This Selected Issues paper analyzes the factors behind the relatively strong performance of the Philippines in recent years (1990s) as well as the remaining reform agenda. The paper highlights that the Philippines has received considerable attention in recent years as it “emerged” in the early 1990s from a long period of slow growth and economic imbalances, and then managed to escape the “Asian crisis” relatively unscathed. The paper examines the public finances for the Philippines. It also analyzes the monetary sector, external sector, and the banking sector reforms.

Abstract

This Selected Issues paper analyzes the factors behind the relatively strong performance of the Philippines in recent years (1990s) as well as the remaining reform agenda. The paper highlights that the Philippines has received considerable attention in recent years as it “emerged” in the early 1990s from a long period of slow growth and economic imbalances, and then managed to escape the “Asian crisis” relatively unscathed. The paper examines the public finances for the Philippines. It also analyzes the monetary sector, external sector, and the banking sector reforms.

II. ECONOMIC GROWTH1

A. Introduction

1. The Philippine economy has long been set apart from many of its Asian neighbors by its weaker growth performance; however, its performance in recent years offers grounds for optimism.

  • In 1950, Philippine per capita income placed it in the center of a group of its Asian neighbors; by 1996, it had slid to the bottom of the group (Figure II.1). Relative to the United States, Philippine per capita income fell from one-seventh in 1950 to one-tenth by 1996, while all its neighbors improved, some spectacularly. Over the last three decades, per capita real GDP growth in the Philippines averaged less than one percent a year, compared with close to 3 percent per year in the ASEAN-3 (Indonesia, Malaysia, and Thailand).

  • However, the country’s more recent economic performance offers grounds for some optimism; in particular, the period 1993–98 was marked by a performance better than the historical average, at a time when performance in the ASEAN-3 countries had slipped below historical norms (Table II.1).

  • At the height of the Asian crisis, Philippine real GDP declined only 0.5 percent in 1998 (despite a very bad year for agriculture), compared with declines of 6–14 percent in the ASEAN-3.

Figure II.1
Figure II.1

Philippines: Real GDP per Person in 1950, 1973, 1996

(Constant prices based on PPP exchange rates -- 1990 international dollars)

Citation: IMF Staff Country Reports 1999, 092; 10.5089/9781451831221.002.A002

Source: IMF, World Economic Outlook database.
Table II.1.

Growth in the Philippines and ASEAN-3:

Historical Average and Recent Performance

article image
Source: Perm World Tables, updated using growth rates from World Bank and IFS databases.

2. This chapter reviews the growth experience of the Philippines, with a view to assessing the outlook and policy requirements for rapid growth over the medium term. Section B presents stylized facts about Philippine economic growth, recent changes in economic performance, and the role of policies and institutions in shaping growth. Section C describes growth performance during the Asian crisis. Section D contains a statistical analysis of the sources of growth, with particular focus on the contribution of trade liberalization. The chapter concludes with an assessment of the medium-term growth outlook.

The main findings of each section are as follows:

  • Role of policies: The increasingly outward-oriented and market-based policies adopted since 1986 appear to have borne fruit. Historically, the salient features of Philippine growth have been: (i) its unevenness or “boom-bust” nature; (ii) absence of productivity gains; and (iii) a distorted industrial structure (biased towards very capital-intensive, import-competing production). However, the turnaround in the orientation of economic policies since 1986 appears to have yielded results in the form of (i) strong economic performance—particularly since 1993—and the resilience exhibited during the Asian crisis; (ii) signs of pick-up in productivity growth; and (iii) some movements towards a less distorted industrial structure.

  • Sources of growth: Statistical analysis indicates that the pick-up in performance in recent years is due to increasing openness, stable investment rates, and an increase in foreign direct investment (FDI). Empirical estimates of the determinants of growth for the period 1970 to 1995—building on Fischer (1993) and Barro and Sala-i-Martin (1995)—show that the weaker performance of the Philippines, on average, relative to East Asian countries is due to its lower degree of openness and lower (and uneven) investment rates, including FDI. By the same token, the recent pick-up in Philippine performance can be attributed to some catching-up with East Asian countries with respect to these determinants.

