Abstract
This Selected Issues paper analyzes the macroeconomic impact of workers’ remittances on Moldova. The paper focuses on Moldova’s labor emigration since the late-1990s using survey data designed to shed light on the economic and social consequences of migration. The survey results are broadly consistent both with the findings from balance-of-payments data and with the stylized facts in the labor migration literature. The paper also examines various indicators to assess the appropriateness of the current exchange rate level.
I. Macroeconomic Consequences of Workers’ Remittances in Moldova1
A. Introduction
8. Inflows of workers’ remittances have become increasingly important to Moldova. The large-scale emigration process characterizing Moldova today took off in the wake of the 1998 regional crisis, and since then both emigration and remittances have grown steadily. The domestic economic situation was already difficult before the crisis, as the transformation from a centrally-planned economy to a market-based economy implied drastic structural changes, resulting in output contraction and massive job losses. When hit by the 1998 crisis, many workers did not have any other viable alternative than to seek job opportunities abroad to support their families. This emigration trend has intensified, and in 2003 the total officially-estimated gross inflows of workers’ remittances reached almost 25 percent of GDP, up from 5 percent of GDP in 1996 (Figure 1). This is very large compared to most other countries, including in the Commonwealth of Independent States (CIS) and in Central and Eastern Europe (CEE), where remittances significantly exceed 5 percent of GDP only in a few countries.
Moldova: Gross Workers’ Remittances
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: National Bank of Moldova.9. The purpose of this chapter is to assess the main characteristics of remittances, their macroeconomic consequences, and policy implications. The large inflows have a profound impact on the Moldovan economy. They (i) drive growth through consumption; (ii) reduce labor supply and put pressure on wages; (iii) finance a large and widening trade deficit; (iv) put the exchange rate under appreciation pressure; (v) fuel inflationary expectations; (vi) contribute to higher tax revenues; and (vii) threaten the sustainability of the pension system. The analysis focuses on macroeconomic issues, but suggests that many macroeconomic problems reflect structural bottlenecks in the economy, requiring structural policy solutions.
10. The chapter is organized as follows. The next section summarizes the main stylized facts on remittances, as identified in the current literature. Section C outlines the main patterns of migration and remittances in Moldova—discussed in detail in Chapter II. Section D reviews the most important macroeconomic effects of remittances, and Section E discusses their policy implication. Section F concludes.
B. Stylized Facts on Remittances
11. Remittances have become increasingly important as a source of external financing worldwide. Ratha (2003) reports that they are the second most relevant external financing source in developing countries, after FDI, and are more critical than foreign aid. They are mainly used for consumption and, to a lesser extent, for saving and investment. Investment, moreover, tends to be “unproductive”, and often concentrated in real estate. Although remittances finance investment in education, their net effect on human capital development could be negative, since emigration can potentially lead to brain drain.
12. Remittances affect economic agents and the economy in many ways. In particular, the evidence regarding their contribution to long-term growth is disappointing. When analyzing panel data across 113 countries, Chami et.al. (2003) find that remittances are negatively correlated with economic growth, mainly because they are generally not used to finance productive investment. The authors suggest that one important explanation for this result is moral hazard, since remitters cannot directly observe the behavior of recipients. Short-term effects on growth are more positive. Part of remittances-induced consumption spending benefit domestic producers, thereby creating a Keynesian-type multiplier effect. Indeed, several studies report positive multiplier effects of remittances on short-term growth. Remittances also have a positive impact on poverty reduction, especially because they are transferred directly to households, instead of channeled through the government—as is the case of foreign aid. Moreover, remittances can also contribute to macroeconomic stability: since they tend to be relatively stable and countercyclical, it is possible to avoid the Dutch-disease-type risks associated with large windfall gains from positive (and quickly reversible) terms of trade shocks. Finally, remittances are not debt-creating flows.
13. There are other potential negative effects generally associated with emigration and remittances. While emigration reduces labor supply, the related remittances increase demand for goods and services, putting pressure on wages and on the real exchange rate and increasing costs for domestic producers. This discourages investments in sectors that do not benefit from remittances (including the export sector). Moreover, if the inflows of remittances were to dry up quickly, the economy would become vulnerable, much in the same way as an oil exporting country suffers when oil prices plummet.
C. Patterns of Remittances and Migration in Moldova
Definition and measurement
14. Workers’ remittances are defined as the sum of two components in the balance of payments: (i) compensation of employees in the income account; and (ii) workers’ remittances in the transfer account. Compensations capture workers’ wages, salaries, and other benefits earned by non-residents (temporary workers).2 Remittances are transfers from migrants residing abroad (permanent migrants, or workers staying or expected to stay for a year or more). A third balance of payments component—migrants’ transfers in the capital account—could be included, but these transfers should be formally contra-entries to flows of goods and changes in financial items that arise from migration and, as such, they do not capture money sent to Moldova from workers abroad. This type of transfer is, moreover, insignificant in the case of Moldova.3
15. Workers’ remittances are inherently difficult to measure, as they involve transactions between and within households, often outside the formal economy.4 This chapter focuses on officially-recorded remittances, as reported in the balance of payments, since they are the only available source for a time series on workers’ remittances. Furthermore, these data, compiled by the National Bank of Moldova (NBM), appear to be of reasonably good quality, considering the difficulties in collection. In addition to banking statistics, the NBM estimates workers’ remittances outside the banking system using information on individuals’ exchange of foreign currency into lei at foreign exchange bureaus and banks’ foreign exchange counters, and on transactions where foreign exchange is typically used (e.g., homes and cars). Many, if not most, other countries do not make such efforts. That said, it is important to stress that statistics on workers’ remittances are incomplete and must be interpreted with great caution.
