2.9	3.5	2.9	2.5	2.7	4.3	3.0	3.6	2.7	2.1	2.4
Colombia	0.1	0.2	0.1	0.0	0.2	0.1	0.0	0.1	0.1	0.0	0.2	0.1
Denmark	0.6	0.7	0.6	0.6	0.8	0.7	1.3	1.1	1.2	0.8	0.7	0.7
United Kingdom	3.9	4.8	4.4	4.4	5.7	5.0	4.4	5.8	5.2	5.0	5.3	5.2
Hong Kong SAR	1.2	1.2	1.2	0.9	1.3	1.1	1.4	1.5	1.4	0.9	0.9	0.9
Croatia	0.5	0.4	0.5	0.1	0.1	0.1	0.1	0.2	0.2	0.0	0.1	0.1
Hungary	1.3	1.1	1.2	1.0	1.0	1.0	2.9	1.2	1.9	1.3	0.9	1.1
Japan	1.6	3.3	2.5	1.4	3.7	2.5	2.4	3.8	3.2	1.3	2.3	1.8
Morocco	0.2	0.4	0.3	0.7	0.7	0.7	0.0	0.2	0.1	1.1	1.3	1.2
Mexico	0.2	0.9	0.6	0.1	0.8	0.4	0.4	0.9	0.7	0.2	1.1	0.6
Poland	2.8	2.4	2.5	2.0	2.0	2.0	4.6	3.0	3.7	1.7	1.9	1.8
Russia	1.2	2.3	1.8	0.1	1.8	0.9	0.5	2.7	1.8	0.1	1.1	0.6
United States	3.2	7.8	5.7	3.4	8.2	5.7	4.6	9.0	7.2	3.0	5.8	4.3
South Africa	0.2	0.6	0.4	0.1	0.6	0.4	0.5	0.7	0.6	0.3	0.5	0.4
Sum of these countries	22.2	31.3	27.2	18.4	31.7	24.7	28.3	35.7	32.6	19.2	26.4	22.6
Remaining extra-euro area countries	22.0	27.0	24.7	15.2	27.8	21.2	23.2	29.0	26.6	17.6	22.4	19.8
Euro-area countries ${}^{(1)}$	55.8	41.7	48.1	66.4	40.5	54.1	48.5	35.3	40.8	63.2	51.2	57.6
Total	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0
Weight of the export-based PCI in the overall PCI ( $\alpha$ )	54.4			58.6			47.4			46.6

Source: [1, 19, 20]. Notes: (1) Luxembourg is not included.

References

Felettigh

Giordano

Oddo

Romano

. New indicators to assess price competitiveness developments in the four largest euro-area countries and in their main trading partners. J Econ Soc Meas. 2016; 41(3): 203-235.

Schmitz

De Clercq

Fidora

Lauro

Pinheiro

. Revisiting the effective exchange rates of the euro. J Econ Soc Meas. 2013; 38: 127-158.

Bayoumi

Lee

Jayanthi

. New rates from new weights. IMF Staff Papers. 2006; 53(2).

Klau

Fung

. The new BIS effective exchange rate indices. BIS Quarterly Review. 2006; March.

Lynch

Whitaker

. The new sterling ERI. Bank of England Quarterly Bulletin. 2004; Winter.

Loretan

. Indexes of the foreign exchange value of the dollar. Federal Reserve Bulletin. 2005; Winter: 1-8.

Turner

dack

Van’t

. Measuring international price and cost competitiveness. BIS Economic Papers. 1993; 39.

Chinn,

. A primer on real effective exchange rates: determinants, overvaluation, trade flows and competitive devaluation. Open Economies Review. 2006; 17: 115-143.

Berry,

Graf

Stanger

Yla-Jarkko

. Price Statistics Compilation in 196 Economies: The Relevance for Policy Analysis. IMF Working Paper. 2019; 163.

10.

ECB. Developments in the Euro Area’s International Cost and Price Competitiveness. Monthly Bulletin. 2003; August: 67-74.

11.

Deutsche Bundesbank. Purchasing Power Parity Theory as a Concept for Evaluating Price Competitiveness. Monthly Report. 2004; June: 29-42.

12.

Giordano

Zollino

. Shedding Light on Price- and Non-price Competitiveness Determinants of Foreign Trade in the Four Largest Euro-area Countries. Rev Int Econ. 2016; 24(3): 604-634.

13.

Amici

Bobbio

Torrini

. Patterns of convergence (divergence) in the euro area: profitability versus cost and price indicators. Italian Economic Journal. 2018; 4(3): 367-384.

14.

Fidora

Giordano

Schmitz

. Real exchange rate misalignments in the euro area. Bank of Italy Working Papers. 2018; 1162.

15.

Giordano

. Price and cost competitiveness misalignments of the euro area and of its main economies according to a quarterly BEER model, 1999-2017. Bank of Italy Occasional Papers. 2018; 444.

16.

Bernard

Jensen

Redding

Schott

. Global Firms. J Econ Lit. 2018; 56(2): 565-619.

17.

Giordano

Lopez-Garcia

. Firm heterogeneity and trade in EU countries: A cross-country analysis. ECB Occasional Papers. 2019; 225.

18.

Bugamelli

Fabiani

Federico

Felettigh

Giordano

Linarello

. Back on track? a macro-micro narrative of italian exports. Italian Economic Journal. 2018; 4: 1-31.

19.

Felettigh

Giordano

Oddo

Romano

. Reassessing price-competitiveness indicators of the four largest euro-area countries and of their main trading partners. Bank of Italy Occasional Papers. 2015; 280.

20.

Felettigh

Giordano

. Revisiting prices and markets underlying the Bank of Italy’s price-competitiveness indicators. Bank of Italy Occasional Papers. 2018; 447.

21.

Schott

. Across-product versus within-product specialization in international trade. Quarterly Journal of Economics. 2004; 119(2): 647-678.

22.

Correa-López

Doménech

. The Internationalisation of Spanish Firms. BBVA Working Papers. 2012; 30.

23.

Dustmann

Fitzenberger

Schonberg

Spitz-Oener

. From Sick Man of Europe to Economic Superstar: Germany’s Resurgent Economy. Journal of Economic Perspectives. 2014; 28: 167-88.

24.

Emlinger

Jean

Vicard

. L’étonnante atonie des exportations françaises: retour sur la compétitivité et ses déterminants. CEPII Policy Brief, 2019; 24.

A novel three-market view of price competitiveness

Abstract

Keywords

1. Introduction2

2.1 A recap of the current methodology

Table 1 Price-competitiveness developments in the four largest euro-area countries by sub-period (percentage changes of annual averages)

Table 2 Export-based price-competitiveness developments in euro-area and non-euro area markets by sub-period (percentage changes on annual averages)

6. Conclusions

Footnotes

Annex A. A comparison of different methods used by selected organizations to calculate PCIs

Annex B. Additional information on the Bank of Italy’s PCIs

References

Table 1
Price-competitiveness developments in the four largest euro-area countries by sub-period (percentage changes of annual averages)

Table 2
Export-based price-competitiveness developments in euro-area and non-euro area markets by sub-period (percentage changes on annual averages)