Agricultural Sector Policies and Poverty in the Philippines: A CGE Analysis

Transcription

1 Modeling and Policy Impact Analysis (MPIA) Network Agricultural Sector Policies and Poverty in the Philippines: A CGE Analysis Erwin Corong Philippines A paper presented during the 4th PEP Research Network General Meeting, June 13-17, 2005, Colombo, Sri Lanka.

2 Agriculture Sector Policies and Poverty in the Philippines: A CGE Analysis 1 Caesar B. Cororaton and Erwin L. Corong 2 Very Preliminary Final Report Abstract The Philippines has undertaken substantial trade reform since the 1980s with little knowledge of the likely poverty impacts. A detailed CGE model is used in the analysis to estimate and explain these impacts. The actual tariff reduction that occurred between 1994 and 2000 is found to increase poverty, especially among the rural households. This results from the contraction of the agricultural sector as lower import prices induce consumers to substitute towards it. Keywords: JEL: Agriculture, International Trade, Poverty, Computable General Equilibrium, Microsimulation, Philippines D58, E27, F13, F14, I32, 013, 015, 024, 053, Q10 1 This work was carried out with the aid of a grant from Poverty and Economic Policy (PEP) Research Network, financed by the International Development Research Center (IDRC). 2 International Food Policy Research Institute (c.cororaton@cgiar.org) and De La Salle University (coronge@dlsu.edu.ph). We are grateful to Ismael Fofana, John Cockburn and Bernard Decaluwe. The usual disclaimer applies. 1

3 Summary The Philippines has undertaken substantial trade reform since the 1980s with little knowledge of the likely poverty impacts. This study employed a detailed CGE microsimulation model in order to assess the economic and poverty effects of the actual tariff reduction that occurred between 1994 and Simulation results show that the price of agricultural imports fall the most since its initial import-weighted average tariffs are higher than industry. Agriculture sector contracts as the substantial decline in import prices induce consumers to substitute towards it. On the other hand, the lower price of imported inputs reduces the local cost of production benefiting the outward-oriented-import-dependent industrial sector. Overall, a freer trade appears to benefit the service sector as wholesale and trading sub-sector expands. National poverty headcount marginally decreases. However, both the depth (poverty gap) and severity (squared poverty gap) worsens implying that the poorest of the poor become much more immerse in poverty. Poverty stricken rural areas suffer the most due to a decline in average income (as a result of agricultural contraction), compounded by an increase in average consumption prices. In contrast, poverty situation improves in the National Capital Region (NCR) or Metro Manila because of its proximity to major industries and the expanding service sector. In conclusion, tariff reduction appears likely to increase poverty especially in the rural areas and among the rural low educated. The tariff reduction between 1994 and 2000 hurts poor the most. This suggests that the government should consider a regional complementary policy aimed at addressing regional imbalances. In addition, a human capital development program must be instituted as simulation results point out that education and skill are the best ally against poverty. 2

4 Introduction The Philippines has undertaken substantial trade reform since the 1980s. Tariff rates have been reduced, tariff structure simplified, and quantitative restrictions tariffied. However, the impact of all these developments on agriculture and on the poor is not very clear, thus becoming the subject of an intense policy debate. What is the impact of tariff reduction on the agricultural sector, and the entire Philippines? Is trade liberalization favorable or harmful for the poor? What alternative or accompanying policies may be used in order to ensure a more equitable distribution of the gains from freer trade? What are the channels through which these changes are most likely to affect the poor? This study employs a 35-sector CGE microsimulation model calibrated to Philippine data to analyze the impacts of tariff reduction in the Philippines. The model consists of 24,797 Filipino households based on the 1994 Family Income and Expenditure Survey. Given the macro and micro nature of the study, specific focus is given on macroeconomic impacts, resource allocation, factor demands and factor prices, household income, consumer prices and poverty. Background on Philippine Agriculture The agriculture sector employs about 35 percent of the labor force and accounts for roughly 20 percent of GDP. If linkages with all other sectors such as agriculture related processing and non-farm agriculture inputs are accounted for, agriculture contributes roughly 40 percent of GDP and employs two thirds of the labor force (David, 1997). The sector s growth performance is generally weak because of low productivity growth. Growth decelerated from an annual average of 6.7 percent in the 1970s to 1.1 percent in the first half of the 1980s (Table 1). Although the second half of the 1980s saw some recovery, agriculture again lost steam in the 1990s to register an annual growth rate of two percent. Agriculture was the most promising sector in the 1970s. It grew rapidly mainly due to the Green revolution. However, the inherent policy bias against agriculture, coupled with the collapse in world commodity prices, halted this growth momentum. David (2003) concludes that the negative impact of government's anti-agriculture policy bias was greater than that of declining world commodity prices. The policy bias towards import substitution and against agriculture and exports until the 1970s led to market distortions that promoted rent seeking activities and distorted economic incentives against investments in agriculture. In particular these biases were: (a) the imposition of agricultural export taxes to generate government revenue; (b) policy of maintaining an over-valued exchange rate that resulted to negative protection rates in agriculture thereby reducing the rates of return to investments; and (c) government intervention in agriculture that created government marketing agencies that siphoned off the gains from trade, hence creating to rent-seeking activities. 3