Levels, trends, and distribution of workers’ remittances
16. On a net basis, remittances inflows have become the single most important source of foreign exchange in Moldova. Remittances are sizable even if we consider only net inflows (almost 23 percent of GDP in 2003), or only inflows coming through the banking system (almost 15 percent of GDP in 2003—Figure 2). Remittances have been growing rapidly also when compared to exports of goods and services; this ratio corresponded to 15 percent in 1998, but edged up to more than 45 percent in 2003.
Moldova: Workers’ Remittances through the Banking System
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: National Bank of Moldova.17. The strong upward trend in remittances may have been somewhat weaker than indicated by the data. Trust in banks has improved, and confidence in the leu has strengthened over time as the Moldovan economy stabilized after the 1998 regional crisis. In addition, wire transfers from abroad have become more accessible and cheaper. This could exaggerate the upward trend in two ways. First, stronger banks and cheaper transfers raise incentives to use official channels and, as a result, balance of payments statistics now capture a larger share of all workers’ remittances. Second, a stronger leu and stronger banks encourage people to exchange their idle foreign exchange cash holdings (“money in the mattress”) into lei, and deposit the money in banks, increasing estimates of inflows outside the banking system. It should be noted, however, that some of those cash holdings may emanate from unrecorded remittances in earlier years, which only surfaced when confidence in the leu strengthened. Therefore, some unrecorded remittances may have been captured by official statistics in a later year.
18. The main source of remittances are temporary workers. About 70 percent of all remittances originate from Moldovans working abroad only part of the year. Many of these are likely to be seasonal workers in, for example, agriculture and construction (Figure 3).
Moldova: Distribution of Temporary and Long-term Workers’ Remittances
(In percent)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: National Bank of Moldova.19. Available balance of payments data do not reflect trends in the geographical origin of remittances, since they cover only inflows coming through the banking system. For instance, they do not capture information about Moldovans working in Russia prior to 2000, when money transfers started to be used. Before 2000, non-residents in Russia were not allowed to make any money transfers abroad; since then the maximum amount allowed has been gradually increased. At the same time, money transfers have become much more available in Russia, as several money transfer systems have been introduced. Money transfers have also increased owing to security reasons. Reportedly, it is dangerous for migrants to bring cash to Moldova across the Russian and Ukrainian borders.
20. Most remittances transferred through the banking system originate from the European Union (EU). In 2003 almost 60 percent of these inflows came from the EU, and about a quarter from the CIS (Figure 4). Of EU transfers, 47½ percent are from Italy, and 17 percent from Portugal. Almost all transfers from the CIS originate from Russia (97 percent).
Workers’ Remittances to Moldova by Region of Origin
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: National Bank of Moldova; and Fund staff estimates.Comparison with other countries
21. The amount of workers’ remittances sent to Moldova is large both from a regional and an international perspective. As illustrated in Figure 5, Moldova clearly stands out compared to all CIS and CEE countries, and also fares highly compared to the rest of the world.5 Ratha (2003) ranks Moldova as one of the top ten receivers of workers’ remittances in terms of GDP.
Gross Workers’ Remittances as a Share of GDP in 2003
(In percent)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: IMF; and Fund staff estimates.D. Macroeconomic Consequences of Migration and Remittances
Economic growth
22. Workers’ remittances have played an instrumental role in propelling growth in recent years, through their effect on consumption. As shown in Figure 6, real GDP growth has been driven by consumption, in turn fueled by remittances. Figure 7 shows a strong correlation between real consumption growth and remittances growth during the period 1999-2003, with a correlation coefficient of 0.98. This high correlation is a direct reflection of the fact that the large inflows of remittances boost income. While GDP has increased relatively fast since 2000, Gross National Income (GNI) and Gross National Disposable Income (GNDI) have grown even faster (Figure 8).6 Between 2000 and 2003, GDP per capita grew by 13½ percent on average, GNI per capita by 15½ percent, and GNDI per capita by 18 percent, mainly reflecting large and rising workers’ remittances.