  • Medium-term growth outlook: Prospects depend on maintaining the broad thrust of policies adopted since 1986, while deepening considerably the reforms in several areas. In particular, robust growth is contingent on:

  • continued trade liberalization, accompanied by a shift towards broad-based incentives and complementary investments in infrastructure;

  • maintenance of high and stable investment rates (including FDI, but supported as well by higher domestic savings);

  • a low inflation environment, which in turn requires fiscal consolidation and a continuation of prudent monetary policies;

Under this scenario, the estimates of potential output growth in the Philippines range between 3½ percent and 5 percent for the period 2000–05.

B. The Role of Policies and Institutions in Shaping Growth

Stylized facts about Philippine growth

3. Growth since the 1970’s has been uneven (Table II.2). The rapid growth of the 1970’s proved unsustainable and culminated in a sharp reversal during the first half of the 1980’s. The period since has been marked by slower but somewhat steadier growth. Performance since 1994 has been particularly encouraging, with the economy achieving increasing growth rates up until the outbreak of the Asian crisis in 1997 (see Section C for a discussion of the crisis period).

Table II.2.

GDP Growth by Sub-Periods, 1969–98

article image
Source: IMF, World Economic Outlook database.

4. Despite recent progress, per capita income remains low (Figure II.2). Per capita GDP is below its level in 1980 and only 3 percent higher than that in 1990. The debt crisis in the early 1980s, the near-crisis in 1991, and the mixed performance in the intervening years took their toll from which the economy has yet to recover fully.

Figure II.2
Figure II.2

Philippines: GDP per Capita, 1970-98

Citation: IMF Staff Country Reports 1999, 092; 10.5089/9781451831221.002.A002

Sources: Perm World Tables; World Bank; and IMF, World Economic Outlook.

5. For East Asian countries in general, empirical studies point to capital accumulation, rather than productivity gains, as the more significant contributor to growth. In traditional models, per capita income growth can be driven by improvements in aggregate productivity ascribed to technological progress, or increased factor inputs (physical and human capital). Attempts to measure the relative contributions of factor inputs and technological progress have been the focus of extensive research. A commonly used approach is to deduct from growth in output per worker a weighted average of the accumulation of physical and human capital per worker, and to interpret the residual as total factor productivity (TFP) growth—the increase in productivity brought about by technological advances and greater organizational efficiency. In the case of East Asian economies, the most common finding is that capital accumulation has made the largest contribution, while productivity growth has made smaller but still significant contributions. Some studies suggest that almost all of the region’s growth was due to capital accumulation, with productivity growth contributing almost nothing.2

6. In the case of the Philippines, productivity growth has historically made little—or even a negative—contribution to output growth; but there is some evidence of a pick-up in recent years. Most studies indicate that TFP growth in the Philippines has been negative (Table II.3). When TFP growth is estimated over more recent periods, there is evidence of improved performance in the 1990s, with TFP growth averaging between ¾ percent and 1 percent a year.

Table II.3:

Philippines: Estimates of Total Factor Productivity Growth

article image

7. At the sectoral level, aggregate growth has been driven mainly by the industry and service sectors; agriculture has rarely contributed more than 1 percent of growth each year over the past 20 years (Figure II.3). Agriculture still accounts for a large proportion of output, and employment in the sector remains high, at 43 percent of total employment. While other countries in the region have benefited from continuing shifts in the output and employment structure away from agriculture, the Philippines has lagged behind (Table II.4).

