From Evidence to Policy: Supporting Nepal’s Trade Integration Strategy Policy Note 1 Trade Imbalances and Remittances: Ensuring Macro Stability Alberto Portugal Emir Zildzovic Executive Summary The report aims at addressing the following questions: what are the underlying drivers of the trade imbalances and their relative contributions to the deficit (e.g. domestic private expenditure fueled by remittances versus public expenditures, relative prices, etc.)? What are the paths to be expected for the trade balance and the current account in the medium run, based on reasonable forecast assumptions for the main identified determinants? What are the implications of high remittances for the competitiveness in the short and long-run? How can Nepal maximize the positive impact of remittances in the economy? Nepal’s trade balance in goods and services has been in deficit for many years. The large deficit has not resulted in a substantial accumulation of net foreign liabilities, as it has been largely financed with workers’ remittances from abroad. Indeed, remittances are the largest component of the current account. In the case of Nepal, the current account balance is roughly equal to the trade balance plus remittances; hence, the report focuses mostly on the trade balance and remittances. More specifically, we estimate the impact of determinants of the trade balance, and the last section focuses on the impact of remittances in the real exchange rate. We found that the trade deficit is highly influenced by changes in remittances and terms of trade. In fact, a large part of the increase in the trade deficit is explained by the remittance inflows. Nepal’s trade deficit is moderately persistent and sensitive to policy malleable variables, such as government expenditures or credit. On the short-run policy front, government expenditures had a significant contribution on the trade deficit, though at a lower scale. We simulated future paths of the trade balance and the current account balance over 2014-2019 under five scenarios for the following determinants: (i) remittances (ii) government expenditures, (iii) output gap, (iv) oil imports, and (v) real effective exchange rate (REER). Overall, higher levels of domestic demand, inflation, government expenditures and remittances related to reconstruction works are likely to lead to higher trade deficits. The current account is expected to turn sharply into a deficit in 2016 due to reconstruction expenditures and improve gradually thereafter. Nepal experienced a significant increase in remittances inflows after 2000, which reached 25-30 percent of GDP over the past years. The literature has noted the positive impact of remittances on an economy. First, the literature found that remittance inflows reduce volatility of output and stimulate economic growth. Second, remittances support the development of the financial sector. By relaxing credit 2 constraints, they stimulate investments and future growth. Yet, remittance inflows may exert negative effects in the economy. Large inflows can lead to a real exchange rate appreciation and the subsequent loss of export competitiveness. This Dutch-disease effect operates through two channels. First, higher inflows of remittances lead to increased spending in both tradeable and non-tradeable goods (spending effect). As prices of tradeable goods are set in international markets, increasing demand leads to higher prices of non-tradeable goods. The relative increase in prices of non-tradeable goods results in an appreciation of the real exchange rate. Second, an increase in remittances stimulate the transfer of resources from the tradeable to the non-tradeable sectors (resource movement effect) as the relative profitability of producing tradeable goods fall. This is due to an increase in wages and the cost of other production factors led by the higher demand for non-tradable goods, which puts additional pressure on the real exchange rate to appreciate. We assess the impact of remittances on Nepal’s real exchange rate and attempt to distinguish between their short and long term impact. Potential negative effects of remittances on the competitiveness can be offset in the long run if these inflows boost capital formation and investments. Some countries implemented measures aimed to stimulating more remittance inflows and shift their structure towards investment. We found that the exchange rate adjusts to changes in workers’ remittances in the long-run. The estimated coefficient of remittances implies that a 10 percent nominal increase in remittances leads to a 0.5 percent of real appreciation in the long run and the subsequent loss of export competitiveness. The potential negative effects of remittances on the competitiveness can be offset in the long run if these inflows boost capital formation and investments. Yet, a large amount of remittances inflows in Nepal seem to be spent in consumer goods, including importing consumer goods. Governments have often offered incentives to increase remittance flows and to channel them to productive uses, but such policies can also generate unintended effects. Tax incentives may attract remittances, but they may also encourage tax evasion. Matching-fund programs to attract remittances from migrant associations may divert funds from other local funding priorities. Efforts to increase savings and improve the allocation of remittances should also be accomplished through improvements in the overall investment climate and the business environment. 