Economic growth and balance of payments constraint in Viet Nam

.VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review Economic Growth and Balance of Payments Constraint in Vietnam Pham Sy An* Introduction ABSTRACT This paper focuses on the question of whether economic growth was constrained by Vietnam’s balance of payments during the 1990-2004 period, by using the model developed by Thirlwall. Based on quarterly and annual data from the period, we found that economic growth was indeed constrained by the country’s balance of payments, although deficits in the trade and current account were partly relieved by external inflows of capital such as foreign direct investment, official development assistance, and debt. This is evidence that the Vietnamese Government must adopt policies that can relieve the balance of payments and foster economic growth. Economic growth is usually the primary goal of an economy. To accelerate economic growth, it is worth billions to identify where the source of economic growth comes from and this has been the subject of much controversy among economists. On one side, economists such as Solow (1957), Romer (1986), Lucas (1988), and Krugman (1989) suggest that economic growth originates from the supply side, such as labor, capital, total factor productivity (TFP), and research and development (R&D). In their eyes, factors like supply and productivity play a major role in economic growth and the differences in economic growth among countries are due to differences in factors like supply and productivity. On the other side, economists such as Keynes (1936) and Thirlwall (1979) believe that demand will induce or constrain economic growth. Thirlwall (1979), a post-Keynesian economist, considered demand, especially international trade, as the principal factor accelerating or constraining growth. He suggests that the dominant constraint on demand is balance of payments in an open economy and his model, the so-called Balance of Payments Constrained Economic Growth model, stresses primarily on balance of payments as a constraint of economic growth. In the 1998-2001 period, for instance, after the Asian financial crisis, Vietnam experienced the phenomenon of supply exceeding demand, and the economy fell into stagnation. Growth was low and deflation and disinflation appeared. In confronting the situation, the government * Pham Sy An is a research fellow at the Vietnam Institute of Economics, Vietnam Academy of Social Sciences. The author would like to thank Dr. Vo Tri Thanh (CIEM), Nguyen Cao Duc (CAF), and James Donald (ANU) for their valuable comments on and support for the completion of this paper. 51 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Vietnam economic management review Number 1 Winter 2006 52 determined to stimulate domestic demand. Expansionary monetary policy and relaxed fiscal policy were implemented through credit expansion, low interest rates, and increases in government spending. This is clear evidence indicating that supply does not guarantee growth. Moreover, unemployment and capital waste prove that available production factors are underutilized in reaching potential output (Vietnam Institute of Economics 2002). In the case of supply exceeding demand, sources will be underused, while when demand exceeds supply, sources from other places will flow into the economy to satisfy demand. Until now, the relationship between the balance of payments and economic growth has not been recognized in Vietnam. Researchers have focused on looking for sources of economic growth in Vietnam under the form of production function (supply side), such as Tran Vo Hung Son and Chau Van Thanh (1998), Le Dang Doanh et al (2002), and Chu Quang Khoi (2003). Therefore, this paper is the first attempt at investigating economic growth in Vietnam from the demand side. The purpose of the paper is to answer the question of whether or not economic growth was constrained by the balance of payments in Vietnam during the 1990-2004 period. The paper applies Thirlwall’s balance of payments constraint model. The paper is structured into five sections and will be presented in two issues of the Review. Following the Introduction, Section 1 briefly reviews the picture of Vietnam’s economy and considers economic growth in relation to balance of payments during the 1990-2004 period. Section 2 develops the Balance of Payments Constrained Economic Growth Model. Section 3 estimates the growth constrained by balance of payments in the Vietnamese economy using quarterly data from the 1990-2004 period. Finally, Section 4 provides conclusions and policy implications. 