Figure II.3
Figure II.3

Philippines

GDP GROWTH BY SECTOR, 1980–98

(Changes in percent of previous year’s GDP)

Citation: IMF Staff Country Reports 1999, 092; 10.5089/9781451831221.002.A002

Source: Data provided by the Philippine authorities.
Table II.4:

Structure of Output and Employment, 1970–96

article image
Sources: World Bank and IMF; Leipziger (1997), Dolan (1993), Mill (1995), Jomo (1990), Krongkaew (1995).

Role of policies and institutions

8. As in many developing countries, government policies have played an important role in shaping the nature and extent of growth in the Philippines. In particular, government policies in the period following independence were characterized by (i) import-substitution; (ii) a strongly interventionist stance, as reflected in a range of fiscal incentives and directed credit; and (iii) the promotion of capital-intensive industries in a labor-abundant country. Despite piecemeal attempts at reform, the above policies prevailed through the 1970s and much of the 1980s.

There were fundamental problems with these policies.3

  • The financing for these policies was difficult to secure on a sustained basis. As a consequence, growth was uneven. The economy suffered through episodes of severe macroeconomic imbalances as manifested in balance of payments crises, large fiscal deficits, and high inflation. In particular, the rapid build-up of external debt prior to 1982 helped feed an unsustainable boom which ended in the debt crisis.4

  • Government intervention in the economy hindered productivity growth and the efficient allocation of resources.5 The trade regime was protectionist and discretionary. While there were export incentives, the policy tilt was still in favor of capital-intensive industries that could not compete in international markets. The central bank provided subsidized credit to enterprises, and later rescued them from financial difficulties. The pricing policies of marketing boards in the agricultural sector did not encourage efficient production,6 nor did they encourage increased processing of products before export. Land reform was held back by the vested interests of traditional landowners. During this period, the role of the public sector expanded as public corporations were established, encroaching into more and more areas of economic activity.

9. The period since 1986 has been marked by a turnaround in the orientation of policies, although policy implementation was hampered by political instability. When the Aquino government took power in 1986, it adopted an ambitious reform agenda which included liberalization of the trade and exchange regimes. The quantitative restrictions on imports were gradually phased out or tariffied, and the tariff structure was simplified and rationalized. A new, more liberal Foreign Investment Act was passed. Many public corporations were privatized. A new program of land reform was also promised as part of the Comprehensive Agricultural Reform Program (CARP). Some of the reforms were implemented early on, but others (mainly land reform) were subsequently reversed or watered down. Political instability—there were six coup attempts in the first three years of the Aquino administration—was an impediment to full implementation of the reform program, and a series of shocks to the system (natural disasters, power shortages) contributed to continued macroeconomic imbalances.

10. The Ramos government, which assumed office in 1992, adopted a host of reforms designed to put the economy on a rapid sustainable growth path. During this period most restrictions on current and capital transactions were lifted and tariff reforms continued (Chapter V). A new Central Bank Act was passed in 1993, establishing price stability as the key objective of monetary policy and strengthening the legal framework for improved bank supervision. At the same time, the Central Bank was recapitalized, with the portfolio of nonperforming assets hived off to a separate entity (Chapters IV and VI). Marketing monopolies were abolished, and producer price controls were eliminated. Electricity generation and telecommunications were opened to the private sector, foreign participation was allowed in the banking sector, and the domestic shipping and oil sectors were liberalized. A Comprehensive Tax Reform Package was passed in 1998 (see Chapter III). During this period, the external debt burden was brought down to more manageable levels, helped by debt restructuring (which had begun in the 1980s and was completed in 1994) and prudent fiscal management.

11. The reforms since 1986 appear to have had a favorable impact on economic performance. Per capita growth has picked up, particularly since 1993. Productivity growth has picked up as well, and the industrial structure is showing signs of dynamism. As noted in IMF (1995), “this improvement in TFP is consistent with enhanced efficiency arising from the reform program.”7 In particular, the role of the public sector in the economy has been reduced through fiscal consolidation, the divestment of many public enterprises, and structural reforms such as trade liberalization, financial sector liberalization, and private sector participation in activities which were previously in the public domain (e.g., power generation and infrastructure projects). Market incentives have replaced the more direct hand of the government.