3 Context Nepal’s external sector is under-performing and increasingly vulnerable. Nepal’s trade deficit in goods and services reached 20 percent of GDP in 2007, and since then, it has grown steadily, as shown in Figure 1. A large deficit has not resulted in a substantial accumulation of net foreign liabilities, as it has been largely financed with workers’ remittances from abroad. Figure 1. Current account balance and its components Trade balance Income balance Current transfers Current Account 40% 30% 20% 10% 0% -10% -20% -30% -40% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Nepal had a significant increase in remittances inflows after 2000 . As a result of the massive labor emigration, remittances reached 22 percent of GDP in 2009, previous the global recession, and went down slightly subsequently. 1 Since 2010, remittances as a share of GDP have increased continuously until reaching almost 28 percent of GDP in 2013. 1 They are expected to have increased substantially in the aftermath of the earthquake that devastated the Kathmandu Valley in April 2015, although data are not yet available. 4 Nepal has become less integrated in the global marketplace. While Nepal’s overall openness to trade was similar to that of other countries at similar levels of income during the 1990s (i.e. with openness levels above Uganda or Bangladesh), it has fallen well below the average thereafter. In the last 10 years the country’s export performance has been poor: merchandise exports growth have fell from an average of 19 percent per year in the 1990s to 0.65 percent in 2000s. Nepalese exports in 2013 were barely 1.1 times of 2000’s, while imports quadrupled over the same period. Figure 2: Openness to Trade 2.a Average 1996-98 Openness to Trade (Average 1996-1998) 250 200 Trade to GDP (%) 150 TJK 100 KGZ BTN LAO NPL 50 BOL UGA CHN BGD BDI IND 0 4 6 8 10 12 Log of GDP per capita (PPP, av.) 2.b. Average 2009-2013 Openness to Trade 2009-2013 200 Trade to GDP (%) 150 KGZ 100 BTN TJK LAOBOL UGA 50 BDI BGD IND CHN NPL 0 6 8 10 12 Log of GDP per capita (PPP, av.) 5 Nepal has historically been a net exporter of services, but imports of services have increased substantially in recent years. While in 2000, exports of services were about two and a half times imports of services, by 2013, they were only 21 percent higher. Box 1 WHAT DOES THE CURRENT ACCOUNT BALANCE SHOW? The current account balance of a country comprises three subcomponents: the trade balance, the income balance, and the transfer balance. The trade balance records all transactions with the rest of the world related to the exchange of goods and services. The income balance records net interest and dividend payments and earnings of domestically owned firms operating abroad, while the last component, the transfer balance reflects net payments (that do not correspond to purchases of any good, service, or asset) received from the rest of the world. Notably, the transfer balance includes remittances received from nationals working abroad minus remittances sent abroad by foreigners working in the domestic economy. Current Account = Trade Balance + Income Balance + Current Transfers Balance The current account can also be seen as the difference between what an economy produces and what it consumes and invests in a given period (or alternatively, the difference between saving and investment). When a country consumes and invests more than what it produces, it needs to borrow from the rest of the world to finance that gap. Hence, the current account deficit reflects what a country borrows from the rest of the world to finance its investment and consumption in excess of what it produces. Current Account Balance= [Output (GDP) – Consumption] – Investment Or Saving – Investment If Saving > Investment  The economy is a net lender of the rest of the world If Saving < Investment  The economy is a net borrower of the rest of the world The saving and investment can be decomposed further into portions attributable to the public and private sectors. Thus, the current account balance can be expressed as the sum of the private sector’s surplus (private saving minus private investment, Sp-Ip) and the government’s surplus (tax revenues minus government expenditures, T-G). This is why current account deficits and fiscal deficits are often referred as the “twin deficits”. Given the private sector balance (Sp-Ip), increases in fiscal deficits co-move with increases in current account deficits. Current Account Balance = (Savingp - Investmentp) + (Tax – Government Expenditures) From the above relationship, one can infer that a positive or a negative current account balance is not bad per se, as it depends on the country’s specific circumstances. For example, a relatively poor country in the process of investing in infrastructure but constrained by low levels of domestic saving will finance its infrastructure projects through foreign capital and will run a current account deficit. In this context, the deficit may be an inevitable path to economic development. On the other hand, current account deficits also arise as a consequence of public sector deficits. When governments consume more than what they collect in taxes, countries run current account deficits, all other things being held equal. Yet, these deficits are unlikely to be associated to faster future growth, and concerns about current account sustainability will be raised. These simple examples stress the importance of identifying the underlying sources of current account deficits. For instance: are they mainly driven by fiscal deficits, by low private saving rates, or by high private investment? 