1. Economic growth in Vietnam’s economy during the 1990-2004 period 1.1. Economic overview In the early 1980s, Vietnam’s economy was essentially a centrally-planned economy. The State controlled and intervened in the means of production as regards physical inputs, output, and prices. It established trade barriers, especially in foreign trade, and set dual foreign exchange rates as well as interest rates. The result was hyperinflation and slow economic growth. At the end of the 1980s, the Vietnamese Government launched the “doi moi” (economic renovation) process at the Sixth Congress of the Communist Party, with the nature of the process being a movement away from a centrally-planned economy towards a market-oriented economy. In the first three years of doi moi, changes took place in foreign investment policy, land policy, foreign trade policy, and banking policy. However, the failure of these efforts to stabilize the economy until 1989, as well as the expectation of aid from the former Soviet Union drying up, created the strongest pressure on reform. In March of 1989, Vietnam adopted a radical and comprehensive reform package aimed at stabilizing and opening up the economy, and enhancing freedom of choice for economic units in order to fundamentally change the economic management system in the country. The reform measures consisted of complete price liberalization, large devaluation and unification of the exchange rate, increases in interest rates to .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review positive levels in real terms, substantial reductions in subsidies to the SOE sector, encouragement of the private sector and a return to family farming on the basis of long-term leases. The 1989 reform package was considered the most successful since the basic conditions were created for the transformation into a market-oriented and open economy. After the 1989 reform package was implemented, hyperinflation fell rapidly from three digit levels to two digit levels, economic growth gradually improved, growth rates in exports and imports increased, and Vietnam became a rice exporting country after a long period of importing rice. These initial achievements encouraged the economic renovation program and were an initial stepping stone on the path to success during 1992-1997. The successful years began with the Seventh Party Congress in June 1991, which decided to continue doi moi by establishing and developing basic markets such as the labor market, capital and money markets, and an official foreign exchange market, and allowed prices and interest rates to be determined by market forces. In this period, economic growth was impressive, recording 8.77% on average. Inflation also fell quickly to single-digit levels, except in 1992 (17.5%). Inflation was thus under control, contributing significantly to macroeconomic stabilization and creating good conditions for high economic growth. Growth rates in exports and imports in these successful years were quite high and played an important part in promoting high economic growth. Moreover, high foreign investment flows and high savings to investment ratios were achievements in the period and this time marked a new phase of international integration, with Vietnam becoming a full-member of ASEAN in 1995. The Asian financial crisis suddenly broke out in Thailand and quickly spread its negative effects to other Asian countries such as Indonesia, Malaysia, and South Korea in 1997. Although the financial crisis did not impact directly on Vietnam’s economy, it had considerable indirect influence and curtailed the high economic growth of previous years. Growth fell unexpectedly to 5.76% in 1998 and continued to be low from 1998 to 2001. Inflation in the period after the Asian financial crisis was strangely low (a disinflation rate), and a minus figure in 2000 (a deflation rate of -0.6%). This phenomenon is a symbol of stagnation (low rates of inflation and growth). In this period, Vietnam became a member of APEC in 1998 and signed the Bilateral Trade Agreement with the US in 2000. Vietnam was also actively preparing the groundwork for its WTO accession. After the repercussions from the Asian financial crisis eased, economic growth was partly restored and inflation inched upwards (4% in 2002, 3% in 2003, and 9.5% in 2004). As existing difficulties pass, though, new ones arise. Although the Asian financial crisis was a good lesson for Vietnam, it is now being forced to reform its institutions and economy under pressure of international competitiveness if it wants its economy to integrate into the global economy with general laws and recording high growth as well as stability. 1.2. Economic Growth in the 1990-2004 Period In the 1990-2004 period, economic growth varied according to policy responses and changes in internal as well as external environments. As regards macroeconomic policies, the 1990-2004 period can be divided into two sub-periods: 1990-1995 and 1996-2004. The former was for 53 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Vietnam economic management review Number 1 Winter 2006 54 stabilizing macroeconomic policies and the latter for macroeconomic stabilization to record higher growth. However, for the purpose of this paper the 1990-2004 period will be divided into three sub-periods: 1990-1997, 1998-2001, and 2002-2004, to reflect the status of economic growth and the relationship between economic growth and balance of payment constraints. Business cycle of economic growth in the 1990-2004 period Traditional theories of the business cycle that considers cycles to be self-sustaining have been substituted by modern theories that attribute cyclical fluctuations to the cumulative effect of shocks and disturbances that continually affect the economy (Chatterjee 2000). Based on this shock-based view of the business cycle, we can examine the cumulative effect of shocks