Moldova: Contribution of Consumption and Fixed Capital Formation to Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: Moldovan authorities; and Fund staff estimates.Moldova: Workers’ Remittances and Real Consumption Growth
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: Moldovan authorities and Fund staff estimates.Moldova: GDP, GNI, and GNDI per Capita
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: Moldovan Authorities; and Fund staff estimates.23. By contrast, investment contribution to economic recovery has been very modest. Fixed capital formation is still quite low as a share of GDP and, compared to GNI and GNDI, has remained virtually flat since 2000 (Figure 9). Investment has not yet recovered from the 1998 recession: while at that time fixed capital formation accounted for 22 percent of GDP, this share fell to 17 percent of GDP in 2003 and is projected to be about 18 percent in 2004. With disposable income boosted by large inflows of remittances, investment could have been growing more rapidly as a share of GDP had the investment climate been more favorable. Without higher investment, Moldova’s long-term growth prospects are grim.
Moldova: Fixed Capital Formation
(In percent)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: Moldovan authorities; and Fund staff estimates.24. The current situation, with very high remittances and somewhat disappointing levels of investment, suggests that Moldova could be saddled on a path where migration and remittances reinforce each other. The cost for Moldovan workers to work abroad is comparatively low for several reasons: (i) Moldova is located close to both the EU and the CIS; (ii) there are tight connections between Moldova and other CIS countries; and (iii) since many Moldovans already are working abroad, they can facilitate emigration of others.7 The current trend of persistent emigration leading to rising remittances is, therefore, to be expected under the circumstances. With a favorable investment climate, remittances could be used to finance investment, creating jobs, and making it more attractive for Moldovans to stay at home. In a less favorable environment, however, the potential return on investment may not be high enough to compensate for risky projects, making it more attractive for recipients to use remittances to finance the emigration of one or more family members.8 At this point, Moldova appears to be in a situation where emigration is a better option than investing at home.
25. A potential problem is moral hazard—remittances may reduce incentives for reforms aimed at addressing the very problems that motivated emigration in the first place. Workers emigrate because economic circumstances are unfavorable in Moldova, and transfer money back home because they need to support their families. These remittances lift household income, alleviate poverty, and boost GDP growth through higher consumption. With such favorable developments, the pressure on the government to implement structural reforms—which would make it more desirable to work and invest at home—is reduced.
Labor market
26. Emigration and remittances have significantly affected the labor market, raising both the equilibrium and the reservation wage. As argued above, emigration costs are relatively low, and, barring introduction of restrictive legislation in host countries, it may become easier to emigrate as emigration increases. From the perspective of a Moldovan worker, the possibility of working abroad fundamentally alters work prospects, raising the equilibrium wage rate in Moldova. Moreover, remittances seem to have also increased the reservation wage, with recipients demanding a higher wage at home.
27. Emigration has reduced unemployment and led to labor shortages. According to the official statistics, labor supply has decreased markedly owing to emigration, while real and dollar wages have increased very rapidly. Statistics on the number of emigrants show that, by the fourth quarter of 2003, 322,000 people had left Moldova to find a job abroad (Figure 10). By the third quarter of 2004, this number had expanded to 367,000, which corresponds to about ¼ of the economic active population. These numbers should be seen as floors, or low estimates, since existing data only cover workers who officially declared that they were emigrating. At the same time employment, unemployment, and the labor force have all declined substantially (Figure 11).
Moldova: Labor Migration 1/
(In thousands)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: Moldovan authorities.1/ The number of people who have declared to have migrated to find a job.Moldova: Labor Market Indicators 1/
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: Moldovan authorities.1/ ILO methodology.28. The reduced labor supply has pushed up wages. Starting from 1999 the average real wage increased by 70 percent through 2003, and by almost 80 percent through the first ten months of 2004. Average monthly dollar-wages rose by 126 percent through 2003, and by almost 200 percent through the first ten months of 2004. By contrast, real GDP is estimated to have grown by 33 percent from 1999 through 2004, and nominal GNDI (expressed in dollars) is estimated to have grown by 154 percent during the same period. There is also anecdotal evidence of labor shortages in some sectors (construction, transportation, and agriculture), where employers are not able to find enough workers unless they are prepared to pay very high wages by Moldovan standards.9
29. Another potential implication of emigration is brain drain although this in practice may not have been in a problem in Moldova (see Chapter II). Many of those with higher education and specialized working skills may find it financially very attractive to leave Moldova. The resulting decline in human capital is likely to hamper economic growth. In this context, it is important to note that investing in a higher education may fetch a very high rate of return. Thus, recipients of remittances in Moldova may conclude that the best way to use the money is to invest in own education or children’s education, with a view to find a job abroad.10
Balance of payments and the exchange rate
30. Remittances have also had a major impact on the balance of payments. Exports of goods and services have been growing quite rapidly since 2000, but as imports have grown even faster, the balance of trade in goods and services has deteriorated from about 15 percent of GDP in 1999 to over 30 percent in 2003 (and was most likely over 30 percent in 2004 as well). Remittances covered about ¾ of the trade deficit in 2002 and 2003 (in 2004 the coverage may be even higher). As actual inflows could be larger, the coverage may have been even higher, implying that Moldova’s current account deficits were significantly lower than reported.