5 Table 1: Growth Rates of Agriculture, Fishery and Forestry Agriculture Crops - Palay Corn Sugarcane Coconut Banana Other Crops Livestock Poultry Agricultural Services Fishery Forestry Agriculture, Fishery and Forestry Source: National Statistical and Coordination Board In spite all these, the agriculture sector continued to serve as the country s backbone in the 1970s. It was a net exporter, contributing two-thirds of total exports and representing only 20 percent of total imports, thus providing the foreign exchange needed to support the import dependent manufacturing sector (Intal and Power 1990). However, the 1990s saw a clear change in agricultural trade patterns as exports stagnated and imports increased dramatically to the point that the Philippines became a net importer of agricultural goods. David (2003) attributes this evolution to the country s fading comparative advantage and low productivity levels in agriculture. Essentially, this can be traced to primary agricultural goods where exports have gone from 1400 percent of imports in 1970 to 50 percent in Table 2 Contribution of Agriculture to GDP (%) Agriculture, Fishery, and Forestry a. Agriculture Palay Corn Coconut including Copra Sugarcane Banana Other Crops Livestock Poultry Agricultural activities & Services b. Fishery c. Forestry Industry Services GROSS DOMESTIC PRODUCT

6 Source: Economic and Social Statistics Office, National Statistical and Coordination Board The combined impact of low productivity growth and declining comparative advantage took its toll on agriculture s contribution to GDP. Table 2 shows that the entire agriculture sector (including fishery, and forestry) contribution to GDP declined by 7 percent from 21.7 in 1993 to 14.5 in Notably, the contribution of Palay, Corn, and Sugar decreased in the same period as well. Table 3 shows the structure of agriculture crops in the last 10 years. More than 50 percent of agricultural area is planted with Palay and Corn, while sugar occupies about 3 percent. In terms of the value of production Palay and Corn contributes more than 40 percent. Palay alone contributes 38 percent of the total value of agricultural production in 2002 Table 3. Agriculture Production (percentage distribution, %) /p Area Quantity Value Area Quantity Value Area Quantity Value A. Cereals Palay Corn B. Major Crops Coconut Sugarcane Banana Pineapple Mango Other major crops Other Crops Total , ,630 12, , Source: Philippine Statistical Yearbook /p: preliminary Philippine Trade Reform The first phase of the trade reform program (TRP) started in the early 1980s with three major components: (a) the tariff reduction; (b) the import liberalization program (ILP); and (c) the complimentary realignment of the indirect taxes. Maximum tariff rates were reduced from 100 to 50 percent. During the period sales taxes on imports and locally produced goods were equalized. The mark up applied on the value of imports (for sales tax valuation) was also reduced and eventually eliminated. The implementation of ILP however was suspended in the mid 1980s because of a balance of payments crisis. In fact, some of the items that were deregulated earlier were re regulated during the period. When the Aquino government took over the administration in 1986 the TRP of the early 1980s was resumed, resulting in the 5

