“The impact of remittances on household savings in the Baltics”

Proper understanding and monitoring of household savings are crucial to effec- tive macroeconomic policies targeted at balanced and sustainable economic growth. Remittances, as a financial flow of foreign capital, can create a vital part of private sav- ings. This paper is aimed at identifying whether remittances contribute to household savings in the Baltics along with other macroeconomic variables in a post-crisis period, during which the relative smoothing and convergence of economic development of the Baltic countries after the sharp financial distress in 2009 can be observed. The following methods of panel data regression analysis were employed: fixed effects and OLS. The results of the econometric analysis based on both fixed effects and OLS meth- ods reveal that remittances are an essential driver of savings in the Baltics in the long run. Savings in the Baltics are not significantly influenced in the short term by sharp economic fluctuations, but are dependent on demographic factors and foreign capital, which can bring instability in economic development and financial flows of the region.


INTRODUCTION
Remittances account for a significant part of foreign capital flows in developing countries. Remittances are considered to have a positive effect on the economy and be beneficial for development (e.g. Aguinas, 2006, Ghosh, 2006. However, it is not clear enough how remittances affect different aspects of wealth of a recipient economy. The study of the Asian Development Bank and the World Bank (2018) has documented that remittances are acyclical regarding the recipient state. Remittances appear to be less volatile than other types of financial flows, including both foreign direct investment (FDI), and official development assistance (ODA) and export (Gavurova et al., 2020). However, they are more procyclical than ODA and less procyclical than financial flows. Remittances demonstrate flexibility during sudden financial crises and breaks and remain steady, whereas total financial flows drop abruptly during such episodes, thus, offsetting the effects of volatile financial flows. Households that receive international remittances spend more on health, education and housing than those that do not. Remittances also contribute to smoothing out consumptions of receiving households and facilitating them to mitigate sudden economic disaster better.
The main issue of remittances regarding development is that the majority of remittances are spent on consumption, and minimal share is invested (Basu & Rajan, 2018). It should be noted that recently the spending on education and health of remittance-receiving households is considered as investment in human capital as in the long run they contribute to the capacity development of labor force in the country.

The effect of remittances on household savings
The empirical literature in most cases indicates a positive association between household savings and remittances across countries ( has studied the impact that remittances have on investment, household savings and economic growth. It was found out that remittances had a positive influence on growth through stimulating family consumption that may rise imports, which cuts savings under the low levels of income.

Remittances in the Baltics
There is vast literature that explores the impact of remittances on economic growth indicating their relatively positive influence on the economy of receiving countries and the development of the countries of different income levels ( Azam (2016) has studied the impact of remittances on poverty alleviation in a global perspective, including Baltic countries, and found out there are substantial possible benefits related to foreign remittances for poor people. The regional research of Zuniga (2011) has shown that remittances play a vital role in the economy of countries, including the Baltics. Independent of the role of institutions, remittances positively influence economic growth; however, they are used for consumption rather than savings. The results of Kumar and Stauvermann (2014) have demonstrated that in Lithuania, the remittances have a dynamic short-run and a significant long-run effect on the output per worker. Moreover, the causality analysis has shown a mutually underpinning effect between the output per worker and capital per worker, as well as unidirectional causality from remittances to output per worker.
Thus, remittances have an undoubted effect on the economy. However, there is no consensus about the nature of their influence on the savings rate, hence giving grounds for further research.

Economic factors
Unemployment rate and inflation are the most widely used indicators for assessing the influence of uncertainty on household savings. When anticipating economic distress, households tend to save more as a mitigating measure. An increase in the unemployment rate and inflation increases the volume of household savings (Kostakis, 2013; Syed et al., 2017; Temel Nalın, 2013a). Temel Nalın (2013b) investigated the drivers of household saving and portfolio choice behavior in Turkey using the inflation index and concluded that inflation increased the probability of household saving. Niculescu-Aron and Mihăescu (2012) employed first-order differences in the inflation rate in two panels of the states -Western Europe and Central and Eastern Europe. They found out in both panels that a rise in inflation was positively associated with household savings.
Fiscal policies may have a considerable effect on household savings. Active fiscal policies aimed at stimulating consumption can adversely affect the savings. To stimulate the economy, the government can increase government expenditure and run a budget deficit, which may decrease present household consumption expenditure if households anticipate future tax rises to support public debt (OECD, 2018). An assumed consumption smoothening, a drop in the expected future in-come will decrease current consumption and rise current household savings. The fiscal policy of the states is assessed by the rate of the government surplus in % of GDP, rate of public debt in % of GDP, social protection expenditures by the government in % of GDP and the portion of indirect taxation in total taxation.
The net inflow of FDI is negatively associated with household saving. Hence, FDI seems to be a substitute for domestic saving (Rocher et al., 2015).