31. The increasing importance of remittances in financing trade deficit stands in sharp contrast to the disappointing performance of foreign direct investment (FDI). While workers’ remittances have been growing rapidly since 2000, FDI has lagged behind, covering only 10 percent of the trade deficit in 2003. This is yet another indication that the investment climate in Moldova is not conducive to private sector activity. As a result, remittances have become much more important than FDI as a source for foreign exchange inflows (Figure 12).
Moldova: Workers’ Remittances and Foreign Direct Investment
(In millions of U.S. Dollars)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: National Bank of Moldova.32. Remittances inflows have created a substantial appreciation pressure on the exchange rate. Since the second quarter of 2003 the leu has been under a fairly constant appreciation pressure, in particular against the dollar, but also against other currencies. As a result, the leu has strengthened vis-à-vis the dollar as well as against a trade-weighted currency basket, the Nominal Effective Exchange Rate (NEER) (Figure 13). At the same time, the NBM has stepped up its purchases of foreign exchange, mainly dollars, to dampen the actual appreciation of the leu. Without those interventions, leu appreciation would have been even stronger. Workers’ remittances are the only large foreign exchange net-inflows that can explain this strong appreciation pressure.
Moldova: The Moldovan Leu Exchange Rate
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: National Bank of Moldova; and Fund staff estimates.33. While at first sight, the external situation suggests that the leu should be under a depreciation pressure, there are important unexplained factors in the balance of payments supporting the appreciation trend. The recorded current account has deteriorated significantly since 2002, and the government is running arrears on some of its external debt obligations, with an accompanying low credit rating. On the other hand, this period has seen a large accumulation of reserves. It is not surprising that, at the same time, large errors and omissions were recorded in the balance of payments. Since the second quarter of 2003, errors and omissions have been quite large and positive; in the last three quarters of 2003, they amounted to $133 million or almost 7 percent of annual GDP, and through the first three quarters of 2004 they reached $78 million or about 3 percent of annual GDP. It is reasonable to guess that remittances explain a significant share of these unexplained inflows. If all errors and omissions were unrecorded remittances, the current account deficit in 2003 would have been $55 million, or less than 3 percent of GDP (compared to almost 7 percent officially reported in the balance of payments). Part of the errors and omissions could also be due to under-invoicing of exports and over-invoicing of imports to avoid the existing repatriation requirement, which would also imply that the current account is much stronger than reported. However, the trade deficit would still be very large.
34. The leu appreciation has also encouraged higher demand for domestic currency, making the appreciation self-reinforcing. With a stronger leu, households may be inclined to exchange their dollar-savings (e.g. from remittances) into lei. Similarly, the leu has become more attractive as a transaction currency. Consequently, there has been increasing demand for lei at foreign exchange bureaus; the appreciation pressure has been stronger there than in the interbank market, prompting the NBM to intervene in the cash market in an unprecedented fashion in 2004. There is also anecdotal evidence suggesting that the leu has become more desirable as a transaction currency in 2004.11
Monetary policy and inflation
35. Although its official monetary policy objective is to maintain the stability of the currency, in practice the NBM attempts to achieve several, sometimes conflicting, goals:
Keep inflation low;
Preserve external competitiveness by preventing or at least limiting excessive nominal appreciation of the leu against the dollar;
Accumulate foreign exchange reserves on a precautionary basis, to be able to meet external debt service and to reach a level corresponding to three months of imports; and
Keep interest rates low, to support private sector development and limit government domestic interest payments.
The NBM is also concerned about its profit level, since it affects the strength of its balance sheet and the amounts that can be transferred to the budget.
36. The large inflows of workers’ remittances has complicated the task of monetary policy. It has naturally proven to be very difficult to keep inflation down, while at the same time achieving the other monetary policy goals in the face of strong inflows of foreign exchange. The NBM can choose to intervene in order to prevent the leu from appreciating by buying large amounts of foreign exchange in the open market, with the added benefit that foreign reserves are built up. However, this boosts money supply and fuels inflationary pressures. Sterilization operations can in principle help, with the undesired consequence of pushing interest rates up. Short-term interest rates (maturities less or equal to a year) are directly affected by NBM interventions in the money market, and result in higher interest payments for the government. Commercial credits also become more expensive, as a large share of them are short-term, thus discouraging private sector activity. Long-term interest rates could in principle fall, as higher short-term interest rates help lower inflation expectations, but no such effect has been observed in Moldova. Moreover, sterilization operations are costly for the NBM, reducing its profits and its transfers to the budget.
37. Since late 2004 the NBM has given less priority to inflation. Until then, the NBM had been trying to strike a balance between keeping inflation down and preventing appreciation, while at the same time building reserves and keeping interest rates down. However, as the inflows of foreign exchange have persisted and the pressure on the exchange rate has intensified, the NBM appears to have given highest priority to preserving competitiveness in recent months. The appreciation of the leu against the dollar was halted in mid-2004 through historically large purchases of foreign exchange; the leu even depreciated somewhat before stabilizing toward the end of the year. Sterilization efforts have been stepped up, but not enough to prevent money growth from accelerating (Figure 14). Inflation has not yet picked up dramatically, but the inflation goal for 2004 was missed and there is clearly a risk of higher inflation in 2004 (Figure 14).