6 Source: Authors’ elaboration. Nepal has ample public and private saving that could be mobilized for investment. Compared to other South Asian countries, such as Bangladesh, India, and Sri Lanka, Nepal’s aggregate saving has been among the largest in the region controlling for its size. Nepal saving rate was almost 34 percent of GDP in 2014. Figure 3: Aggregate saving rates (percent of GDP) Bangladesh India Myanmar 40.0 Nepal Sri Lanka Vietnam 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 1990 1993 1996 1997 1999 2000 2002 2003 2006 2009 2012 1991 1992 1994 1995 1998 2001 2004 2005 2007 2008 2010 2011 2013 2014 Two features may raise concerns about the current account sustainability. First, export earnings are concentrated in a narrow set of export products and markets. Second, Nepal is highly dependent on remittances to finance its increasingly larger trade deficit. 7 What is behind Nepal’s trade balance deficit? In this section, we analyze the determinants of trade deficits (TD) over the period 1994-2013. The section is structured as follows. First, we discuss the relationship of Nepal’s trade deficits to a number of prospective determinants by estimating a model2. Second, we discuss about the variables that contribute the most to trade dynamics in Nepal, as estimated by the model. The current account is equal to the sum of the trade balance, income balance, and transfers balance, as discussed previously. In the case of Nepal, the income balance has been close to zero during most of the covered period, and remittances account for the transfer balance. In fact, remittances are the largest component of the Nepalese current account. Thus, the Nepalese current account is roughly equal to the trade balance and the inflows of remittances. Given the particular configuration of the Nepalese current account, we apply the methodology to of current account determinants (described in Box 2) but focus on the trade balance and exclude remittances. Indeed, remittances, the other large component of the current account, are largely determined at the household level based on more lucrative opportunities abroad and family factors, instead of macroeconomic variables. In next section, we combine the estimates of the determinants of the Nepalese trade balance provided by the model with assumptions regarding the expected evolution of these determinants to compute their potential contribution to the dynamics of trade balances as well as current account balances. Box 2 CONCEPTUAL FRAMEWORK FOR THE ANALYSIS OF DETERMINANTS OF THE CURRENT ACCOUNT BALANCE The current account balance reveals the difference between national saving (private and public) and national investments (private and public), as described in Box 1. To identify the underlying drivers of the current account balance, we relate to the factors affecting private and public savings and investments with other related factors that may affect the current account directly (with indirect effects on saving and investment): CA X CA   S P  X S   I P  X I   S G  I G 2 For more on the methodology, see Box 2 and the Toolkit for the Analysis of Current Account Imbalances (Cusolito and Nedeljkovic, 2013). 8 where XS are private consumption/saving determinants, XI denotes factors that affect private investment and XCA denotes factors that may influence the current account directly (for example, the export/import determinants, past FDI inflows). Assuming exogeneity of the trade drivers XCA the current account balance is defined as: CA  g  X S , X I , S G , I G , X CA  and the function g(•) is assumed linear. The prospective determinants are classified according to their sensitivity to policy decisions, and the sign of their expected effect on the current account balance. PROSPECTIVE DETERMINANTS OF THE CURRENT ACCOUNT BALANCE EXPECTED EFFECT ON THE CURRENT ACCOUNT BALANCE Positive Negative Ambiguous SENSITIVITY TO POLICY DECISIONS Fiscal Balance, REER, Sensitive to policy in the short run credit Sensitive to policy in the Openness, FDI medium run Relative income, Macro Relative GDP growth Sensitive to policy in the long run uncertainty External Terms of trade Others Lagged CA/GDP NIIP Source: Authors’ elaboration based on Cusolito, A. and M. Nedeljkovic (2013). The dependent variable in the analysis is the ratio of the current account balance to GDP. The explanatory variables and their relationship with the current account balance are described below:  Lagged current account balance: At an annual frequency, CA balances tend to show high persistence, associated to habit formation in consumption and saving, or agglomeration effects in investment.  Lagged Net International Investment Position (NIIP): Net foreign assets can affect the current account balance in two ways. First, a large stock of foreign liabilities will require a country to run current account surpluses to pay them off. Second, the country will still pay interest on those liabilities, and thus the current account will become more imbalanced.  