and disturbances that brought about fluctuations in economic growth during the 1990-2004 period. After implementing structural and orthodox reforms in 1989, natural and financial resources were allocated more efficiently, creating new incentives to produce goods and services. In the first two years of this period, economic growth averaged 5.45%. Then the cumulative effect of institutional reforms spurred economic growth to high levels during the 1992-1997 period, until the second shock came in the form of the Asian financial crisis in July 1997. While the first shock buffeted economic growth through the reallocation of natural and financial resources, the second shock impacted on economic growth through trade and investment. Prior to the Asian financial crisis, Vietnam’s economy was considered a small, open one in terms of the ratio of trade to GDP, and it connected with the outside through trade and investment. The shocks from the crisis influenced Vietnam’s economy through these channels, especially as its trade was concentrated primarily on Asian countries and substantial foreign direct investment came from Asian countries. There is a third channel that can affect Vietnam’s economic growth: the financial channel. Vietnam’s capital account in the balance of payments was controlled and restricted, so the shocks from the Asian financial crisis could not buffet Vietnam’s economy through portfolio investment and, consequently, the impact of the shocks on economic growth were not so severe. The effect of the shocks on economic growth then died out, Vietnam’s economy recovered and trade and FDI were also restored. Since 2002, economic growth has been high and other macroeconomic indicators such as the budget deficit and inflation have been stable. Period of economic stabilization and high growth: 1990-1997 The 1990-1997 period was characterized by high growth rates and stability in the socio- economic climate. High economic growth was only seen during 1992-1997, because in the first two years Vietnam had yet to break out of the “bad” period in the 1980s. During 1990-1991, Vietnam’s economy recovered its balance to motivate the following years of 1992-1997. Thus, the 1990-1997 period can be divided into two sub-periods: 1990-1991 and 1992-1997. For these reasons, the period can be called the period of economic stabilization and high growth. The need to confront potential macroeconomic crises was increasing, and in the beginning of 1989 the government decided to audaciously combine structural reform and orthodox stabilization measures. Structural reform brought about the liberalizing of most domestic prices, .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review while orthodox stabilization measures included raising interest rates, restraining credit expansion, and devaluating the exchange rate (Liunggren 1994). The result was an astonishing fall in inflation (from 393.8% in 1988 to 34.7% in 1989). However, by the end of 1989 high inflation had returned and continued upwards in 1990 and 1991 (67.1% and 67.5%, respectively). The balance of payments became serious and the collapse of the Soviet Union, one of Vietnam’s major donors and trading partners, resulted in grave shortages of necessary imported inputs for production, particularly petroleum and fertilizer. Growth was just 5.09% and 5.81% in 1990 and 1991, respectively (Liunggren 1994). All of these were synonymous with a latent return of macroeconomic instability. Reduction in aid from the Soviet Union, devaluation to reach the prevailing market exchange rate, and trade liberalization motivated exports and earned foreign exchange. This in turn increased imported inputs for production. Economic growth and the inflation rate surprisingly averaged 8.77% and 9.65%, respectively. Indeed, the stabilization period (1990-1991) was a springboard for the high growth seen in 1992-1997. In this period, the renovation process continued with several proposed macroeconomic reforms, such as (1) establishing and developing basic markets like labor, official exchange rate, capital and money markets; (2) allowing prices, interest rates, and exchange rates to be market determined; and (3) innovating administrative procedures. Growth rates for industry and construction as well as services were high during this period, especially industry and construction, at 13.25% per annum. Industry was considered the engine of growth. Moreover, the proportion of agriculture, forestry, and fishery fell compared to the proportion of industry, construction and services. Structural changes in the economy by increasing industrial and manufactured goods and decreasing agricultural products in the GDP showed a positive tendency in the process of development. Table 1: Growth rates of agriculture, industry, and services (%) (1994 prices) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 GDP 5.09 5.81 8.7 8.08 8.83 9.54 9.34 8.15 5.76 4.77 6.79 6.84 7.04 7.3 7.6 Agriculture, forestry and fishery 1 2.18 6.88 3.28 3.37 4.8 4.4 4.33 3.53 5.23 4.63 2.79 3.62 3.9 3.3 Industry and Construction 2.27 7.71 12.79 12.62 13.39 13.6 14.46 12.62 8.33 7.68 10.07 10.32 9.95 10.32 10.3 Services 10.19 7.38 7.58 8.64 9.56 9.83 8.8 7.14 5.08 2.25 5.32 6.13 6.31 6.32 7.3 Source: The General Statistics Office (GSO) (1996, 2000, 2005) The current account deficit as a percentage of GDP was relatively high compared with other periods and the trade deficit increased during the high growth period (see Figure 1). To speed up economic growth, a large trade deficit is inevitable as developing countries often import capital goods and intermediary inputs for production while exporting primary products and low value- added goods, and this was the case with Vietnam. 