Moldova: Money Growth and Inflation
(Year-on-year, percent)
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Source: National Bank of Moldova.Monetary policy and competitiveness
38. Preserving external competitiveness by preventing excessive exchange rate appreciation has received increasing importance as a monetary policy goal. Under these circumstances, two questions are relevant: (i) does Moldova have a competitiveness problem?; and (ii) what can and should monetary policy do about competitiveness?
(i) While at first glance, competitiveness does not seem to be a major problem in Moldova, the export sector is clearly below its potential.12 The large trade deficit is not, in itself, evidence that that the exchange rate is misaligned. Exports of goods and services are growing at a healthy pace. Import growth is much higher, but imports are fully financed by workers’ remittances, suggesting that Moldova has a comparative advantage in exporting labor. Admittedly, the current account deficit is relatively large, but if the sizeable errors and omissions in the balance of payments were taken into account, the deficit would be much smaller, and declining. On the other hand, it is possible to argue that Moldova is not competitive in a more fundamental way. Considering that the economic environment is not conducive to private investment, and that work opportunities are scarce, pushing workers abroad, the export sector cannot fully develop its potential. In a more favorable business environment, small and medium-sized enterprises could thrive, leading to higher investment and increase private sector activity. Moldovan labor, skilled and unskilled, would then find better opportunities at home and exports of goods and services could replace exports of labor.
(ii) There are several reasons why monetary policy cannot do much to improve competitiveness in Moldova. First, and most importantly, monetary policy only affects competitiveness in the short-term, and the end-result of attempts to prevent a nominal appreciation of the leu may be nothing but higher inflation and an unchanged real exchange rate down the line. Higher inflationary expectations imply that the effect of monetary policy on competitiveness is likely to be short-lived, and an undesirable wage-price spiral may develop. Second, considering that the exchange rate is likely to be undervalued (see Chapter III) risk of inflation may be a more important concern at this point. Third, Moldova as chosen to adopt a flexible exchange rate regime, and the NBM has clearly stated its intention to keep inflation down. It is important to live up to this commitment in order to build up credibility, and let the market, by and large, set the exchange rate. There may be room for interventions to smooth short-term fluctuations, but this should not be at the expense of accelerating inflation and damaging credibility.
39. For the same reasons, there is a limit to what monetary policy can do to make Moldova a more attractive place to invest and work. While a depreciation would reduce the cost of labor, it is not likely to make Moldova more attractive, since there is not much slack in the labor market, and lower dollar-wages could not be sustained. Nominal lei-wages would have to increase and the end-result would only be higher inflation. What is more, to the extent that dollar-wages are lowered for a period, it will then become even more attractive to work abroad, which may encourage additional emigration. What is needed is higher productivity in Moldova, and higher real return on investment. That can only be achieved through structural policies to improve the business environment.
Fiscal impact
40. The exodus of Moldovan labor force and the related inflows of remittances have shaped fiscal performance in recent years, and will likely impact on fiscal policy in the medium-to-long term. In the short run, fiscal performance has been affected through the impact of emigration and remittances on (i) the labor market; (ii) revenue collection; and (iii) the exchange rate. In the long run, emigration raises fiscal sustainability issues, through its impact on the demographic dependency ratio and on the contribution base.
(i) Labor market changes: As emigration has reduced unemployment, and remittances have provided a social safety net to the poor, pressures on the budget related to unemployment have been alleviated (Figure 15).
Moldova: Unemployment, Unemployment Benefits, and Dependency Ratio
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: Moldovan authorities; and Fund staff estimates.(ii) Revenue collection: The revenue base has grown dependent on remittances. Overall collection has improved significantly over the last few years, largely owing to rising import taxes. Although the decline in labor force and employment has constrained growth in value added, and hence income subject to taxation, rising remittances inflows have boosted imports and import-related tax collection (Figure 16). This greater reliance on indirect taxation reflects a faster growth in domestic absorption relative to incomes generated in the domestic economy (or faster growth of national disposable income relative to GDP).13
Moldova: Tax Revenue Composition, 1999–2004
Citation: IMF Staff Country Reports 2005, 054; 10.5089/9781451825046.002.A001
Sources: Moldovan authorities; and Fund staff calculations.(iii) Exchange rate changes: The appreciation of the leu resulting from remittances inflows affects the government’s fiscal position in the short run. To illustrate, a 10 percent appreciation of the leu in nominal effective terms would lead to a deterioration in the government fiscal position of about 0.6 percent of GDP.14 This is a combination of a (negative) revenue impact of 0.9 percent of GDP, and a reduction in annual external debt service of 0.3 percent of GDP (assuming that all government spending is on domestically-produced goods). Allowing for a partial demand response (higher demand for imports, assuming price elasticity of 0.5), the net effect would be smaller—about 0.2 percent of GDP. In the long run, the initial net negative effect of an appreciation on the fiscal accounts would likely dissipate, particularly if the share of government spending on imports was relatively large.