Terms of trade (ToT): A positive ToT shock can improve the current account via increased saving due to larger current income relative to permanent income (the Harberger-Laursen-Metzler effect). On the other hand, ToT shocks can also affect the optimal capital stock and change investment plans, leading to more current account deficit. The greater the persistence of the shock, the more dominant is the investment effect. For an oil importer like N, oil prices directly affect the oil import bill, thus the CA. 9  Fiscal balance: To the extent to which Ricardian equivalence hold, there should be no relationship between the fiscal and the CA balance. However, empirical studies in both developed and developing economies typically reject the hypothesis and suggest positive relationship between the two deficits.  Openness: Trade openness has ambiguous effects on the current account balance. Less open economies may import less, which may reduce the current account deficit. However, the same countries may have difficulties servicing external liabilities, resulting in higher debt service costs and a greater current account imbalance. On the other hand, greater openness typically allows countries to undertake more investment and to finance the resulting current account deficits with capital flows from abroad. Also, international trade is an important conduit for the transfer of technology, leading in the long run to economic development, thereby improving the current account balance.  Real Effective Exchange Rates (REER): REER appreciations induce an expenditure switching effect away from domestic goods and into foreign goods, for a given level of expenditure, which increases the CA deficit, all else equal.  Foreign Direct Investment (FDI): FDI has ambiguous effects on private domestic investment and the current account. It can crowd out domestic investment when local and offshore firms compete for scarce domestic resources (e.g. labor or finance). FDI may also generate local spillover that ‘crowd in’ domestic investment. Gross FDI may also worsen the current account, depending on import content and the amount of profits repatriated and the export orientation of multinationals.  Relative GDP growth: The effect on the CA balance depends on agents’ expectations about the implications of gr owth for future income. If agents consider it permanent, then saving rates could decline, increasing the CA deficit. If instead it is perceived as temporary, saving will increase and the current account balance will increase.  Credit to the Private Sector: Proxies financial deepening, and aggregate demand. Relaxed borrowing constraints can reduce private saving (increase CA deficits). Also, if reduces transaction costs and improves risk management, may encourage private saving (decrease CA deficits).  Relative income: Small, developing economies will run current account deficits as they accumulate capital goods. Eventually, the country will be sufficiently developed to pay its debts by running CA surpluses.  Demographics: Faster expected aging is related with an improvement in the CA. If agents expect an increase in the share of dependent population in the future (increase in the dependency ratio relative to the current one), they will be expected save more, which will improve the CA balance.  Macroeconomic uncertainty: A more uncertain macroeconomic environment is expected to raise precautionary saving and reduce investment, according to the buffer stock theory. Some of the factors proposed as potential determinants are in fact jointly determined with the current account balance. The most notorious case is that of the real exchange rate, for example. To deal with this ‘simultaneity’ problem, we instrument variables that are likely to be endogenously determined with the current account balance. The instruments consist of lagged values of the variable in question, which are predetermined at time t. 3 Pre-determination, however, does not necessarily imply ‘exogeneity’ if anticipation effects are present. For these reasons, a word of caution is in order. The results presented here should be read as conditional associations between variables, rather than strict causal relationships. Source: Authors’ elaboration based on “Toolkit for the Analysis of Current Account Imbalances” (Cusolito, A. and M. Nedeljkovic). 3 This methodological choice is based on data availability (both in terms of the scope of potential additional explanatory variables, and on the time period available). 10 How sensitive is the Nepalese trade balance to its determinants? The model used to estimate the trade balance determinants performs well on average. Figure 4 shows the actual and the predicted trade balance dynamics. Overall, the difference between the actual trade balance and the one predicted by the model is small over the entire period. The model underestimates the size of trade deficit in 2009 and 2010, but predicts better the more recent evolution. Table 1 reports the estimated coefficients for each of the prospective determinants, along with their standard errors and the standardized coefficients. Standardized coefficients show the effect of a change in one standard deviation in one of the trade-balance determinants (measured also in terms of standard deviation of the trade balance). To facilitate reading the results, Table 2 reports the sign of all estimated coefficients and the relative importance of their effects on the trade balance. All coefficients are significant and have expected signs and magnitudes. Figure 4: Trade balance: actual level vs predicted level 0.0% Actual TB Predicted TB -5.0% -10.0% -15.0% -20.0% -25.0% 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Source: author’s estimates Table 1: Estimated effects of trade balance determinants in Nepal Model used in simulations Standard Standardized Variable Coefficient