7 reduction of the number of regulated items from 1,802 in 1985 to 609 in Export taxes on all products except logs were also abolished. In 1991 the government launched TRP II, which was an extension of the previous program that realigned tariff rates over a five year period. The realignment involved the narrowing of the tariff rates through a series of reductions of the number of commodity lines with high tariffs, and an increase in the commodity lines with low tariffs. In particular, the program was aimed at clustering of tariff rates within the percent range by This resulted in a near equalization of protection for agriculture and manufacturing by the start of the 1990s, reinforced by the introduction of protection to "sensitive" agricultural products. Despite the programmed narrowing of the tariff rates, about ten percent of the total number of commodity lines were still subjected to a 0-5 percent tariff or a 50 percent tariff by the end of the program in In 1992, a program of converting quantitative restrictions (QRs) into tariff equivalents was initiated. In the first stage, QRs of 153 commodities were converted into tariff equivalent rates. In a number of cases, tariff rates were raised over 100 percent, especially during the initial years of the conversion. However, a built in program for reducing tariff rates over a five year period was also put into effect. QRs were removed for a further 286 commodities in the succeeding stage. At the end of 1992 only 164 commodities were subjected to QRs. There were some policy reversals along the way though. In 1993, QRs were re-introduced for 93 items, largely as a result of the Magna Carta for Small Farmers in In 1994, the government started implementing TRP III at the same time as it was admitted to the WTO. Tariff rates were successively reduced on: capital equipment and machinery (January 1, 1994); textiles, garments, and chemical inputs (September 30, 1994); 4,142 manufacturing goods (July 22, 1995) and non-sensitive components of the agricultural sector (January 1, 1996). Through these programs, the number of tariff tiers was reduced, as were the maximum tariff rates. In particular, the overall program was aimed at establishing a four-tier tariff schedule: three percent for raw materials and capital equipment that are not available locally; ten percent for raw materials and capital equipment that are available from local sources; 20 percent for intermediate goods; and 30 percent for finished goods. This further reduced the anti-agriculture tariff bias with the EPR for agriculture exceeding that of industry for the first time in 1995 (Habito 1999). Indeed, EPRs in agriculture and industry went from nine and 44 percent, respectively, in 1979 to 25 and 20 percent in 1999, and to 24 and 15 percent by the year 2000 (Bautista, Power and Associates 1979; Manasan and Pineda 1999; Habito 2002). By 1998, TRP IV was undertaken to recalibrate the tariff rate schedules implemented under previous rounds of TRPs. This was primarily due to the result of a tariff review process that evaluated the pace of tariff reduction in line with the competitiveness of the local industry. Initially, EO 465 was implemented on January 22, 1998 to adjust the tariff rate schedules of twenty-two industries that were identified as globally competitive (Tariff Commission, 2004). Subsequently, EO 486 was passed on July 10, 1998 to amend the tariff schedule of residual items as well as reduce the number of tariff lines subject to quota from 170 under TRP III to 144 under TRP IV. In line with TRP IV 6

8 moreover, EO 334, which took effect on January 1, 2001 provided for an amended tariff schedule on all product lines (except sensitive agricultural products) within the years 2001 to By and large, EO 334 provides for a tariff band of 0 to 5% starting the year Government Revenue Between 1994 and 2000, the overall weighted nominal tariff declined by 66.9 percent (Table 4). The decline in the industry tariff (-65.3 percent) was much greater than in agriculture (-48.8 percent). The largest drop in tariff rates was in mining (-88.9 percent) while the smallest decline was in "other agriculture" (-19.9 percent). In terms of the average sectoral tariff rate level in 2000, food manufacturing still had the highest rate of 16.6 percent, whereas other agriculture had the lowest tariff rate of 0.2 percent. Table 4: Nominal Tariff Rates Percent change Crops Livestock Fishing Other Agriculture AGRICULTURE Mining Food manufacturing Non-food manufacturing INDUSTRY TOTAL Sources of data for calculation: Various issues of Foreign Trade Statistics, and Manasan and Querubin (1997). Table 5. Sources of Revenue and Balances of National Government * Tax Revenue Taxes on Net Income and Profits Excise and Sales Taxes Import Duties and Other Import taxes Other Taxes Non_Tax Revenue Grants Total Total Revenue (P Billion) Total Expenditure (P Billion) (Deficit)/Surplus (P Billion) (37.2) 12.1 (134.7) (187.1) (Deficit)/Surplus (% of GNP) (3.5) 0.6 (3.9) (3.6) Source: Selected Philippine Economic Indicators (BSP) and Yearly Statistics (Bureau of Treasury) 7