55 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Figure 1: GDP growth rate and Trade balance/GDP ratio -15.00 -10.00 -5.00 0.00 5.00 10.00 15.00 1990 1991 1992 1993 1994 1995 1996 1997 Trade balance/GDP (%) GDP growth rate (%) Source: GSO (1996, 2000, 2005) The coefficient of the correlation between economic growth and the ratio of the trade balance to GDP in the 1990-1997 period was -0.7. This was negative and quite high. It raises the question: what happens if there is a shortage of foreign exchange for imports? If this was to occur, imports would be reduced and domestic resources such as labor and machinery would be underemployed, and economic growth would be not be as high as usual. Fortunately, this did not happen in Vietnam, where the shortage of foreign exchange was supplemented mainly by inflows of FDI and partly by debt and net transfers. The coefficient of the correlation between GDP (%) and FDI was 0.53. This is moderate and positive. In the high growth period, FDI was high on average as well and the reduction in economic growth in 1997 compared with 1996 was accompanied by a simultaneous decline in FDI levels. Thus, FDI partly impacted intuitively on economic growth. In summary, in order to impulse economic growth, developing countries need to import the necessary inputs for production, but the level of exports required to earn foreign exchange aimed at financing imports are rarely met, so inflows of capital such as debt and foreign investment are very important. In the case of Vietnam, the scarcity of foreign exchange was fortunately supported by FDI, and partly by net transfers from remittances and debt. Period of economic slowdown after the Asian financial crisis: 1998-2001 The Asian financial crisis in 1997 put an end to Vietnam’s high economic growth. Growth fell suddenly from 8.15% in 1997 to 5.76% in 1998. Although Vietnam’s economy was not seriously affected directly like other Asian countries, it was hit indirectly. In the same period, Vietnam’s economy also suffered from recession, with symptoms of disinflation and low economic growth. In this period, growth and inflation averaged approximately 6% and 2.38%, respectively. The growth rate of industry, considered “the engine of growth”, unexpectedly declined. The Asian financial crisis resulted in a slowdown in the economies of influential countries and the unavoidable consequences were reductions in exports to and imports from these countries as well Vietnam economic management review Number 1 Winter 2006 56 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review as their investment to Vietnam. Furthermore, Vietnam’s currency devaluation was sufficient to enhance competitiveness compared with other countries in Asia. In this period, the trade balance and current account were slightly in deficit and even surplus in 1999-2001, but economic growth was low compared with the previous period. It is worth noting that narrowing the gap in the trade deficit can be a bad sign for developing countries, as they must import capital goods and intermediate inputs for domestic production and for export. Vietnam’s imports, as well as exports, fell significantly in the period, and unemployment concurrently increased. It is also important to understand that in order to reduce the trade deficit and still accelerate economic growth, it is not necessary to restrain imports because this will affect domestic production and exports. So the import of necessary inputs for production can continue to be encouraged, and exports will simultaneously expand. However, less developed countries usually export primary goods that have low prices and income elasticities in demand, while importing manufactured goods that have high prices and income elasticities in demand, so the trade deficit will widen unless reductions in economic growth drive decreases in imports. In the case of Vietnam, the period of high economic growth was also the period of a large trade deficit, as well as the current account deficit. The foreign exchange gap was filled by foreign financial resources. However, if the trade deficit continuously widened and the gap was not bridged by financial resource due to unsustainable warnings about Vietnam’s balance of payments in the international community, then there would not be the minimum amount of foreign exchange required to achieve targeted growth rates and economic growth would be constrained by the balance of payments. Of course, this has not happened in Vietnam’s case at present. It is a reminder of the important theory that shows that reductions in the trade deficit due to limiting or decreasing imports will constrain economic growth, as in 1998-2001 period. In the period, inflows of foreign investment into Vietnam decreased because foreign investors feared that