VoxEU Column Exchange Rates International trade

Why isn't the use of yen more widespread? A puzzle of invoicing currency choice in Japanese exporters

Why do so many Japanese firms risk their profits by invoicing their exporters in foreign currencies? This column, based on firm-level interview data, suggests that such invoicing may be motivated by the fact that many Japanese firms export their goods to foreign subsidiaries facing local competitors. Invoicing in foreign currencies concentrates currency risk at the company headquarters.

Whenever the yen appreciates substantially, Japanese exporters scream and the government becomes under pressure to do something. This has been the story many times in the past, and the story is repeating itself now. That raises the question: Why don’t Japanese firms invoice their exports in yen to avoid sudden losses of profits?

Japan is well known for its unique pattern of trade invoicing. According to the "stylised facts" on the choice of invoice currency, which are based on 1970s empirical research such as Grassman (1973) and Page (1977), trade between two advanced countries tends to be invoiced in the exporter's currency. Trade between advanced and developing countries are generally invoiced in the advanced country’s currency.

Japan’s currency invoicing pattern puzzlingly differs from these stylised facts.

  • First, Japanese exporters have a strong tendency to choose the importer’s currency for their exports to advanced countries such as the US and EU.
  • Second, dollar invoicing is prevalent in Japan’s exports to Asia. As shown in Figure 1, the share of dollar invoicing is higher than that of yen invoicing in exports to Asia from 2006 to 2009.

Figure 1. The share of the yen and dollar invoicing in Japan’s exports (%)

a) Exports to World

b) Exports to Asia



Notes: The data for 1999 is not available. The September data is used for 1992-97, the March data for 1998, and the 2nd half of the year data for 2000-09. Source: Bank of Japan, Yushutsu Shinyojo Tokei (Export Letter of Credit Statistics); MITI, Yushutsu Kakunin Tokei (Export Confirmation Statistics); MITI, Yushutsu Hokukosho Tukadate Doko (Export Currency Invoicing Report); MITI, Yushutsu Kessai Tsukadate Doko Chosa (Export Settlement Currency Invoicing); the website of Japan Customs.

The plot thickens

The puzzle takes on another dimension when we consider the internationalising process of Japanese firms and the yen over the last three decades.

On one hand, since the Foreign Exchange and Controls Law was amended in 1980, numerous attempts have been made to liberalise internal and external financial transactions in Japan for the purpose of promoting the yen’s internationalisation further, especially the use of the yen in trade transactions. The yen became a fully convertible currency and is still the only currency in Asia that meets the full convertibility.

On the other hand, in response to the sharp appreciation of the yen with respect to the dollar in the mid-1980s, Japanese firms accelerated the shift toward overseas production, and cross-border intra-firm transactions increased substantially. Japanese firms have expanded overseas business especially in Asia and have built a regional production network over the last two decades.

When the yen’s internationalisation was actively debated up to the mid-1990s, it was expected that the use of the yen would be growing as a trade invoice currency if intra-firm trade increased between Japan and Asia. As shown in Figure 1, however, the share of yen invoicing has not shown a steady upward trend from 1983.

Evidence from interview analysis

There has been a growing literature on the determinants of currency invoicing. These studies typically use data on the share of currency invoicing at a country level (Goldberg and Tille 2008, Kamps 2006) whereas destination- or industry-level data is rarely used. One notable exception is Goldberg and Tille (2009), who use the highly detailed Canadian import data at the customs level with rich information on the source country, invoice currency, and value of transactions. Another exception is Friberg and Wilander (2008), who conduct questionnaire surveys of Swedish exporting firms for empirical tests on the determinants of currency invoicing.

In order to solve the puzzle we would need firm-level data on trade invoice currency – data that, alas, are not publicly available. We therefore employ a different approach – an interview survey of leading Japanese exporters (see Ito et al. 2010, which is based on our research at the Research Institute of Economy, Trade, and Industry).

We interviewed finance directors of 23 Japanese companies from four major export industries (automobile, electrical machinery, general machinery, and electronic components) between autumn 2007 and autumn 2008. A notable advantage of interview survey is that we obtain firm-level information on:

  • exchange-rate-risk management,
  • destination breakdown data on the choice of invoice currency, and
  • the underlying reason for a company’s invoicing choice through face-to-face discussion with each company.