C. Growth Performance During the Asian Crisis

Growth performance

12. The Philippines has managed to escape the Asian crisis relatively unscathed. In particular, output experienced a slight contraction in 1998 while in the other ASEAN-5 countries and Korea the decline in output has been severe (Table II.5).

Table II.5:

Economic Growth in the Crisis Countries

(Percent change)

article image
Source: IMF, World Economic Outlook.

13. The decline in output in the Philippines was largely a result of adverse weather conditions brought about by El Niño. In 1998, agricultural output fell by over 6 percent, while non-agricultural output grew by 1 percent. Had the weather conditions been normal, the Philippines economy could have avoided a recession.

14. External demand continued to grow during the crisis period, while domestic demand contracted. Domestic investment experienced a large decline, but the drop in domestic demand was muted by continued consumption growth (particularly by the private sector). On the external side, rapid export growth continued, albeit at a slowing pace, while imports contracted reflecting the investment decline.

15. How is it that the Philippine economy was able to avoid the large declines in output experienced by other crisis countries?

Possible explanations

16. The stage of the economic cycle may have played a role in the muted impact of the external shock on the Philippines. At the onset of the Asian crisis, the country was still in the initial stages of an economic boom (as opposed to the most heavily affected crisis countries which had experienced very rapid growth for much longer periods). While there existed elements of a bubble economy (two years of rapid economic growth, with a boom in real estate and construction supported by large capital inflows and rapid bank lending), the expansion was still in its early stages, and the adjustment to the reversal of capital flows was not as painful as for the other countries.

17. The reforms carried out since 1986 seem to have helped to improve the resilience of the economy. The increased market orientation of the economy and institutional strengthening made domestic systems more robust. In particular, reforms of the financial sector and trade liberalization seem to have played an important role.

  • The reforms have strengthened the structure of the economy and the financial system, building greater resilience against shocks. The experience of the external debt crisis in the 1980s triggered the associated domestic financial crisis and measures to open up the banking system to more competition (including from abroad), creation of a new and independent central bank (Bangko Sentral), and improved banking supervision and prudential regulations. As a result, the banking system was better able to handle subsequent shocks.

  • The winding down of government interference in business decisions triggered further reform and changed the behavior of the private sector. In particular, the backlash against “crony capitalism” and the BSP’s virtually complete retreat from directed lending activities avoided much of the inefficiencies in financial intermediation experienced in other crisis economies. The business environment became increasingly market-oriented as the reforms took hold. As a result, the stronger elements of the private sector have been able to better withstand difficult economic conditions.

  • Trade liberalization exposed domestic manufacturers to more competition forcing them to learn to operate successfully in international markets. A favorable impact of trade liberalization on Philippine growth is suggested by the discussion in the next section.

D. Sources of Growth in the 1990s

Trade liberalization and growth

18. The positive impact of trade liberalization on growth has been well documented in the empirical literature (see, for instance, Vamvakidis (1999)). Both cross-country and time series studies generally show that open economies grow faster and have higher investment shares on average. These results are robust to the choice of a large variety of openness measures, such as trade shares, tariff and nontariff barriers and indices of trade distortion. Further evidence comes from the fact, noted above, that many of the economies that followed import substitution policies performed badly during the 1980s and 1990s. Theoretical growth studies have provided the foundations for these empirical findings, arguing that openness influences growth through more investment and technological progress.

19. From the late 1980s, the market-oriented reforms are reflected in changes in Philippine trade shares. Figure II.4 shows that the ratio of exports to the GDP increased fairly consistently during the 1980’s, with the trend accelerating in the 1990s. The ratio of imports to the GDP shows a similar pattern with the trend starting to accelerate in the late 1980s.