instability in the regional socio-economic environment could affect Vietnam’s economy. In the 1990-1997 period, FDI increased strongly, at an average of 66.2%, but then fell by 12% in the 1998-2001 period despite the new Law on FDI in 1996 and its amendment on 9 June 2000 to improve the investment climate and attract inflows of foreign investment. Another economic event to occur in 1999-2001 was recession. This stemmed from supply exceeding demand in the domestic economy. The industrial structure leaned towards the import- substitution industry, which had a high effective rate of protection, and the growth of heavily protected industrial output has far outdistanced manufactured export growth (Dapice 2002). Confronting this situation, the government executed a demand-stimulus policy1 aimed at stimulating domestic demand. Although the government implemented several measures such as reducing tax rates and interest rates and expanding credit, domestic demand was not stimulated as expected and supply still exceeded demand because of poor management and lack of coordination in policies (See Le Xuan Sang (2003), Nguyen Van Tau (2001)). In summary, the low growth rate during the period was due to weaknesses in the domestic production structure and adverse external shocks. Although the trade balance, the current account 57 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Vietnam economic management review Number 1 Winter 2006 58 and the overall balance were in slight deficit and even surplus, economic growth was low. This period was a lively example of economic growth being constrained by demand in an open economy. Reduction in external demand (exports) resulted in a decrease in demand for imports of capital goods and an increase in unemployment in export activities. This in turn brought about reductions in income and, subsequently, domestic demand. To stimulate demand, the government must increase its expenditure but, of course, budget revenue was reduced due to the recession. This created a large budget deficit. Moreover, the gap between savings and investment also increased. This period quite clearly indicated that in an open economy, economic growth will be constrained by balance of payments and one gap in foreign exchange can bring about two other gaps, in savings and budget. Period of economic recovery and integration into the global economy: 2002-2004 The repercussions from the Asian financial crisis subsided but new difficulties arose. Economic growth began to recover and the inflation rate also increased slightly in 2002 and 2003, suddenly jumping to nearly 10% in 2004. Vietnam was also preparing to commit to WTO membership requirements, so it was urgent to quickly liberalize trade and finance as well as reform state-owned enterprises and administrative procedures. In this period, economic growth recovered and the trade balance was in deficit because the value of imports exceeded that of exports, although growth in exports was restored, at 25.3% and 28.9% in 2003 and 2004, respectively. Again, the opposite relationship between the trade balance and economic growth is clearly visible. A large trade deficit brings about high economic growth. And the foreign exchange gap is mainly financed through FDI inflows, and partly by net remittances and debt. In terms of export destinations, Vietnam still exports primarily to Asia, followed by the US and Europe. Vietnam imports chiefly from Asia. Lack of diversity in the origin of imports and the destination of exports can result in a high level of risk when adverse domestic shocks occur that negatively affect the balance of payments and then economic growth. In summary, economic growth recovered during this period and the trade deficit was high. Exports were mainly primary goods and land-based products. The import of machinery, equipment, materials, and fuel dominated total imports. The Vietnamese Government continued to reform the economy on the threshold of international economic integration, but many weaknesses were revealed and challenged the process of reform, especially the dubious nexus and kinship among government, state-owned enterprises, and commercial banks. In conclusion, the following remarks on Section 2 of this paper can be made. Firstly, economic growth went through three sub-periods: the periods of stabilization and acceleration (1990-1997), recovery (1998-2001), and integration (2002-2004). Each sub-period had specific characteristics and policy responses to adapt to changes in the internal and external environment. Secondly, the economic structure changed slowly. This was unexpected in the process of industrialization and improvements in competitive capacity. Thirdly, economic growth and the trade balance, as well as the current account, have an opposite relationship. And these deficits were filled by FDI, debt, and net transfer. Thus, economic growth would have been severely .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review constrained by balance of payments if no capital inflows bridged the gap in the trade balance and current account. 