Economic growth, current account dynamics, and growth regimes in the Baltic states

ABSTRACT

This article considers the growth performance of the Baltic states from the mid-1990s to 2021. Economic growth was rapid before the global financial crisis but slowed markedly after the crisis. Panel data estimations using seemingly unrelated regressions suggest that the dynamics of the current account balance are important for short and medium-term growth in the Baltic states but that there is a break signifying a change of short-term growth regime around the time of the global financial crisis. Before the crisis, rapid growth was supported by domestic demand that was made possible by large current account deficits. After the crisis, economic growth was supported by external demand reflected in improvements in current account balances. The shift in the short-term economic growth regime after the global financial crisis has brought lower rates of economic growth but also reduced financial vulnerability.

KEYWORDS:

• Economic growth
• economic convergence
• current account balance
• global financial crisis

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Barro and Sala-i-Martin (2003) is a standard reference book that discusses theories of long-term economic growth and presents empirical estimates of growth.

2. Some of these models may be considered alternative and not part of mainstream economic thinking, while others are or have been part of mainstream policymaking, for instance, at the World Bank and the International Monetary Fund.

3. Latvia was arguably the hardest hit by the global financial crisis, and it was the only Baltic state to enter a borrowing program with the IMF, the World Bank, the EU, and neighboring countries. The program was agreed upon at the end of 2008 and made €7.5 billion available for Latvia.

4. The data were downloaded on 16 December 2022. The data from AMECO, the Annual Macroeconomic database of the European Commission, are from the November 2022 vintage used for the Autumn 2022 Forecast of the European Commission.

5. Income levels are generally lower in Europe than in the USA; the EU15 average is around 75% of the US level throughout the period considered.

6. Notice that the crisis period 2007–2012 covers not only the deep falls in GDP but also the subsequent rebounds in Estonia and Lithuania in 2011 and Latvia in 2012.

7. The convergence rate for a given period is calculated as the average annual growth rate that brings the GDP per capita PPP relative to the EU15 from its initial level to the level at the end of the period considered.

8. Hansen and Hansen (2004) is an early study considering whether large and growing current account deficits were sustainable.

9. Harkmann and Staehr (2021) use data from central and eastern Europe and show that the drivers of the current account balance in those countries depends on the exchange rate regime. External factors are of key importance for the current account dynamics in countries like the Baltic states that have fixed exchange rates or are in a currency union. In contrast, external factors are less important in countries with floating exchange rates.

10. It is noticeable that not only the Baltic states but also many other countries in southern Europe and central and eastern Europe experienced very large current account deficits, while there were large current account surpluses in many northern European countries. The current account deficits in the Baltic states were part of broader asymmetric developments in Europe before the global financial crisis (see Giavazzi and Spaventa 2011; Kang and Shambaugh 2016) for a broader discussion of current account imbalances in Europe.

11. Quarterly data would require detailed modeling of the dynamics and interaction of the two main variables of interest. This could be achieved in a structural vector autoregressive model, but such a modeling strategy is impeded by the volatility of the quarterly data in the Baltic states and is beyond the scope of this paper.

12. The first stage of the SUR methodology consists of OLS estimations with country-fixed effects, from which the residuals are recovered to compute the cross-equation covariance matrix. The second stage is then a standard GLS estimation where the computed covariance matrix is used as a weighting matrix.

Additional information

Notes on contributors

Karsten Staehr

Karsten Staehr is a professor of macroeconomics at the Department of Economics and Finance at Tallinn University of Technology, Estonia, and a part-time research advisor at the Bank of Estonia. He carries out research and policy analysis within macroeconomics, public finances, international economics, public economics, European integration, and comparative economics.