41. In the long-run, emigration is likely to affect the balance between the taxable base and the demand for social spending in the domestic economy. The shrinking contribution base resulting from emigration adds to the challenge of safeguarding the viability of Moldova’s pension system in the context of an ageing population. The challenge is compounded by uncertainty about the ultimate size of the old-age population expected to depend on the Social Fund pension benefits, which will be determined by emigrants’ decision whether or not to retire in Moldova. While the reform of the pension system initiated in 1999 was an important step toward restoring the system’s short-term stability, securing its long-term fiscal sustainability is likely to necessitate adjustments in the future (see Box 1).
Social Insurance and Fiscal Sustainability—Conceptual Framework
Background
Moldova’s social insurance is administered through the Social Insurance Fund. The pay-as-you-go (PAYG) pension system is its largest component. The 1998 regional financial crisis impacted heavily on Moldova’s costly and poorly targeted social protection system, which was already under severe strain. Problems emerged particularly in the pension system, including rising contribution arrears, delays in payments, lack of funds, and growth in in-kind payments.
A pension system reform, launched in 1999, laid the foundation for transforming the system into a sustainable insurance program. The main features of the new system include: a new benefit formula, which bases future pensions more on individual contributions than on reported wages and years of service; elimination of most early retirement privileges; a gradual increase in retirement age; and an increase in the minimum required contribution record.
While the reform helped restore the short-term financial stability of the system, ensuring its fiscal sustainability remains a challenge. PAYG systems need to ensure an ongoing balance between contributions and benefits. Demographic changes, including an aging population, as well as increases in life expectancy, require changes to contribution rate or to the system benefits. In the case of Moldova, by raising the demographic dependency ratio, emigration is expected to place additional strain on the system’s fiscal sustainability.
Framework
In a PAYG system, current period benefits are financed from current revenues, typically via a payroll tax. For revenues to equal expenditure, the payroll tax rate (α) has to equal the pension bill divided by the wage bill, or the ratio of pensioners (M) to active contributors (N) times the replacement rate (β), where the replacement rate is defined as the ratio of the average pension to the average wage. Thus we can write
Allowing the possibility of budget transfers, (1) becomes
where τ is the ratio of budgetary transfers to pension expenditures (τ >0 implies transfers from the budget to the pension fund; τ <0 implies a transfer of the pension fund surplus to the budget). More generally, the ratio M/N can be expressed as a function of the demographic dependency ratio (M*/N*), where M* is the number of people 60 years and older and N* is the number of people between 15 and 59 years of age:
where γ is termed as the pension system coverage ratio, with its value depending on the maturity of the system, retirement policies, labor force participation rates and other labor market conditions.
Source: Castello-Branco (1998).42. To illustrate the possible adjustments in the pension system as a response to demographic changes, two alternative scenarios are presented in Table 1 below. The scenarios are conceived as a change relative to the situation projected for 2005. Under Scenario A, the number of contributors would decline by 10 percent and the number of beneficiaries would rise by 10 percent. Under Scenario B, the number of contributors would also decline by 10 percent (again, relative to 2005), but the number of beneficiaries would increase by 50 percent, simulating the return of a large number of future retirees to Moldova. As outlined in Box 1, under a PAYG system, an adverse demographic development—such as an increase in the ratio of the number of beneficiaries to contributors (M/N)—necessitates an adjustment in policy parameters—either the contribution rate (α), the replacement rate (β), or the fiscal transfer rate (τ)—to preserve the long-term sustainability of the system. Table 1 shows the required response in these three policy parameters.15
Moldova—Pension System’s Response to Demographic Changes 1/
The implied transfer from the budget in 2005 reflects the need for additional financing of the Social Fund, currently provided through drawndowns of accumulated deposits.
Moldova—Pension System’s Response to Demographic Changes 1/
| Base (2005) | Scenario A | Scenario B | |
|---|---|---|---|
| M/N | 1.21 | 1.48 | 1.82 |
| α | 0.29 | 0.35 | 0.44 |
| β | 0.26 | 0.21 | 0.17 |
| τ | 0.08 | 0.25 | 0.39 |
The implied transfer from the budget in 2005 reflects the need for additional financing of the Social Fund, currently provided through drawndowns of accumulated deposits.
Moldova—Pension System’s Response to Demographic Changes 1/
| Base (2005) | Scenario A | Scenario B | |
|---|---|---|---|
| M/N | 1.21 | 1.48 | 1.82 |
| α | 0.29 | 0.35 | 0.44 |
| β | 0.26 | 0.21 | 0.17 |
| τ | 0.08 | 0.25 | 0.39 |
The implied transfer from the budget in 2005 reflects the need for additional financing of the Social Fund, currently provided through drawndowns of accumulated deposits.