Figure II.4
Figure II.4

Philippines: Exports and Imports, 1960-98

(In percent of GDP)

Citation: IMF Staff Country Reports 1999, 092; 10.5089/9781451831221.002.A002

Sources: IMF, International Financial Statistics; and data provided by Philippine authorities.

20. Statistical tests confirm that the period 1989 to 1992 was marked by a significant change in the behavior of export shares. To determine formally if the structure of trade has changed during the recent decade, we follow a method—developed in Ben-David and Papell (1997, 1998) and described in Annex II.1—that determines when structural change in a variable has occurred by estimating the year of a break in the trend of the variable. (Note that the year of the break is not pre-determined, but estimated through time series regressions.) Application of this method to export and import data for the period 1960–97 yields the following results:

  • The estimates show that there has been a change in the trend of these variables: the hypothesis of “no break” can be rejected at the 5 percent significance level for both variables.

  • As shown in Table II.6, the break year for the export share is estimated to be 1992; the export share averaged 20 percent before 1992, and nearly doubled in the period thereafter. The break year of the import share is estimated to be 1989, also with a near-doubling of the import share from its average of 21 percent before 1989.8

Table II.6:

Structural Breaks in Trade Shares, 1960–97

article image
Source: Fund staff estimates.

21. The estimates above show a structural change in trade patterns at a time when economic performance also improved significantly. While this provides suggestive evidence, a more formal investigation of the impact of trade liberalization on growth has to control for other variables that also influence growth. This is attempted below.

Evidence from panel regressions

22. This section provides estimates of the impact of various determinants—policies and structural factors—on Philippine economic growth. The estimates are based on panel (i.e., cross-country, time-series) regressions and thus provide an estimate of the average impact of a particular determinant on growth.9 These estimated average impacts can be used to shed light on: (i) why the Philippines, historically, has done worse than other country groups, particularly in East Asia, and (ii) which factors account for the recent improvement in performance.10

23. Slower growth in real per capita GDP in the Philippines relative to other country groups mirrors the behavior of the likely growth determinants. Table II.6 presents average values over the period 1970–95 of the growth determinants for the Philippines, and for the groups of developing countries (countries with GDP per capita less than 5,000 in 1995, measured in 1987 U.S. dollar), developed countries, and East Asia (excluding Japan). Comparing the Philippines with the other country groups leads to the following conclusions:

Table II.6:

Country Comparisons of the Determinants of Growth

article image
Sources: Data are from World Development Indicators, with the exception of the capital market openness measure.
  • The Philippine economy remains less open than the other country groups, despite recent liberalization. A variety of trade variables are used to test for the robustness of this conclusion, since trade measures are not always highly correlated with each other.11 All indicators show that the Philippines, despite the trade reforms in the 1990’s and the large increase of the trade share, is less open than all other country-groups, and especially compared to the rest of East Asia.

  • The investment share of the Philippines is similar to that of other country groups but lower than in the rest of East Asia. This is true both of domestic investment and FDI. However, while the average share of net FDI to GDP was relatively small on average, it has increased considerably in recent years (reaching the ratio in developed economies in 1995). In addition, the ratio of savings to GDP has been relatively low in the Philippines (21 percent on average in 1970–95, compared to 26 percent in developed countries and 30 percent in East Asia). Higher savings, especially domestic savings, would be necessary to finance a higher rate of investment.

  • Capital markets in the Philippines are less open than in developed countries and in the rest of East Asia, as measured by an index of free capital mobility (as well as the average share of net FDI to GDP, as noted earlier). The index—constructed by Gwartney, Lawson and Block (1996) using data from the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions—measures both the freedom of foreigners to invest within the country and the freedom of citizens to invest abroad. The higher the value of the index, the freer are capital transactions with foreigners. Measured on a scale of 1 to 10, the Philippines averaged 2, compared with 4 for East Asia and 5.6 for developed economies.

  • Inflation has been lower, on average, in the Philippines than in other developing countries.