Demographic factors
The concept of consumption smoothing over a lifetime suggests that young people tend to save less than adults. Elderly individuals are likely to dissave. Thus, countries with high age dependency are likely to have lower household savings (Edwards, 1995;Loayza et al., 2000).
An increasing life expectancy may also affect households savings. Demographic projections show that the population is rapidly ageing in many of the EU member states. Therefore, households may rise their savings as they face a longer retirement. Thus, life longevity can also be a potential driver to capture this effect (Bloom et al., 2002;Carvalho et al., 2017). Often, older people live alone, which drops their disposable income. Therefore, assuming that women are over-represented among the older age group of the population, their saving rates may be negatively related to life expectancy (OECD, 2017).

The effect of financial literacy on household saving behavior
Financial literacy is another critical factor that influences households saving behavior. Shortage of primary financial education may be related to the lack of sufficient retirement planning and lack of wealth accumulation (Reisch & Zhao, 2017

Model specification
Based on the theoretical grounds and empirical findings, a regression model was used to assess the impact of remittances on household saving behavior in the Baltics along with other macroeconomic determinants (Table 1)  The transition to a market economy after recapturing independence shadowed a similar pattern in the Baltics. All three countries followed a common goal to join the EU and the monetary union.  According to statistics, the Baltic countries also demonstrate similar household savings dynamics ( Figure 3).
It should be noted that households saving rates need to be interpreted carefully due to data soundness and limited international comparability. Saving rates seem to be in divergence with other economic variables. In particular, household saving rates have been negative over the last twenty years in Romania and Bulgaria, as shown in Figure  3. It means that households in these EU member states spend considerably and interminably more than they earn. However, economic reality does not provide evidence to this, since households debt-to-income ratios stay relatively low in these states and have even been declining since 2009.
Given the typical path of economic development and saving dynamics, the Baltics represent a stable research panel. The next section studies the influence of remittances on household savings in the Baltics.

RESULTS AND DISCUSSION
The regression analysis has been conducted by employing panel data with the use of fixed effects and Source: Eurostat.  OLS methods. Summary statistics are represented in Table 2. Regression analysis results are shown in Table 3. The collinearity test is presented in Table  4, and tests for differing group intercepts, heteroscedasticity and error distribution are shown in Table 5.    Fixed effects estimation demonstrated that the uncertainty indicators -unemployment and inflation -were not significant determinants of household savings in the Baltics. However, an OLS estimation has revealed a positive association between unemployment and savings, which confirms the assumption. Government debt has appeared to be an insignificant regressor in the fixed-effects estimation. However, OLS estimation has revealed a negative link between this variable and household savings, which supports the assumption.
The FDI inflow in fixed effects estimation appeared to be negatively linked to savings, which is following the assumptions. However, in the OLS estimation, this variable was not significant.

CONCLUSION
This paper identified the influence of remittances on household savings in the Baltics, along with other macroeconomic variables in the post-crisis period. The relative smoothing and convergence of economic development of the Baltic countries can be observed after the sharp economic distress in 2009. The significance of remittances for household savings in the Baltics was studied along with other macroeconomic variables: unemployment rate, inflation, FDI inflows, and government debt and demographic variables: life expectancy, old-age dependency, and the introduction of the National Strategy of financial education as a proxy for financial literacy.
Fixed effects and OLS methods were employed for the regression analysis. Both estimations have revealed that remittances are an essential driver of savings in the Baltics. The estimation with fixed effects association between demographic indicators did not support the hypothesis advanced in this paper. However, the issue lies in the complexity of savings due to high subjectivity.
According to the fixed effects estimation, uncertainty indicators -inflation and unemployment -appeared to be insignificant factors of saving. Among other employed factors, only FDIs are significant. However, OLS estimation results have shown that government debt and unemployment do influence savings.
The fixed effects results indicate that in general savings in the Baltics are not significantly influenced in the short term by sharp economic fluctuations, but are dependent on demographic factors and foreign capital, which can bring instability in economic development and financial flows of the region. Thus, in the light of this, remittances, given their nature described above, are a vital source of savings in the long run.