2. The balance of payments constrained growth model Thirlwall, a post-Keynesian economist, uses his model, the so-called Balance of Payments Constrained Economic Growth Model, similar to a dynamic version of Harrod’s (1933) Foreign Trade Multiplier, to explain the position of trade and the balance of payments in economic growth. Thirlwall (1979) reckoned that economic growth is ultimately constrained by the balance of payments. In his original model, international trade, particularly elasticities in demand for exports and imports in respect to price and income, plays a considerable role in explaining the differences in growth rates across countries. It is necessary, therefore, to revisit Thirlwall’s growth law and its other modified models. The initial equilibrium condition is given by: MEPXP fd = [1] MEPFXP fd =+ [2] MEPRXP fd =+ [3] mfxd IDPMEPIDPXP +=+ [4] DMEPFXP fd +=+ [5] In which X is the quantity of exports, is the price of exports in domestic currency, M is the quantity of imports, is the price of imports in foreign currency, E is the exchange rate measured as the domestic price of foreign currency, is the value of nominal net capital inflows measured in domestic currency, R is value of remittance, is the value of nominal revenues, in domestic currency, of invisible services related to production factors, the so-called “ dP fP F xIDP IDP revenues”, is the value of nominal expenditures, in domestic currency, of the invisible services related to production factors, the so-called “ mIDP IDP expenditures”, and D is debt service. Recognizing that a country can obstinately face a trade deficit, Thirlwall and Hussain (1982) developed the “Extended Model”, with the initial equilibrium condition [2]. Elliot and Rhodd (1999) highlighted the importance of debt servicing in the equilibrium condition, so the initial condition can be included in [5]. And other economists can modify the initial equilibrium condition in putting forward the importance of certain factors in the balance of payments such as capital flows, interest payments (Brid 2001), interest, dividends, and profits of the current account of balance of payments (Ferreira and Canuto 2003) in accordance with studied countries. In [3], due to the importance of remittance in filling the deficit in the trade balance in many developing countries, it is necessary to add remittance to the initial condition. Taking logarithm and rates of growth infers: 59 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT empxp fd ++=+ [6] empfxp fd ++=−++ )1()( θθ [7] emprxp fd ++=−++ )1()( γγ [8] mfxd idpempidpxp )1()()1()( ααωω −+++=−++ [9] dempfxp fd )1()()1()( ρρθθ −+++=−++ [10] In which, lower case letters represent rates of growth of the respective variables; θ and ( θ−1 ) represent the shares of exports and capital flows as a proportion of these receipts, γ and )1( γ− are the shares of exports and remittance as a proportion of these receipts, ω and ( ω−1 ) are the share of exports of goods and IDP revenues as a proportion of total receipts in the current account, α and ( α−1 ) show the share of imports and IDP expenditures as a proportion of total payments in current account, and ρ and )1( ρ− are the shares of imports and debt service on total expenditure. The normal multiplicative import and export demand functions have the forms: [ ] πψ YPEPaM df= [11] [ ] εη ZEPPbX fd= [12] In which a and b are constants, ψ is the price elasticity of demand for imports (ψ < 0), η is the price elasticity of demand for exports (η < 0), Y is domestic income, Z is the level of world income, π is the income elasticity of demand for imports, and ε is the income elasticity of demand for exports. Taking logarithm and rates of growth give: )()( ypepm df πψ +−+= [13] )()( zeppx fd εη +−−= [14] Substituting [13] and [14] into equations [6] to [10] and implementing some algebraic manipulations, we obtain: πεψη ]))(1[( zeppy fd +−−++= [15] πψθηθθθε )])(()1([ eppeppfzy fdfd −−++−−+−+= [16] πψγηγγγε )])(()1([ eppepprzy fdfd −−++−−+−+= [17] απαψωηαααωωωε )])(()1()1([ eppepidppidpzy fdfmdx −−++−−−−+−+= [18] ρπρψθηρρρθθθε )])(()1()1([ eppepdpfzy fdfd −−++−−−−+−+= [19] The real exchange rate, by assumption, does not change very much in the long run2 ( ) or the Marshall-Lerner condition is just satisfied (0=−− epp fd 1=+ηψ ) and zx ε≡ , equations [15] to [19] become: Vietnam economic management review Number 1 Winter 2006 60 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review πxy =0 [20] πθθ )])(1([1 dpfxy −−+= [21] πγγ )])(1([2 dprxy −−+= [22] απαωω )])(1())(1([3 dmdx pidppidpxy −−−−−+= [23] ρπρθθ )])(1())(1([4 dd pdpfxy −−−−−+= [24] Finally, we gain several forms of the Balance of Payments Constrained Economic Growth Model under different initial equilibrium conditions. According to McCombie and Thirlwall, the standard Thirlwall growth law explains well why developing countries grow at lower rates than developed countries and without the convergence predicted by classical economists. Differences in the rate of growth can be explained by differences in growth rates of export and income elasticity of demand for imports. The rate of growth of exports in turn depends on income elasticity of demand for exports and world income (1994, 1997). However, because of the different characteristics of each country in economic growth, other modified models also are applied to adapt to these characteristics. In developing countries, primary products for export