  • The share of government consumption in GDP is lower in the Philippines than in other country groups. Empirical studies have found that government consumption can reduce growth if it crowds out private sector investment; if it increases rent-seeking behavior; and if it distorts market incentives (in contrast, government investment can promote economic growth if it improves infrastructure).

  • The age dependency ratio of the Philippines is comparable to the average in developing countries, but considerably higher than the average in developed countries and in the rest of East Asia.

  • With respect to school enrollment ratios, the Philippines is ahead of all other country groups for primary schooling, and behind only the developed countries group for secondary schooling.

24. Panel regression analysis indicates that greater openness and higher investment shares (including FDI) have a significant influence on growth—which may explain the recent pick-up in Philippine performance. The results also indicate that a low inflation environment is conducive to growth. Methodology and estimates are described in Annex II.2; the main findings are as follows:

  • Impact of trade liberalization: The estimates indicate that, on average, opening to international trade leads to faster growth; the trade variables are statistically significant at least at the 10 percent level in most specifications, except the export duty ratio. Based on these estimates, trade liberalization since the 1980’s is estimated to have contributed between 0.2 and 0.7 percentage points to Philippine growth. These estimates should be considered as a lower limit of the impact of trade on growth, since they measure only the direct impact. Trade may also influence growth indirectly through other growth determinants. For example, Levine and Renelt (1992) and Vamvakidis (1999) have shown that open economies invest more than closed economies, and through this effect grow faster.12

  • Impact of investment shares:

    • Higher domestic investment also contributes to faster growth. The Philippine investment share has been quite volatile and in 1995 was below the historical average. The estimates suggest that an increase of the Philippine investment share from 22 percent to 25 percent would contribute to faster growth of 0.6 percent.

    • Foreign direct investment has a positive and statistically significant impact on growth.13 The Philippine FDI to GDP ratio increased from -0.4 percent in 1970 to 2.0 percent in 1995. The estimates imply that this increase should be correlated with faster growth by about 0.8 percent. If it had reached the average level in the rest of East Asia (4,3 in 1995) growth would have been higher by 1.5 percentage points.

  • Impact of other variables:

    • The coefficient for inflation is always statistically significant, but small, suggesting that episodes of very high inflation have a negative impact on growth.

    • The regressions indicate the existence of conditional convergence, since the coefficient of the initial GDP per capita is always negative (though not always statistically significant).

    • The estimates of the free-capital-mobility variable are not significant in any of the specifications, and therefore, the impact of financial openness on growth could not be established empirically.

    • The age dependency ratio, the ratio of government consumption and the schooling variables do not have a statistically significant impact, after controlling for differences in the other independent variables. However, these variables are significant in cross-country regressions.

E. Medium-Term Growth Outlook

25. The findings of the previous sections suggest that sustained growth is contingent on continuing with the outward-oriented and market-driven policies adopted since 1986. In particular, the following developments would improve the medium-term growth outlook: (i) continued trade liberalization; (ii) raising saving and investment rates, including FDI; (iii) a low inflation environment; and (iv) improved allocation of resources across sectors. Each of these developments would, in turn, require a deepening of reforms, as described below (and in other chapters of the paper). Under this scenario, potential output growth in the Philippines could rise close to 5 percent for the period 2000–05 (see Box II.1).

  • Continued trade liberalization: The regression results suggest that openness is critical for growth. As noted earlier, despite recent progress, the Philippines is still less open than other country groups; consequently, further trade liberalization is essential. In addition, the experience of other countries suggests that continued growth benefits from trade liberalization require a shift to the provision of broad-based incentives (rather than a “picking-winners” approach), and complementary investments in infrastructure.14

  • Higher investment (including FDI):

    • The regression results indicate strong association between investment shares and GDP growth. The Philippines’ investment share lags behind that of East Asia (though the crisis has emphasized the need to pay attention to quality of investment, not just quantity).