Error coefficient Dependent variable Trade balance (net of oil balance) Trade Balance (1 year lag) 0.18 0.01 0.17 11 Remittances (1 year lag) -0.52 0.02 -0.68 Terms of trade (log, change) 0.09 0.01 0.24 Government expenditures (share of GDP) -0.16 0.01 -0.05 Relative GDP PPP per capita (lag, log) 0.03 0.01 0.01 Relative openness 0.00 0.01 -0.01 Aging speed (lag, 2 year average)* 0.04 0.00 0.03 Real exchange rate (CPI based, change) -0.20 0.06 -0.10 FDI (% of GDP, 2 year average) 0.29 0.53 0.01 Output gap (lag)** -0.41 0.23 -0.07 Credit change (share of GDP) -0.17 0.06 -0.11 Political and macroeconomic uncertainty** 0.00 0.00 0.01 (lag, 2 year average) R squared 0.96 Notes: + (-) denotes a standardized coefficient between 0 and 0.1 (0 and -0.1). * Defined as the current dependency ratio relative to the one in 20 years. ** Real GDP- potential GDP ***Calculated as the first principal component of inflation volatility, unemployment, VXO, and World Bank Governance Indicators for corruption, government effectiveness, political stability, rule of law, regulatory quality and voice and accountability. Nepal’s trade deficit is highly influenced by changes in remittances and terms of trade, as measured by their standardized coefficients. The strong and negative effect of remittances is what expected: higher remittances will increase demand for foreign goods, and will also have an indirect impact through the appreciation of the real exchange rate. The last section of the report provides a more complete discussion on the impact of remittances on the real exchange rate. Unsurprisingly, changes in the terms of trade have a high and positive impact on the trade balance. The trade deficit is moderately persistent, which is associated to habit formation in consumption and saving. The persistence effect of 17 percent is smaller than typically found across groups of countries of all levels of income, but is in line with the results found when implementing this type of current account imbalances analysis for other developing countries, such as Morocco (17 percent), Georgia (23 percent), or Turkey (16 percent). From a conceptual point of view, persistence may be related to habit formation in consumption and saving, and suggests a certain degree of inertia in the current account and the trade balance. 12 Nepal’s trade deficit is sensitive to policy malleable variables such as change s in credit or government expenditures. The results suggest that a one percent (as a share of GDP) increase in the fiscal expenditures increases the trade deficit by 0.16 percent. This result strengthens a case on the importance of prudent fiscal policy to prevent expanding the trade deficit. Expansions of credit to the private sector are also associated with trade deficits. Relative income per capita (measured as GDP at PPP) has a positive effect on the trade balance in Nepal. As the country becomes richer, it tends to save more in order to pay for investment made during the catching up period, in line with the convergence hypothesis. Table 2: Signs and relative importance of estimated effects of trade balance determinants in Nepal Variable Estimated effect Trade balance (1 year lag) ++ Remittances (1 year lag) --- Terms of trade (log, change) +++ Government expenditures (share of GDP) - Relative GDP PPP per capita (lag, log) + Relative openness - Aging speed (lag, 2 year average)* + Real exchange rate (CPI based, change) - FDI (% of GDP, 2 year average) + Output gap** (lag) - Credit change (share of GDP) -- Political and macroeconomic uncertainty*** + (lag, 2 year average) Note: (+/-) denotes a positive/negative relationship between the determinant and the CA balance. +++ (---) denotes a standardized coefficient above (below) 0.2 (-0.2), ++ (--) denotes a standardized coefficient between 0.1 and 0.2 (-0.1 and -0.2), + (-) denotes a standardized coefficient between 0 and 0.1 (0 and -0.1). * Defined as the current dependency ratio relative to the one in 20 years. ** Real GDP- potential GDP 13 ***Calculated as the first principal component of inflation volatility, unemployment, VXO, and World Bank Governance Indicators for corruption, government effectiveness, political stability, rule of law, regulatory quality and voice and accountability. Although small, FDI inflows had a positive impact on the trade balance. Although foreign capital is small in Nepal, foreign companies having entered the Nepalese market had a bigger positive effect on exports than on imports (through increased demand for foreign intermediates and capital goods). Demographic changes, measured by the aging speed, have a small and positive effect on the trade balance. Given the long horizon over which the demographic trends materialize, the overall impact of the demographic change in Nepal on the trade balance is positive, in line with the decline in the dependency ratio. Macroeconomic uncertainty is also estimated to have a small positive impact on the trade balance. Actually, as uncertainty mounts agents tend to increase their precautionary savings, which reduces the consumption, including the consumption of foreign goods. 