43. The scenarios highlight the trade-offs that may need to be contemplated in response to possible demographic shifts. If the required adjustment were shouldered fully by contributors, the contribution rate would have to rise from 29 percent to 35 percent in Scenario A, and to 44 percent in Scenario B. Alternatively, the replacement rate would need to fall to 21 and 17 percent in Scenarios A and B, respectively, from 26 percent in the base year. Without changes in the contribution and replacement rates, the state budget would need to increase its transfers to the pension plan from 8 percent of pension outlays to 25 and 39 percent under Scenarios A and B, respectively.
E. Policy Implications
Structural policies
44. Structural policies will be key when addressing most, if not all, policy issues associated with workers’ remittances. To moderate, and eventually reverse the current emigration trend and Moldova’s increasing dependence on workers’ remittances, establishing a good business environment is crucial. An improved business environment would (i) strengthen the incentives for recipients to invest in Moldova; (ii) increase the expected return on investment in Moldova, encouraging FDI and bringing much-needed know-how; and (iii) make it more likely for the higher educated labor force to seek and find work opportunities in Moldova. Many structural policies, such as legal reforms and anti-corruption measures would, furthermore, directly improve the quality of life and make it more desirable to live in Moldova. The fact that migration and workers’ remittances are both on a rising trend is not an exogenous shock, but is endogenously linked to the economic environment in Moldova. To bring about a change, the environment needs to be improved.
45. It has to be recognized that migration and remittances are very likely to continue to be important in Moldova, even with a much improved economic climate. The upward trend may end and even be reversed, but there will always be attractive opportunities abroad for many Moldovans, as long as barriers to work in other countries (regulatory, logistical, financial, psychological etc.) continue to be low. With reasonable economic prospects in Moldova, the difference would be that workers would leave to find better opportunities abroad rather than to avoid the problems at home, which appears to be the case today.
Monetary policy
46. Supporting structural policies will be necessary to reduce the pressure on monetary policy. Contributing to create a stable macroeconomic environment will be a big challenge for monetary policy if structural policies continue to lag. As argued above, monetary policy cannot fundamentally change Moldova’s competitiveness; only structural reforms can. Monetary policy is important in creating a stable macroeconomic environment, which certainly is important for the investment climate, but without effective structural policies, the effectiveness of monetary policy will continue to be severely hampered.
47. Under these circumstances, a flexible exchange rate regime, paired with a clear focus on low inflation as the overriding goal of monetary policy, appears to be a reasonable strategy ahead. Moldova is sensitive to real exogenous shocks (e.g. good or bad harvests, and changing economic conditions in the countries where remittances originate and exports are destined to); a flexible exchange rate regime helps in absorbing these shocks. Maybe more importantly, there is no clear better alternative at this juncture. Switching to a fixed exchange rate regime would require a clear commitment to policies consistent with such arrangement, not only from the NBM, but also from the government, and it would have to be backed up by strong political support. At this point, that does not appear to be a realistic alternative, and such a strategy would be risky. Of course, a flexible exchange rate regime also works better the more prudent and predictable economic policies are, but to give a false sense of stability by fixing the exchange rate could lead to dangerous imbalances. It is probably better if economic agents in Moldova learn how to cope with the existing uncertainties. In the medium term, a peg to the Euro or a currency basket such as the SDR could be a viable option, in particular if Moldova moves toward closer integration with the EU.
48. Low inflation should be the overriding goal of monetary policy and economic policies should be designed to be consistent with the publicly stated inflation goals. While at this point, Moldova does not fulfill all the requirements for a successful implementation of an inflation targeting regime, focusing on inflation seems to be the best option for monetary policy.16 Not only the required institutional commitment to price stability seems to be absent, but money demand is very uncertain, making it difficult to predict the effects of monetary policy on inflation. Still, since a fixed exchange rate regime does not appear as a viable option, focusing on low inflation seems to be the best alternative. Combined with strong structural policies, prudent fiscal policies, and increased de facto and de jure independence of the NBM, such strategy could be successful.
Fiscal policy
49. The short-term beneficial effects of emigration for the government financial position offer a margin for maneuver for a countercyclical macroeconomic policy. Emigration, by alleviating unemployment and providing a safety net to the population, has eased pressures on the budget, while boosting consumption- and import-related tax revenue collection. This newly-created cushion provides an opportunity to strengthen fiscal policy’s countercyclical role without jeopardizing medium-term fiscal sustainability. This strategy will require that tax revenue increases are not automatically utilized for new spending initiatives. Rather, it may be appropriate to increase fiscal saving when the domestic economy is overheating and inflationary pressures are on the rise. Thus, fiscal policy would lend more effective support to the central bank’s efforts to control inflation.
50. While there is uncertainty about the future evolution of the number of contributors and beneficiaries in the pension system, the analysis above suggest some practical steps that could be taken to strengthen its viability:
Efforts should be directed at broadening the contribution base by bringing a greater number of contributors into the system. The government decision to lower contribution rates (from 30 to 28 percent by 2006) is aimed at encouraging greater participation in the plan. In addition, the authorities intend to engage the business sector in a debate over Social Fund reform, including by seeking ways to broaden the contribution base, while lowering rates.