are usually produced by applying labor- intensive methods and elasticities in demand for these goods in respect to prices and incomes are low, and this will result in deteriorating terms of trade for exported goods. Another factor in determining growth is that income elasticity of demand for imports is usually high. Developing countries import capital goods and intermediate inputs of production from advanced countries in support of growth so they tend to increase demand for imports more than one percent if the rate of growth increases one percent. Combining both low export growth rates (in value) and high income elasticity of demand for imports brings about low growth rates in developing countries compared with that in developed countries, and a divergence between them is inevitable. In post-Keynesian demand-led growth, export-led growth alone can lead to balance of payments constraints in the long run if income elasticity of imports is high and sufficient to offset increases in exports. Thus, the post-Keynesian tradition not only puts forward the importance of exports but also sheds light on the importance of income elasticity of demand for imports. From this standpoint, post-Keynesian and structuralist traditions support structural adjustment to avoid much external dependence that require government policies such as establishing trade barriers to protect domestic industries that have positive externalities on other socio-economic activities and have social costs exceeding private costs. 61 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT In conclusion, post-Keynesian and structuralist traditions build demand-led models implying that government policies are indispensable in stimulating demand (or economic growth) but should emphasize balance of payments constraints that can harm growth. 3. Estimating growth constrained by balance of payments in Vietnam during the 1990-2004 period 3.1 Model specification It is necessary to find elasticity of demand for imports with respect to income (π ) in the case of Vietnam during the 1990-2004 period. To find π , as mentioned previously, researchers usually use the multiplicative import function (as equation [11]) and after taking logarithm and rates of growth the function becomes [13]. For convenience, it can be reproduced as: )()( ypepm df πψ +−+= [25] Import growth depends firstly on growth in the terms of trade multiplied by the price elasticity of demand for imports, and secondly on domestic income (as a proxy for expenditure) multiplied by the income elasticity of demand for imports. Adding constant term α , error terms , and subscript t into above the equation, it turns out: tu ttdttftt uypepm ++−++= )()( πψα [26] Setting , Equation [26] can be written dttftt pepp −+≡ tttt uypm +++= )()( πψα [27] This equation will be estimated to find income elasticity of demand for imports π . Before estimating the coefficients of certain equations, it is necessary to go through some important procedures such as testing unit root for stationarity, finding the lag lengths of each variable, and testing Granger causality. Unit root test for stationarity In regression models for time-series data, we usually face the phenomenon of “spurious regression”. Thus, it is necessary to test stationarity before going further. The simplest and most widely used tests for unit roots were developed by Fuller (1976) and Dickey and Fuller (1979), the so-called Dickey-Fuller, or DF, tests; however, these are special cases of augmented Dickey-Fuller tests, or ADF tests. Thus, ADF tests will be employed for stationarity and results can be represented in the table below. Table 3: Unit Root Tests for Stationarity3 Vietnam economic management review Number 1 Winter 2006 62 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review Notation Description Properties tm Rate of growth of import Stationary ty Rate of growth of GDP Stationary tp Rate of growth of relative price Stationary Lag lengths of each variable Finding the lag lengths of each variable is very important in running regression with time series. The “Akaike information criterion” (AIC) and the “Schwarz information criterion” (SIC) will be applied to find the lag lengths of each variable including , , and . ty tm tp According to Davidson and MacKinnon (1993), “with quarterly data it may be wise to start with (in )”. So, we start with the model 4== qp ),( qpARDL ∑∑∑ = − = − = − ++++= 4 1 4 0 4 0 i titi i iti i itit umypm γπψα Then, finding the lag lengths of each variable by using the AIC and the SIC, we obtain ∑∑∑ = − = − = − ++++= 2 1 4 0 0 0 i titi i iti i itit umypm γπψα [28] The Granger test It is necessary to use the Granger causality test for four cases: (1) unidirectional causality from m to y, (2) unidirectional causality from y to m, (3) feedback or bilateral causality, and (4) independence. t i iti i itit vmym 1 4 1 2 4 1 10 +++= ∑∑ = − = − ββα [29] t i iti i itit vymy 2 4 1 4 4 1 30 +++= ∑∑ = − = − ββα [30] The lag length of each variable can be chosen so that the autocorrelation problem is rejected (Vogelvang 2005). For that condition, the lag length of each variable is revealed as four. After running [29] and [30] in the absence of autocorrelation, it can be concluded that unidirectional causality is from to 3y m . Thus, coefficients of equation [28] can be estimated by the method of ordinary least squares. (To be continued in the next issue) Notes 1. The demand stimulus policy had been implemented since mid-1999 with an emphasis on 63 .