14 Box 3 HOW TO READ FIGURE 5 ? To better interpret Figure 5, which plots the contributions of CONTRIBUTIONS OF DIFFERENT FACTORS TO THE different variables to Nepal’s trade deficit, we use the TRADE DEFICIT IN THE FICTIONAL UQBAR REPUBLIC imaginary case of the Fictional Republic of Uqbar, for the (2012-2013) period 2012-2013 plotted in this box. 5% Uqbar run a trade deficit of 4 percent of GDP on average Trading Partners' over the period. The estimated model identified how Growth sensitive Uqbar’s trade deficit was to a number of 4% Relative prospective determinants. Estimates predicted a deficit of 3 openness percent of GDP, one percentage point away from the actual 3% 4 percent. In fact, the prediction of 3 percent of GDP is the REER sum of the contributions of all trade balance determinants. change 2% Notice that some of the determinants had a negative FDI contribution to the trade balance (i.e. a positive contribution 1% to the trade deficit) while others had a positive contribution Export to the trade balance (i.e. offsetting the trade deficit). Prices 0% Altogether, the variables with positive impact contributed to a trade surplus of 5 percentage points of GDP, while the variables with negative impact contributed to a deficit of 8 Government -1% percentage points of GDP. Thus, the net predicted effect was a trade deficit of 3 percent of GDP. -2% The variables with positive contributions to the trade balance are plotted above “zero”, while those with negative Remittan -3% contributions are plotted below zero in the graph of this ces Box. For instance, export prices, FDI inflows, changes in the REER, relative openness and trading partners’ growth were -4% found to offset the trade deficit of Uqbar in different magnitudes (for example, the trading partners’ growth Relative -5% contributed to offset the deficit by 1 percentage point of income GDP). Instead, government expenditures, remittances, relative income and credit growth dynamics contributed to -6% increase the trade deficit. For example, the government expenditures contributed to the deficit by 2 percentage Credit -7% points of GDP. growth -8% Source: Authors’ invention 15 Source: Authors’ elaboration What are the factors that contribute the most to trade dynamics in Nepal? In addition to its sensitivity, it is crucial to understand how each determinant has contributed to the actual trade balance dynamics. The distinction between sensitivity and contribution of a given determinant is important. The sensitivity refers to how much the trade balance changes given a marginal change in one of its determinants. By combining information on the sensitivity to a given determinant with the actual changes of a given determinant over the period, we can estimate how much of the trade deficit has been attributable to actual changes in that determinant. For instance, Nepal’s current account is highly sensitive to remittances (see standardized coefficient in Table 1) which increases the CA balance. Yet, the contribution of remittances to the trade balance of Nepal before 2002 is small and increases thereafter, as shown in Figure 5. Figure 5 CONTRIBUTIONS OF KEY VARIABLES TO THE TRADE BALANCE (AS % OF GDP) 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 10% 5% 0% -5% -10% -15% -20% Terms of trade (change) REER (change) Output gap (lag) Credit change (share of GDP) -25% Trade balance (net of oil, lag, share of GDP) FDI (2 year average) Remittances (lag, share of GDP) Government expenditures (share of GDP) -30% Political&economic uncertainty Relative openness Agging speed (2 year average, lag) Relative income (lag) -35% Actual TB Predicted TB Source: Authors’ calculations. Another way of looking at the evolution of the determinants contribution to the trade balance is to focus on how these contribution changed between 2000 and 2013, as shown in Figure 6. The contribution of remittances to the trade deficit has increased substantially (by 12 percent of GDP) 16 between 2000 and 2013, the end year of our analysis. The contribution of the terms of trade to the trade deficit has increased by 2.8 percent of GDP. Figure 6 CONTRIBUTIONS TO TRADE BALANCE - CHANGE IN CONTRIBUTIONS TO THE TRADE BALANCE FROM 2000 TO 2013 (AS A PERCENT OF GDP)) 2.0% 0.0% Output gap (lag) -2.0% Credit change (share of GDP) -4.0% Political&economic uncertainty -6.0% Trade balance (net of oil, lag, share -8.0% of GDP) -10.0% Remittances Relative openness -12.0% Government expenditures (share of GDP) -14.0% REER (change) -16.0% Relative income (lag) -18.0% Terms of trade FDI (2 year average) -20.0% Agging speed (2 year average, lag) -22.0% change compared to 2000 Source: Authors’ calculations. 17 What are the Likely Paths of Nepal’s Trade Balance and Current Account Balance? How sustainable is Nepal’s external position? What would happen to the trade deficit and to the current account balance if the key determinants continue along their current trend or if they deteriorate? What if the government introduced policy changes? The external sustainability exercise in this section aims at answering these questions. Current account sustainability is defined as the stable state in which the current account balance generates no economic forces of its own to change its trajectory. In particular, the current account balance is sustainable if the continuation of current government policy stance and private sector behavior does not result in future rapid policy shifts (such as, for example, a sudden policy tightening causing a large recession) and/or substantial changes in other economic variables, such as large exchange rate depreciations or interest rates hikes (Milesi-Ferretti and Razin, 1996). The notion of sustainability does not provide a clear criterion for assessing country’s external vulnerability as it incorporates agent’s expectations of future policies rather than the policies themselves. The methodology