Based on this firm-level information, we propose new determinants of invoice currency that have not yet been examined in the literature.

Table 1 lists the total number of respondent companies classified by the type of currency most frequently used when invoicing exports as well as the breakdown figures for each of the four industries. (For instance, [14/21] in the "Eurozone" column indicates that 14 of the 21 respondent companies answered that the "importer's currency" – the euro in this case – is the most frequently used currency in invoicing exports to Eurozone countries). Our interviews precisely capture the two puzzling tendencies of Japanese exporters' invoicing behaviour, showing that although exports to advanced countries with major currencies are generally invoiced in the importer's currency, Japanese companies use the dollar, a third country currency, more frequently than the yen in invoicing exports to Asian economies.

What impedes yen invoicing?

By constructing a data set based on firm-level financial data of sampled companies, we performed a probit analysis to examine the role of:

  • the type of export channel, for instance, whether specific export transactions are intra-firm trade, inter-firm trade, or trade via a trading company;
  • currency hedging costs;
  • the intensity of competition in the export destination markets and the degree of product differentiation; and
  • the structure of production and distribution network in which goods are produced in Asia and shipped to the US as the final destination.

Our novel findings – which provide answers to the two puzzles mentioned above – are three-fold.

  • First, importer currency invoicing is prevalent in Japanese exports to advanced countries because most of their exports are destined to local subsidiaries that face severe competition in local markets. In this case, Japanese parent firms have a strong tendency to take exchange-rate risk by invoicing in the importer’s currency, which is consistent with the pricing-to-market behaviour discussed in the literature. It also makes economic sense to concentrate currency risk at the headquarters, since it is better equipped with risk management expertise and with scale economies.
  • Second, Japanese firms that export highly differentiated products or have a dominant share in global markets tend to choose yen invoicing even in exports to advanced countries.
  • Third, although Japanese firms have shifted their production bases to Asian countries, exports from these Asian bases tend to be invoiced in dollars as long as the final destination market is the US. By invoicing exports to Asian subsidiaries in dollars, Japanese firms can shift currency risk from Asian subsidiaries to headquarters for efficient currency-risk management.
Policy implications

We find that the puzzle of Japanese exporters' invoicing behaviour is closely related to active overseas operations of Japanese manufacturing firms and, among others, growing intra-firm trade. Even though a large part of Japanese firms’ cross-border production network is built in Asia, the final destination market for Asian subsidiaries’ exports is often the US, where dollar invoicing is prevalent. From a medium- to long-term perspective, such foreign-exchange risks will have a non-negligible impact on the earnings and location choices of companies.

Yet as the final market for Japanese Asian subsidiaries shifts more towards Japan and Asia, the dollar fetters start to break. At that point, Asian countries including Japan may find it beneficial to stabilise the intra-regional exchange rates. It is likely that facilitating the local or regional currency invoicing to avoid exchange-rate risks would become an important issue in the region when this shift occurs. Asian countries will find it beneficial to remove regulations on foreign exchange and financial transactions. With such structural changes in trade a possibility, the invoicing currency of Japanese manufacturing firms may well change in the near future.

References

Friberg, Richard and Fredrik Wilander (2008), "The Currency Denomination of Exports – A Questionnaire Study", Journal of International Economics, 75:54-69.

Grassman, Sven (1973), “A Fundamental Symmetry in International Payments”, Journal of International Economics, 3:105-16.

Goldberg, Linda S and Cédric Tille (2008), “Vehicle Currency Use in International Trade”, Journal of International Economics, 76(2):177-192.

Goldberg, Linda S and Cédric Tille (2009), “Micro, Macro, and Strategic Forces in International Trade Invoicing”, NBER Working Paper 15470.

Ito, Takatoshi, Satoshi Koibuchi, Kiyotaka Sato and Junko Shimizu (2010), “Why Has the Yen Failed to Become a Dominant Invoice Currency in Asia? A Firm-Level Analysis of Japanese Exporters’ Invoicing Behaviour”, NBER Working Paper 16231.

Kamps, Annette (2006), “The Euro as Invoicing Currency in International Trade”, ECB Working Paper 665.

Page, SAB (1977), “Currency of Invoicing in Merchandise Trade”, National Institute Economic Review, 33:1241–1264

2,835 Reads