    • Attracting FDT will require continued liberalization and improvements in “governance.” An attractive investment climate includes “economic security” (enforcement of contracts; stable legal and tax regime; reliability of basic economic services).

    • Raising domestic savings is critical to finance the higher investment rates while limiting the dependence on foreign savings and the resulting vulnerability to capital flow reversals (Figure II.5).

  • Low inflation: The regressions results indicate that episodes of very high inflation are detrimental to growth. In recent years, the Philippines has been able to avoid high inflation. Continuing this will require prudent monetary policies and a return to fiscal consolidation following the temporary widening of the fiscal deficit during the crisis.

  • Improved allocation of resources across sectors: As noted in Section II.B, a shift in the pattern of production and employment from agriculture to manufacturing and services would raise growth rates as resources are shifted towards more productive areas of activity. Improving agricultural performance would facilitate this shift (as discussed in Chapter VII).

Figure II.5
Figure II.5

Investment and Saving, 1980-97

(In percent of GNP)

Citation: IMF Staff Country Reports 1999, 092; 10.5089/9781451831221.002.A002

Source: Data provided by the Philippine authorities.

Estimates of Potential Output for the Philippines, 2000–05

Estimates of potential output are a useful input into policymaking. For example, potential output growth provides a guideline for medium-term growth projections. Potential output growth is also used in estimating a cyclically-neutral budget balance, which provides a measure the actual fiscal stance (expansionary or contractionary, relative to a base year) and fiscal impulse (change in fiscal stance between two periods). Finally, estimates of the output gap—generated from potential and actual output—can be used to gauge inflationary pressures in the economy.

The methods used to derive estimates of potential output for the Philippines are:

Production function approach: This method assumes that economies’ production functions can be approximated by the Cobb-Douglas technology with two-factors, capital and labor. Assumptions about growth in these two factors are used to derive the series for potential output.

Hodrick-Prescott (HP) filter: This is a statistical method which removes short-run fluctuations and leaves behind a series whose smoothness is determined by a parameter choice.

Growth accounting approach: This method is based on the panel regressions described in the main text. The estimate of potential output is derived on the basis of the coefficient estimates and assumptions about how a deepening of assumed reform efforts will translate into values for the independent variables.

Estimates of Potential Output Growth for the Philippines: Average, 2000–2005

article image

ANNEX II.1 Trade Liberalization in the Philippines: Results from “Trend Break” Tests

The first step is to determine whether the variable of interest follows a unit root process. If the variable does not contain a unit root, the test consists of estimating the equation:

x=α+βDUt+γt+ϑDTt+Σi=1ncixti+ɛt

where, x is (for example) the ratio of exports to GDP, t is time, DUt is equal to 1 if t > Tbr, where Tbr is the break year, and 0 otherwise, DTt is equal to t - Tbr if t > Tbr and 0 otherwise, and xt-i is the value of x lagged i times.

If x follows a unit root process, a test in first differences improves power (Vogelsang (1997)). In this case x has no trend, and structural break is a break in the mean of the change of x. The estimated model is the following:

Δx=α+βDUt+Σi=1nciΔxti+ɛt

The number of lags in the regression is determined—following the methodology in Campbell and Perron (1991) and Ng and Perron (1995)—by initially estimating with a maximum number of lags and sequentially dropping one lag until the first significant lag is found. If no lag is significant, the estimated model does not include any lags. Following Ben-David and Papell (1997 and 1998), the maximum number of lags is 8, and the test is conducted for the 10 percent significance level.

The data set is for the period 1960–97, but the estimation trimming is 5 years. Hence, the regression is estimated sequentially for each year in the period 1965–92 (1965<Tbr<1992). The value of Tbr. which maximizes the F-value for testing the hypothesis β = 0 is the year of the structural break. If no DUt is found to be statistically significant, then x does not have a structural break. If x follows a unit root and the first model is estimated, the appropriate critical values are taken from Vogelsang (1997). If x does not follow a unit root, the standard F-statistic for the hypothesis β = 0 can be used.