focuses on another aspect of sustainability: the current account deficit may be sustainable as long as there is foreign funding willing to finance it. The analysis does not impose any steady-state assumption on the evolution of the economy, as these assumptions typically do not hold for emerging countries. To simulate future paths of the trade balance over the period 2014-19, we combine projections for the trade balance determinants (mostly from the IMF World Economic Outlook and our own calculations) with the estimated elasticities of the trade balance with respect to each of these determinants. We also add projections of remittances and other transfers (including donations for reconstruction) to the simulated paths of the trade balance to obtain future paths of the current account balance over the same period following the identity: Current Account Balance (CAB) = Trade Balance(TB) + Remittances + Other transfers (incl.donations) Simulations also take into account the expected economic impact of the earthquake that struck Nepal on April 25, 2015, causing widespread damage and devastation. Box 4 below provides IMF’s account on the expected economic impact of the earthquake on macroeconomic variables that are relevant for 18 our simulations, while Figure 7 summarizes how the changes in these economic variables will affect the current account and the trade balance. Box 4 THE ECONOMIC IMPACT OF THE EARTHQUAKE Growth is expected to slow down. On June 8, Nepal’s Central Bureau of Statistics released a revised GDP projection for 2014/15 with growth falling to 3.4 percent in the year to mid-July 2015, compared to staff’s pre-earthquake baseline forecast of 5.0 percent. The tourism sector which generated foreign earnings for about 2½ percent of GDP last year, has been particularly affected. As economic activity recovers and reconstruction gains momentum, growth is expected to gradually rebound to around 5.5 percent in 2016/17. Based also on experience in other fragile countries struck by natural disasters, potential growth is projected to be adversely affected by the earthquake, falling to around 4 percent over the medium term. Inflation pressures are likely to rise. Losses in agricultural production and damage to transport systems will lead to reduced supply of agricultural products, which account for some 40 percent of the CPI basket. Stepped-up foreign aid and higher inflows of remittances would further boost the liquidity in the financial system, putting pressure on the central bank which has been reluctant to sterilize foreign inflows. Over time, however, as agricultural production recovers and transportation infrastructure improves, inflation pressure should ease. The fiscal impact of the earthquake will also be significant. Revenue losses are unlikely to be fully offset by higher duty collection from increased reconstruction-related imports (to the extent these are ODA-financed, they may enter duty free). The much greater impact on the budget will be on the expenditure side because of damage to infrastructure and government properties. In addition to the reconstruction cost in the public sector, the government will likely have to provide financial assistance for the recovery of the business sector and to households, particularly for housing. Financial institutions may also need assistance to help overcome the effects of the earthquake (see last point below). Donor support is expected to help fund a large part of the recovery and reconstruction expenses, but the government may also need to borrow more to meet the increased spending needs. Thus, both the fiscal deficit and public debt could likely increase in the medium-term. The external current account will likely be pushed into deficit. Imports of reconstruction-related materials will rise. Tourism receipts, a key source of Nepal’s foreign exchange earnings, could fall by some 1½ percent of GDP in 2015/16 compared with 2013/14, and experience in other countries suggests that recovery could take several years. A temporary surge of remittances is likely as the Nepalese diaspora and migrant workers send more money home to support the reconstruction efforts. However, these one-off higher inflows will be more than offset by higher imports, pushing the current account to a deficit of about 4 percent of GDP on average during the next 5 years. An urgent balance of payments has arisen, reflected in a financing gap. Without the mobilization of substantial exceptional donor financing, the deterioration in the external current account would cause the central bank’s foreign reserves to fall significantly in 2015/16 and over the medium term. As illustrated in Table 6, without the RCF disbursement and exceptional support from other donors—which could in part be catalyzed by the RCF disbursement—central bank reserves would fall to about 5 months of imports. This is well below Nepal’s reserve adequacy metric suggesting that reserves should be maintained at the current level of about 7 months of imports (Box 2). It is envisaged that with concerted support from the Fund and development partners, Nepal’s official reserves could be maintained at about 7 months of prospective imports (excluding construction-related imports) over the next few years. The financial sector’s asset quality would be expected to deteriorate. The damages and economic disruption caused by the earthquake could affect the loan portfolio of banks, microfinance institutions and cooperatives, particularly in rural areas where borrowers