9 Revenue from import tariff is one of the major sources of government funds. In 1990, the share of revenue from import duties and taxes to the total revenue was 26.4 percent (Table 5). It increased marginally to 27.7 percent in 1995, but dropped sharply to 17.5% percent in 2004, largely due to the tariff reduction program. The reduction in the share of tariff revenue was compensated primarily by an increase in the share of income and profit taxes from 27.3 percent in 1990 to 39.8 percent in Conversely, the share of excise and sales taxes dropped from 27.2 percent in 1990 to 23.4 percent in 1995, recovered slightly to 28.1 percent in 2000, but faltered once again with a 24 percent share in Poverty Profile Figure 1 presents the evolution of the poverty headcount index and the Gini coefficient from 1985 to The poverty headcount index dropped continuously from 49.2 percent in 1985 to 36.9 percent in 1997, but then rebounded to 39.5 percent in 2000 as a result of the 1998 El Nino and the Asian Crisis. El Nino resulted in a 30 percent contraction in agriculture, the greatest drop in more than 30 years. On the hand, income inequality has steadily increased over this period, as the Gini coefficient climbs from 0.42 in 1985 to 0.51 in Figure 1: Income Distribution and Poverty: The Philippines ( ) Gini Headcount Headcount Gini Source: FIES 1985, 1988, 1991, 1994, 1997, Table 6 shows the poverty indices at the base. In 1994, about 45 percent of the population of 67 million was below the poverty threshold. NCR or Metro Manila, where majority of the industries are located had the lowest poverty with headcount, gap and severity of 10.40, 2.01, and 0.60 respectively. While, rural areas have the highest poverty level. 8

10 Three important facts can be inferred from table 7. First, poverty is influenced by spatial factors. Rural households, which represent roughly half of the population, are substantially poorer than urban households 3. In the same vein, households residing in NCR or Metro Manila are less prone to poverty compared to other urban dwellers. Second, the degree of poverty depends on human capital. Household heads with at least a high school diploma (skilled workers), regardless of gender are less susceptible to poverty having better opportunities and options for employment. Third, Gender. Maleheaded households are relatively worse off, and much more vulnerable to poverty than their female counterparts.!" #$ % %%! "" "" # $% % &! % % &&!!' % % &! &&! 3 Although Balisacan (2003) suggest that this arises from the systematic differences in levels of human capital between low and high income groups within a geographic area which translate to considerable difference in earning opportunities, thus disparity in income and human achievement is the major problem not disparity between regions 9

11 The Model Basic Structure. The model was calibrated to the 1994 SAM of the Philippine Economy and has 35 production sectors composed of 13 Agriculture, Fishing and Forestry; 19 Industry; and 3 Services--which include government services. It distinguishes capital, land, and four types of labor inputs namely: skilled (high school diploma) and unskilled agricultural labor, as well as skilled and unskilled production workers. The production structure of the model assuming constant returns to scale is presented in figure 2. Gross output is determined through a linear aggregation of intermediate inputs and Value added. Intermediate input is determined using a fixed Leontief coefficient, whereas value added is a Cobb-Douglas (CD) function of labor and capital. Sectoral capital is fixed. Figure 2. Production Structure Leontief Output Intermediate Inputs Land-Labor-Land CD Land CD Labor Capital Skilled Unskilled The salient feature of the model stems from the way land and labor inputs are treated in agriculture. Land together with skilled and unskilled agriculture labor are agriculture specific, hence can only be employed in the sector. However, production workers employed in agriculture can move in and out of the sector. On the other hand, non-agricultural sectors, except government (which only utilize skilled production labor) employ skilled and unskilled production labor. Figure 3 illustrates the basic price relationships in the model. Output price, px, affects export price, pe, and local prices, pl. Indirect taxes are added to the local price to determine domestic prices, pd, which together with import price, pm, will determine the composite price, pq. The composite price is the price paid by the consumers. Import price, pm, is in domestic currency, which is affected by the world price of imports, exchange rate, er, tariff rate, tm, and indirect tax rate, itx. All prices adjust to clear the factor and product markets. An Armington-CES (constant elasticity substitution) function allocates the demand between local and imported goods, while a CET (constant elasticity of transformation) function determines the allocation of domestic production between 10

