External adjustment with a common currency: the case of the euro area

The USA, for instance, had a negative IIP representing 39% of its GDP at the end of 2017.

In absolute terms Spain holds the second largest net external debtor position in the world amounting 941.507 billion euros (1.108.386 billion dollars). The USA is the country with the largest negative net external position totalling 7.725.002 billion dollars.

There may be second-order effects as the depreciation could affect the terms of trade with the countries outside the currency area.

See for example Atoyan et al. (2013), Kang and Shambaugh (2014), or Eggertsson et al. (2014).

The data availability of the IIP at the quarterly and yearly frequency is different for each country, with the final estimated quarterly samples spanning from 1980: IV to 2017: I for Germany, Italy and Spain. For France the sample begins on 1989:III.

For “other assets” I use the same market weights as those computed for short-term fixed income assets. For “other liabilities” I assume the total value is denominated in local currency using the same returns as those from short term fixed income liabilities.

The other asset classes (FDI, fixed income and other assets) do also show a large share of euro area positions after the introduction of the common currency.

The data sources are the NSEE France, ISTAT Italy, and the central banks of Germany and Spain.

See Evans and Fuertes (2011) and Fuertes (2019).

The analysis does not include the secondary income which has been historically low for the four countries.

In deriving equation (4) I have performed several first-order approximations. To assess the accuracy of those approximations, we can compute the error term from equation (3) which also includes any measurement errors from the original data. The error term is small and stationary for the four countries under analysis.

Dickey–Fuller augmented unit root tests are performed to the three variables introduced in the VAR. As it is expected, the tests for \(\Delta nx\) and \(r^{NFA}\) reject the null of unit root at the 1 % level for all samples. The tests for nxa cannot reject the null of unit root in the case of France and Germany. For Spain and Italy the null is rejected at the 10 % level. There are several reasons to believe that the nxa variables are not unit root processes though. First, nxa is by definition a linear combination of stationary variables given that \(\Delta nx\) and \(r^{NFA}\) are stationary. Second, the unit root tests lack of power when the alternative hypothesis is a very persistence process with high \(\rho \) as it is the case. Third, being nxa a non-stationary process implies that the non-ponzi game condition could be violated, meaning that Germany and France, with some probability, may not repay their external debt, something very unlikely over the last 40 years.

I carried out the procedure with a maximum number of breaks \(m=3\) and a trimming of 0.2, which means that the minimal length required is 50 observations.

The exchange rate mechanism established that currency fluctuations had to be contained within a margin of 2.25% on either side of the bilateral rates (with the exception of the Italian lira, the Spanish peseta, the Portuguese escudo and the pound sterling, which were allowed to fluctuate by \(\pm 6\)%). The UK did also abandon the exchange rate mechanism in 1992.

The trade weighted exchange rates are OECD real effective exchange rates. I calculated the financial weighted real exchange rates using the country portfolio weights that I used to calculate the portfolio returns for each of the different asset classes.

In the case of France in Table 8, due to the small sample available before 1999 I provide the results for the whole sample instead.

Recall that I obtain the exchange rate components by including in the VAR estimation the part of the return differentials and net exports growth contemporaneously related with the trade weighted and financial weighted real exchange rates.

The quick external adjustment expected for Italy and Spain is supported by projections from the International Monetary Fund released in the April 2019 World Economic Outlook. These projections establish that for a group of euro are debtor countries, including Italy and Spain, the net international investment position is expected to improve by more than 25 percentages points of their collective GDP over the period 2017–2024.

Habib (2010) analyses the differential returns between gross foreign assets and liabilities for a sample of 49 countries, including France, Germany, Italy and Spain, using yearly implied returns.