lost lives and livelihoods. Initial estimates of the financial hit to the banks (NR 38 billion or about 1.8 percent of GDP) and the insurance sector (NR 3 billion, net of reinsurance provided by foreign reinsurers) seem manageable. However, more data and diagnostics are needed to allow accurate assessments of the impact of damage to real estate and there could still be a need for budgetary support for the financial sector. Source: IMF (2015), “ Nepal: Request For Disbursement Under The Rapid Credit Facility”, July 17 19 Figure 7 THE IMPACT OF EARTHQUAKE ON FUTURE TRADE AND CURRENT ACCOUNT BALANCES? Source: Authors’ construction based on IMF (2015) We simulate future paths of the trade balance (TB) and current account balance (CAB) over 2014-2019 under five scenarios for the five following determinants: (i) remittances (ii) government expenditures, (iii) output gap, (iv) oil imports, and (v) real effective exchange rate (REER). One of the scenarios is a baseline scenario described in next paragraph. The other four scenarios are constructed around it. Baseline values for relative trade openness and relative income are drawn from IMF WEO. The FDI projection is taken from the IMF Article IV (2014). The aging speed variable is calculated using UN demographics projections. Political and economic uncertainty variables are set to increase in 2015 and decline afterwards. The terms of trade are expected to improve in 2015 due to declining oil prices but to worsen gradually afterwards. Credit growth is expected to pick up backed by the reconstruction. 20 Under the baseline scenario, the assumptions regarding the trajectory of the five determinants are as follows: Output gap is calculated using WEO projections for the real GDP4. After a negative gap in 2014 and 2015, the scenario assumes a positive gap due to increased demand reflecting reconstruction efforts. Real effective exchange rate (REER): After a sharp appreciation in 2015 (due to higher inflation), a real appreciation of 3 percent is expected by the end of the simulation period. Government expenditures are in line with IMF WEO projection. After a decline in 2014, total increase of 5 percent of GDP is expected by 2019 in line with increasing infrastructure investments. Remittances are expected to increase in 2015-2016 as emigrants support reconstruction efforts and to return to their 2014 level afterwards (in line with the empirical evidence, see WB, 2009). Oil imports are expected to follow the forecasted developments of oil prices. We assume that imports increased by 25 percent in 2014 and by additional 20 percent in 2015. From 2016, a gradual decline of 5 percent per year is assumed. These rates imply oil imports of about 5 percent of GDP in 2019. Results: simulations for the trade and current account balances Overall, higher levels of domestic demand, inflation, government expenditures and remittances related to reconstruction works lead to higher trade deficits. Figure 8 shows the forecast for the trade balance under different scenarios for four determinants, as oil imports (the fifth determinant) is included in the trade balance. Namely, the baseline path as discussed in the previous paragraph, two expansion scenarios, and two contraction scenarios for each simulated determinant. For instance, the comparison between the strong increase and the strong slowdown in the output gaps scenarios in Figure 8.a, implies a difference in the trade deficit of two percent in the long run. Figure 8: Simulations for the trade balance (TB) under different scenarios a. TB in alternative output gap scenarios (% of b. TB in alternative real exchange rate GDP) scenarios (% of GDP) 4 Estimates of output gap are obtained using the Hodrick–Prescott filter with smoothing parameter 40, over the period 1990-2019. 21 baseline moderate increase (1pp>weo) baseline (2% apreciation per year) appreciation (2pp>baseline per year) strong increase (2pp>weo) moderate slowdown (1ppbaseline per year) depreciation (2pp>baseline per year) depreciation (4pp>baseline per year) strong slowdown (2ppweo) baseline (2% apreciation per year) appreciation (2pp>baseline per year) strong increase (2pp>weo) moderate slowdown (1ppbaseline per year) depreciation (2pp>baseline per year) strong slowdown (2ppbaseline per year) 2014 2015 2016 2017 2018 2019 2014 2015 2016 2017 2018 2019 6% 6% 4% 4% 2% 2% 0% 0% -2% -2% -4% -4% -6% -6% c. CAB in alternative government expenditures d. CAB in alternative remittances scenarios (% scenarios (% of GDP) of GDP) baseline moderate expansion (2 pp a year) baseline strong expansion (4 pp a year) weo (pre-earthquake) moderate increase (to 30% of GDP in 2019) strong increase (to 33% of GDP in 2019) moderate consolidation (2pp a year) no change 2014 2015 2016 2017 2018 2019 strong decline (to 20% of GDP in 2019) 6% 2014 2015 2016 2017 2018 2019 5% 6% 4% 4% 3% 2% 2% 1% 0% 0% -2% -1% -2% -4% -3% -6% -4% -5% -8% Finally, oil prices dynamics have important implications for the evolution of the trade balance and the current account balance, as shown in Figure 10. Figure 10: Simulations for the trade balance (TB) and current account balance (CAB) under different oil prices scenarios a. TB in alternative oil prices scenarios (% of b. CAB in alternative oil prices scenarios (% of GDP) GDP) 23 baseline (decline by 2015 and increase after 2016) weo moderate improvement (1pp>baseline) moderate improvement (1pp>baseline) strong improvement (2pp>baseline) strong improvement (2pp>baseline) moderate worsening (1pp