12 export supply and local sales. The demand side assumes cost minimization while the supply side assumes profit maximization. In turn, both their first order conditions generate the necessary: import and domestic demand functions, as well as the necessary supply and input demand functions. Figure 3: Basic Price Relationship in the Model Output price (px) Export price (pe) Local price (pl) + Indirect taxes (itx) Import price (pm) Domestic price (pd) Composite price (pq) The model integrates the entire Family Income and Expenditure Survey (FIES) with 24,797 households. Consumer demand is derived from CD utility functions. Model Closure. Nominal government consumption is equal to exogenous real government consumption multiplied by its (endogenous) price. Fixing real government spending neutralizes any possible welfare/poverty effects of variations in government spending. Total government income is held fixed. Any reduction in government income from tariff reduction is compensated endogenously by the introduction of an additional uniform sales or income tax. Thus, the government's budget balance (public savings) is endogenously determined, although the only variations are due to changes in the nominal price of government consumption. Total nominal investment is equal to exogenous total real investment multiplied by its price. Total real investment is held fixed in order to abstract from inter-temporal welfare/poverty effects. The price of total real investment is endogenous. The current account balance (foreign savings) is held fixed and the nominal exchange rate is the model's numéraire. The foreign trade sector is effectively cleared by changes in the real exchange rate, which is the ratio of the nominal exchange rate multiplied by the world export prices, divided by the domestic price index. The propensities to save of the various household groups in the model adjust proportionately to accommodate the fixed total real investment assumption. This is done through a factor in the household saving function that adjusts endogenously. Economic Structure. Table 7 presents the economic structure based on the SAM. The sectoral CES and CET elasticities in the model are derived as one-half of the Armington elasticities for the Philippines in GTAP (Hertel et al 2004). In general, the pattern of trade points out the dominance of the industrial sector. Indeed, it accounts for 11

13 roughly 60 percent of exports outperforming the services and agricultural sectors with 34 and 6 percent share respectively. Nonetheless, total agriculture exports contributed about 15 percent when all agricultural related food processing are accounted for. The principal industrial exports are semi-conductor and textile and garments followed by all processed food exports with a combined 9 percent share. Furthermore, semi-conductor, Coconut processing, Banana, textile and garments, and mining are the most export intensive sectors. Similarly, 99 percent of total imports accrue to the industrial sector with the remainder going to the agricultural sector. This enormous share stems from the low valued-added-import-intensive-assembly type operation in the manufacturing sector (particularly in the semi-conductor and textile and garment sub-sectors). Motor vehicles 4, has the highest import share followed by semi-conductors. The highly importintensive sectors are mining (72.03 percent; mainly due to crude oil imports), semiconductors, machinery, and fertilizer 5. In terms of value added, the agricultural sector generally has highest ratio compared to industry, although its contribution to the overall value added is relatively small. Agriculture contributes about 20 percent of domestic value added (GDP), compared to 31.5 for industry and 48.5 for services. Labor intensity is uniformly higher in the agricultural sectors, with the exception of fishing and "other livestock" 4 All vehicles are assembled using Completely Knocked Down (CKD) imported parts 5 The Philippines does not produce all items in the semi-conductor sector, but instead imports these items. For example, it does not have the facilities to produce wafers (motherboards) and monitors, which are major parts of computers. Domestic production focuses on hard disks, disk drives, processors, and some chips. Thus, while there is substantial domestic production and exports in the semi-conductor sector, there are also substantial imports. 12

14 SECTORS &' Foreign Trade Production Trade Elasticities*** Exports,% * Imports,% * VA Share Lab- Cap Armington CET Share Intensities Share Intensities (VA/X) (VAi / Ratio i VA) *, ** Irrigated Palay Non_irrigated Palay Corn Banana Fruits Coconut Sugarcane Other agricultural crops Hog Chicken_egg and other poultry products Other livestock Fishing Other Agriculture AGRICULTURE Mining Meat Processing Canning and preserving of fruits and vegetables Fish canning and processing Coconut processing Rice and corn milling Sugar milling and refining Beverages_sugar_confectionery and related products Other food manufacturing Textile and garments Wood_paper products Fertilizer Other chemicals Petroleum_related products Metal and related products Semi_conductors and other electronic products Motor vehicles and other machineries Other manufacturing Construction and utilities INDUSTRY Wholesale trade Other service Government services SERVICES TOTAL * Based on the 1994 SAM ** Lab-Cap is labor-capital ratio *** Based on GTAP (Hertel, 2004) 13