The link between contributions and benefits should be strengthened. This was one of the main objectives of Moldova’s pension reform, to be achieved initially through a blended system, where benefits would depend increasingly on past contributions. However, the weight of past contributions in determining pensions remains relatively small. Greater weighting of past contributions would make the link between contributions and benefits more transparent, encouraging greater participation in the scheme. It would also signal to those who choose to stay outside the system—including those who have decided to seek employment abroad—that they would need to assume greater responsibility for financing their own retirement.
F. Conclusions
51. Clearly, labor emigration and workers’ remittances have a very large impact on the Moldovan economy; structural policies will be key when addressing most, if not all, related policy issues. To end the upward emigration trend and the increasing dependence on workers’ remittances, establishing a good business environment is crucial. Monetary and fiscal stabilization policies will only work if the basic policy direction is sound, with effective structural reform implementation laying the ground for robust economic growth.
52. Monetary policy cannot fundamentally change long-term competitiveness without supporting structural reforms. At this point, a strategy to maintain a flexible exchange rate regime, paired with a clear focus on low inflation as the overriding goal of monetary policy appears to be a reasonable monetary policy alternative. Fiscal policy faces both short-term and long term challenges on account of emigration and remittances. The short-term beneficial effects provide an opportunity to strengthen fiscal policy’s countercyclical role. In the longer run, the pension system is likely to come under pressure; therefore, it will be important to improve the current system by broadening the contribution base and strengthening the link between contributions and benefits.
References
Castello-Branco, M., 1998, “Pension Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union (BRO)” IMF Working Paper 98/11.
Chami, R., Fullennkamp, C., and Jahjah, S., 2003, “Are Remittances Flows a Source of Capital for Development?,” IMF Working Paper 03/189.
International Republican Institute Baltic Surveys Ltd.-The Gallup Organization (2004), Nationwide Public Opinion Survey November/December 2004.
IMF, 1993, Balance of Payments Manual, 5th edition, Washington D.C.: IMF
IMF, 1998, Macroeconomic Accounts, Analysis, and Forecasting, Financial Programming Workshop, October 23-28, 1998, Volume 1, Washington D.C.: IMF
Mishkin, F.S., 2004, “Can Inflation Targeting Work in Emerging Market Economies?,” Festschrift in Honor of Guillermo A. Calvo, presented at a conference in honor of Guillermo Calvo, April 15-16, held at the IMF.
Mishra, P., 2004, “Emigration and Wages in Source Countries,” mimeo, Columbia University.
Rapoport, H., and Docquier, F., 2003, “The Economics of Migrants’ Remittances,” Draft chapter for Handbook of the Economics of Reciprocity. Giving and Altruism Eds. Varet, Ythier, and Ythier, North-Holland.
Ratha, Dilip, 2003, “Workers’ Remittances: An Important and Stable Source of External Development Finance,” in Global Development Finance–Saving for Stability in development Finance, Washington D.C.: The World Bank.
Yang, D., 2004, “International Migration, Human Capital, and Entrepreneurship: Evidence from Philippine Migrants’ Exchange Rate Shocks,” mimeo, University of Michigan.
Prepared by Erik Lundback. Milan Cuc contributed the section on the fiscal impact of remittances.
Temporary workers can be divided into two groups: (i) seasonal migrants, and (ii) non-permanent migrants, who have been abroad for less than one year.
See IMF (1993), Balance of Payments Manual, 5th edition, paragraphs 269-272, 302 and 352-355.
The data problems related to measuring workers’ remittances are widely acknowledged in the literature. See for instance Chami et.al. (2003), and Rapoport and Docquier (2003).
Data for Turkmenistan and Uzbekistan are not available.
GNI is defined as GDP plus the net factor income of residents from abroad, and GNDI is defined as GNI plus net current transfers received from abroad (see e.g. IMF, 1998).
According to Rapoport and Docquier (2003), this network-effect is well documented in the sociological literature.
Rapoport and Docquier (2003) present a model of economic growth and remittances, where a country ends up in a long-run equilibrium with continuing migration and little investment if the costs of migration are low, and the entry costs for investing and becoming an entrepreneur are high—a situation very pertinent to Moldova.
There are few formal studies on the effects of emigration on wages. Mishra’s econometric analsys (2004) finds a strong positive effect of emigration on wages in Mexico.
Yang (2004) finds that in the Philippines remittances are indeed invested in education.
As noted in paragraph 10, when foreign currency in the informal sector is brought into the formal balance of payments, this could reflect previous balance of payments transactions (e.g. cash remittances in the late 1990s may appear in the balance of payments in 2004 as errors and omissions).
For a more thorough analysis of Moldova’s external competitiveness see Chapter III.
This trend is expected to be accentuated by the decision to phase in reductions in personal income and corporate profit tax rates in 2004-07 (see Chapter VI).
All figures are on an annual basis.
The simulations assume that only one parameter at a time is selected for adjustment. If the necessary adjustment were to be achieved by adjusting all parameters simultaneously, the required change in each parameter would be less.
See Mishkin (2004) for a discussion of the requirements for inflation targeting.