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Vietnam economic management review Number 1 Winter 2006 64 expansionary fiscal policy and monetary policy. 2. Krugman (1989) observed that there are no trends in real exchange rates in the long run, or, in other words, he asserts that relative prices should be taken as constant in the long run. According to McCombie and Thirlwall (1994), there are three reasons why relative prices remain unchanged in the long run: (1) domestic price changes mirror exchange rate changes; (2) highly competitive markets, and (3) oligopolistic market structures. 3. See Appendix 1 for more details. References ƒ Brid J. C. M (2001), “Capital Flows, Interest Payments and the Balance-of Payments Constrained Growth Model: a Theoretical and an Empirical Analysis”, Prepared for the Conference an Old and New Growth theories: an Assessment Pisa, Italy, October 5-7. ƒ Chatterjee, S. (2000), “From Cycles to Shocks: Progress in Business Cycle Theory”, Federal Reserve Bank of Philadelphia, March/April 2000. ƒ Chu Quang Khoi (2003), Sources of Economic Growth, the Case of Vietnam in the Period 1986-2001, Thesis for Masters Degree, MDE Hanoi. ƒ Dapice, D.O. (2002), “Success and Failure: Choosing the Right Path to Export-led Growth, Center for Business and Government,” Cambridge. ƒ Davidson, R. and Mackinnon, J. G. (1993), Estimation and Inference in Econometrics, Oxford University Press, Oxford. ƒ Dickey, D. A., and Fuller, W. A. (1979), "Distribution of the estimators for autoregressive time series with a unit root," Journal of the American Statistical Association, 74. ƒ Elliot, D.R. and Rhodd, R. (1999), “Explaining Growth Rate Differences in Highly Indebted Countries: an Extension to Thirlwall and Hussain,” Applied Economics 31. ƒ Ferreira, A. L. and Canuto, O., (2003), “Thirlwall's Law and Foreign Capital in Brazil,” Enero-Febrero de 2003. ƒ Fuller, W. A. (1976), Introduction to Statistical Time Series, John Wiley & Sons, New York. ƒ General Statistics Office (1996), Statistical Yearbook 1995 of Vietnam, Statistical Publishing House, Hanoi. ƒ General Statistics Office (2000), Statistical Yearbook 1999 of Vietnam, Statistical Publishing House, Hanoi. ƒ General Statistics Office (2005), Statistical Yearbook 2004 of Vietnam, Statistical Publishing House, Hanoi.Harrod, R. (1933), International Economics, Cambridge University Press, Cambridge. ƒ Keyness, J. M. (1936), The General Theory of Employment, Interest and Money, Macmillan, London. ƒ Krugman, P. (1989), “Differences in Income Elasticities and Trends in Real Exchange Rates,” European Economic Review 33, 5: 1031-46. .VEMR. Economic research ECONOMIC GROWTH AND BALANCE OF PAYMENTS CONSTRAINT Number 1 Winter 2006 Vietnam economic management review ƒ Le Dang Doanh et al (2002), Explaining Growth in Vietnam, CIEM, Paper for Development Network, Hanoi. ƒ Le Xuan Sang (2003), “Demand-stimulating Policies after 4 Years: Success, Limitations, and Policy Implications,” Economic Studies Review No 301-302. ƒ Liunggren, B. (1994), Challenges on Reform Road in Indochina, National Political Publisher, Ha noi. ƒ Lucas, R. E. (1988), “On the Mechanics of Economic Development,” Journal of Monetary Economics 22, 1: 3-42 ƒ McCombie, J. and Thirlwall, A. P. (1994), Economic Growth and the Balance of Payments Constraint, Macmillan, London. ƒ McCombie, J. and Thirlwall, A.P. (1997), "The dynamic Harrod Foreign Trade Multiplier and the demand oriented approach to economic growth: an evaluation", International Review of Applied Economics, January. ƒ Nguyen Thi Hong (1999), The Sustainability of the Current Account Deficit in Vietnam (1989- 1998), Thesis of Master Degree, MDE Hanoi. ƒ Nguyen Van Tau (2002), Effectiveness of demand-stimulating policies: lessons from other countries and experiences of Vietnam in period 1999-2001, MA thesis, Vietnam-Netherlands Project for Master Program in Development Economics, NEU, Hanoi. ƒ Romer, R. M. (1986) “Increasing Returns and Long Run Growth,” Journal of Political Economy 94, 5: 1002-37. ƒ Stiglitz, J. E. (2002), Keynesian Economics and Critique of First Fundamental Theorem of Welfare Economics, in Market Failure or Success – the New Debate, edited by Tyler Cowen and Eric Crumpton, Edward Elgar, Cheltenham, UK: 41-65. ƒ Thirlwall, A. P. and Hussain, M. N. (1982), "The Balance of Payments Constraint, Capital Flows and Growth Rate Differences Between Developing Countries", Oxford Economic Paper. ƒ Thirlwall, A.P. (1979) “The Balance of Payments Constrain as an Explanation of International Growth Rate Differences,” Banca Nazionale del Lavoro Quarterly Review 128, 791: 45-53. ƒ Tran Vo Hung Son and Chau Van Thanh (1998), “Analysis of the Sources of Economic Growth of Vietnam,” CAS Discussion Paper No. 21, CAS-CIMDA. ƒ Vietnam Institute of Economics (2002), Vietnam Economy 2002: New Policy Thinking for New Stage of Development (Unpublished), Ha Noi. ƒ Vogelvang, B. (2005), Econometrics - Theory and Applications with EViews, Pearson Education Limited, the United Kingdom. 65