15 Household Income Table 8 presents the sources of household income. Income generated from labor was the major source for the entire population with 45.5 percent followed by 35.7 percent from capital. Income earned by laborers in the industrial sector and returns to capital in the service sector has the highest share within the labor and capital income block. Table 8. Sources of Household Income (At the Base) ( )* * (* ) ( %%+, %%#, %+, %#, %% % + ( -.. Policy Experiments Two policy experiments were undertaken in the study. Exp_1 Exp_2 Actual tariff reduction that occurred between 1994 and 2000 (based on table 5). Full tariff elimination Both experiments entailed the use of a compensatory tax on direct income applied uniformly to all households who are paying income taxes. Thus, those who are tax exempt are not burdened by this tax. The justification for this is the consistently increasing share of taxes on net income and profits discussed in Table 6. The additional tax is imposed in the following manner ( [ ]) dyh = y 1 dtxr 1+ ntaxr h h h where dyh h is disposable of household h, y h is income, dtxr h is direct income tax rate at the base, and ntaxr is the endogenously determined compensatory tax. This is implemented only for households with dtxr h > 0, 14

16 Simulation Results /* ' SIM_1 SIM_2 Change in overall nominal tariff rate, % Change in Prices, %: Import prices in local currency Consumer prices Local cost of production Real exchange rate change, % Change in import volume, % Change in export volume, % Change in domestic production for local sales, % Change in consumption (composite) goods, % Change in overall output, % Macro Effects: The macro effects for both policy experiments are identical with relatively higher values under the full tariff elimination scenario. The actual tariff reduction (Sim_1) leads to an 8.02 percent decrease in local import prices. Consumer prices decrease by 2.25 percent resulting to a 0.30 percent increase in consumption. Similarly, the decrease in tariff reduces the price of imported inputs, leading to a 3.67 percent decline in domestic cost of production. Real exchange rate depreciates as a result (by 4.24 percent). With this, producers reallocate towards the international market as domestic sales go down by 2.22 percent while exports increase by 9.55 percent. However, imports outpace exports owing from a much deeper decline in import prices. Effectively, import crowds out locally produced goods as consumers substitute towards it. Hence, overall output decreases though marginally by 0.10 percent. 15

17 Table 10a. Effects on Prices and Volumes (SIM_1) Price Changes (%) Volume Changes (%) SECTORS δpm i δpd i δpq i δpx i δpl i δm i δe i δd i δq i δx i Irrigated Palay Non_irrigated Palay Corn Banana Fruits Coconut Sugarcane Other agricultural crops Hog Chicken_egg and other poultry products Other livestock Fishing Other Agriculture AGRICULTURE Mining Meat Processing Canning and preserving of fruits and vegetables Fish canning and processing Coconut processing Rice and corn milling Sugar milling and refining Beverages_sugar_confectionery and related products Other food manufacturing Textile and garments Wood_paper products Fertilizer Other chemicals Petroleum_related products Metal and related products Semi_conductors and other electronic products Motor vehicles and other machineries Other manufacturing Construction and utilities INDUSTRY Wholesale trade Other service Government services SERVICES Where qi : composite commodity pxi : output prices mi : imports xi : total output pli : local prices ei : exports pmi : import (local) prices pqi : composite commodity prices di : domestic sales pdi : domestic prices δ : change 16

18 Table 10b. Effects on Prices and Volumes (SIM_2) Price Changes (%) Volume Changes (%) SECTORS δpm i δpd i δpq i δpx i δpl i δm i δe i δd i δq i δx i Irrigated Palay Non_irrigated Palay Corn Banana Fruits Coconut Sugarcane Other agricultural crops Hog Chicken_egg and other poultry products Other livestock Fishing Other Agriculture AGRICULTURE Mining Meat Processing Canning and preserving of fruits and vegetables Fish canning and processing Coconut processing Rice and corn milling Sugar milling and refining Beverages_sugar_confectionery and related products Other food manufacturing Textile and garments Wood_paper products Fertilizer Other chemicals Petroleum_related products Metal and related products Semi_conductors and other electronic products Motor vehicles and other machineries Other manufacturing Construction and utilities INDUSTRY Wholesale trade Other service Government services SERVICES Where qi : composite commodity pxi : output prices mi : imports xi : total output pli : local prices ei : exports pmi : import (local) prices pqi : composite commodity prices di : domestic sales pdi : domestic prices δ : change 17

19 Sectoral Trade, Output and Consumption: It seems that tariff reduction results to a reallocation from the inward oriented agricultural sector into the outward oriented industrial sector, and the services sector. Table 10a and 10b presents the sectoral impacts for SIM_1 and SIM2 respectively. Agriculture: The decline in local agriculture import prices induces consumers to substitute towards it. In particular, irrigated palay and fruit imports increase by and percent 6. Overall, total agricultural imports rise by resulting to a 0.51 percent dip in agricultural output. In spite of this, Banana stands to benefit from tariff reduction as both its output and exports expand by 9.32 and percent. Industry: Tariff reduction generally favors the import-dependent-outwardoriented industrial sector as the cost of inputs go down resulting to a surge in import demand. Nevertheless, export increases by percent arising from a decline in local production cost. Semi-conductor, textile and garments, and coconut processing emerge as the biggest gainers realizing a hefty increase in both output and export volumes. However, the entire industrial output contracts by 0.62 percent. Service: The service sector appears likely to benefit the most from tariff reduction. The decline in composite prices for both agriculture and industry increases trading activity in the wholesale and retail trade sub-sector. With this, the entire sector s output expands leading to an increase in value added demand by percent (Table 12a). This suggests that resource reallocation will flow from the contracting agriculture and industrial sectors towards the expanding service sector. Factor Remuneration: Table 11a and 11b presents the factor market impacts. Returns to factor decrease as return to capital and over-all wage decreases by 1.9 and 2.1 percent respectively. The much higher decline in wage relative to capital triggers a substitution towards labor, hence increasing labor utilization by percent. The contraction in both agriculture and industrial sectors results to a decline in their value added price and demand. However, the decline experienced by the agriculture sector is much higher than that of the industry. In spite of this, a reallocation towards Banana in agriculture, Semi-conductor and textile and garments in industry occur as both sub-sectors expand. In contrast, the value added price and demand in the service sector increases resulting to a resource reallocation towards it. 6 The share of Palay imports at the base is almost nil 18

20 00'*,!+* 0" Value Added Changes (%) SECTORS δva i δpva i δr i, % Change (%) in Labor Demand L* L1** L2** L3** L4** Irrigated Palay Non_irrigated Palay Corn Banana Fruits Coconut Sugarcane Other agricultural crops Hog Chicken_egg and other poultry products Other livestock Fishing Other Agriculture AGRICULTURE Mining Meat Processing Canning and preserving of fruits and vegetables Fish canning and processing Coconut processing Rice and corn milling Sugar milling and refining Beverages_sugar_confectionery and related products Other food manufacturing Textile and garments Wood_paper products Fertilizer Other chemicals Petroleum_related products Metal and related products Semi_conductors and other electronic products Motor vehicles and other machineries Other manufacturing Construction and utilities INDUSTRY Wholesale trade Other service Government services SERVICES ( $ where li : labor ri: rate of return to capital vai : value added *L aggregate labor δi : change pvai : value added prices **L1, L2, L3, & L4: Labor type 1, 2, 3, & 4 19

21 00'*,!+* 5" Value Added Changes (%) SECTORS δva i δpva i δr i, % Change (%) in Labor Demand L* L1** L2** L3** L4** Irrigated Palay Non_irrigated Palay Corn Banana Fruits Coconut Sugarcane Other agricultural crops Hog Chicken_egg and other poultry products Other livestock Fishing Other Agriculture AGRICULTURE Mining Meat Processing Canning and preserving of fruits and vegetables Fish canning and processing Coconut processing Rice and corn milling Sugar milling and refining Beverages_sugar_confectionery and related products Other food manufacturing Textile and garments Wood_paper products Fertilizer Other chemicals Petroleum_related products Metal and related products Semi_conductors and other electronic products Motor vehicles and other machineries Other manufacturing Construction and utilities INDUSTRY Wholesale trade Other service Government services SERVICES ( $ where li : labor ri: rate of return to capital vai : value added *L aggregate labor δi : change pvai : value added prices **L1, L2, L3, & L4: Labor type 1, 2, 3, & 4 20

22 Household Income: Factor income from for all households decline as a result of agricultural contraction (Table 13a and 13b). Evidently, human capital proves important as only income from skilled industrial labor increase by an average of 1.2 percent. Return to capital in the service sector increase as a result of output expansion and resource reallocation effects. * 2 3 Table 12a Sources of Household Income (% Changes) (SIM_1) ( () ()* ()+ (, (, * (, + (!! -. / / / / / (!! 0. / / / / / 1! -. / 1! 0. / / / / (!! / / / / / / / 1! / / / / / / / - * / / / / / / / Table 12b Sources of Household Income (% Changes) (SIM_2) ( () ()* ()+ (, (, * (, + (!! -. / / / / / (!! 0. / / / / / * 2 1! -. / 1! 0. / / / / (!! / / / / / / / 1! / / / / / / / 3 - * / / / / / / /

23 Poverty: In the FGT calculation, poverty effects come from two sources: (i) from the change in household income; and (ii) from the change in consumer prices, which affects the nominal value of the poverty line. The results of the calculations for the three poverty indices, headcount, gap, and severity, are presented in Table 14a and 14b. Recall that poverty in the Philippines is likely to be influenced by: Spatial factors, human capital or educational attainment, and Gender of household head. The poverty results for SIM_1 and SIM_2 will be discussed separately below. SIM_1. Overall, Poverty headcount seems to decrease marginally as a result of tariff reduction. However, both the gap and severity increases implying that the poorest of the poor become much more immerse in poverty. Urban areas (which exclude NCR) mimic the trend in national poverty. 06!+* 0" / / / / / / / / / / /!" / / / / / / / / / / / / / / / / / / / / / # / / / / / / / / / / / / / / % / / / / / %%! / / Spatial Consideration: Clearly, rural households are worse off than urban households. In particular, rural households who depend on agriculture experience an increase in all measures of poverty. Households residing in the NCR fare better compared to other urban dwellers due to proximity to major industries and the expanding service sector. Human Capital: Highly educated household heads benefit from tariff reduction having the ability to move towards sectors offering higher returns. Indeed, all 22

24 poverty indices for highly educated male and female household heads decline. For instance, poverty headcount decreases by 3.88 and 2.14 respectively for both. Once again, highly educated households residing in NCR gain the most. However, highly educated males in the rural areas experience a decline in poverty as skilled wages and return to capital in agriculture drop by 6.9 and 6.6 percent. Gender: It appears that female-headed households are more receptive to trade liberalization than their male counterparts. The main driving force behind this is because of the expansion: of semi-conductors, textile and garments, and wholesale and retail trade sub-sectors which mainly employ female workers. Female household head s poverty headcount decrease marginally by 0.78 compared to 0.03 for males. In contrast, poverty gap decreases by 0.33 for females whereas males experience a 1.23 increase. Poverty severity increases for both, with a substantial increase among male-headed households. 06 / / / / / / / / / /!" / / / / / / / / / / / / / / / / / / / / / # / / / / / / / / / / / / / / % / / / / / / %%! / / 23

25 SIM_2. The decline in the national poverty headcount is much less under the full tariff elimination scenario. In particular, both the gap and severity worsens by 2.17 and 3.59 compared to 1.11 and 1.84 under the actual tariff reduction. Households residing in rural areas experience an increase in poverty as their labor and capital income from agriculture decrease substantially. This stems from the rise in consumption prices coupled with a decline in average nominal income. Only households living in NCR benefit NCR because of its proximity to major industries and the expanding service sector. Conclusion The two policy experiments conducted in this paper generate quite identical outcome. Tariff reduction results to a decline in over-all output as agriculture and industry contracts. The contraction in agriculture is higher than that of the industry. Nonetheless, sub-sectors such as Banana in agriculture, Semi-conductor and Textile and Garments in industry expand arising from substantial output and export growth. The service sector appears likely to benefit the most from tariff reduction. The decline in composite prices for both agriculture and industry increases trading activity in the wholesale and retail trade sub-sector. With this, the entire sector s output expands encouraging resource reallocation. Tariff reduction results to consumer substitution towards cheaper imports. On the other hand, the lower prices imported inputs reduce the local cost of production benefiting the outward-oriented-import-dependent industrial. National poverty headcount decreases marginally as a result of tariff reduction. However, both the gap and severity worsens implying that the poorest of the poor become much more immerse in poverty. Poverty stricken rural areas suffer the most as all poverty indices increase. This stems from a decline in average income (as a result of agricultural contraction), compounded by an increase in average consumption prices. Poverty in NCR decreases considerably because of its proximity to major industries and the expanding service sector. In conclusion, tariff reduction appears likely to increase poverty especially in the rural areas and among the rural low educated. The tariff reduction between 1994 and 2000 hurts poor the most. This suggests that the government should consider a regional complementary policy aimed at addressing regional imbalances. In addition, a human capital development program must be instituted as simulation results point out that education and skill are the best ally against poverty. 24