It is important to note that this methodology can detect only one structural break. If the variable in question has more than one structural break, this methodology will detect the larger break (if the breaks do not offset each other), but not any other structural breaks. This is a drawback, but it should be more of a problem with longer time series than the ones used here.

ANNEX II.2 Impact of Trade Liberalization on Growth

The regressions are estimated using a random effects estimator for a panel of all countries with available data (54 to 93 countries, depending on the independent variables in the regression), with each data point being a five-year average over the period 1970–95. Since the sample is a panel (i.e., it includes both cross-country and time-series variation), the estimated impacts will be smaller than in the more commonly-used cross-country regressions,15 because most of the growth determinants vary considerably across countries, but to a much smaller extent through time. However, these panel regression estimates will be more reliable for projecting the impact of changes in policies on growth.

As discussed in the main text, the trade liberalization since the 1980’s is estimated to have contributed between 0.2 and 0.7 percentage points to Philippine growth.

  • The numerical estimates of the trade share imply that an increase of this share by 10 percent is correlated with faster growth by 0.1 to 0,6 percent annually (0.3 on average), depending on the specification. The trade share of the Philippines (PPP-adjusted) was equal to 14.2 in 1980 (there are no available data for the PPP adjusted trade share before 1980) and has increased to 19.9 in 1995.16 The estimate of the trade share implies that this increase by 5.7 percentage points is expected to be correlated with an increase of GDP growth by 0.2 percentage points during the same period. If the trade share of the Philippines had reached the level of the average trade share in developed economies in 1995 (which was 77.3), growth would have been higher by 1.7 percentage points, and 2.3 percentage points higher if the trade share had reached the average level in the rest of East Asia (equal to 95.8).

  • The openness index estimates imply that a year of openness leads to faster growth by 0.1 percent. Since the Philippines has been open since 1988, it is expected that the growth contribution of openness should have been 0.7 percentage points.

  • The import duty coefficient is significant at the 10-percent level in only one of the two specifications (and only when the trade share is included in the regression). A possible reason is that the import duty is just a proxy, often not very precise, of the weighted average tariff rate. The average estimate of the import duty is -0.06. This estimate implies that a reduction of the import duty ratio by 10 percent, is correlated with faster growth by 0.6 percentage points. The Philippine import duty ratio has remained almost the same during the period 1970–95. As Table 2 shows, it was equal to 14.8 in 1995. If it had reached the average level in the rest of East Asia, equal to 4.4 in 1995, growth would have been higher by 0.6 percentage points on average, and if it reached the average level in developed countries, equal to 1.7 in 1995, growth would have been higher by 0.8 percentage points.

Annex Table II.1.

Results of Panel Growth Regressions

article image
Sources; Data are from World Development Indicators, with the exception of capital market and trade openness.Note: The above estimates are from random-effects panel data regressions. The data are five year averages for the period 1970–95. All regressions have GDP per capita growth as the dependent variable. The regressions that include the trade share include, in addition, the log of GDP to control for country-size differences.

References

  • Austria, Myrna S. and W. Martin, 1995, “Macroeconomic instability and growth in the Philippines, 1950–87: A Dynamic approach.” Singapore Economic Review, Vol. 40 No. 1, pp. 6581.

    • Search Google Scholar
    • Export Citation
  • Barrell, Ray and James Sefton, 1995, “Output gaps. Some evidence from the U.K., France and Germany.” National Institute Economic Review: No. 151 (Feb.). pp. 6573.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barro, Robert J. and Xavier Sala-I-Martin, 1995, Economic Growth, McGraw-Hill, Inc.

  • Ben-David, Dan, and David H. Papell, 1997, Structural Change and International Trade, Centre for Economic Policy Research, Discussion Paper no. 1568.

    • Search Google Scholar
    • Export Citation
  • Ben-David, Dan, and David H. Papell, 1998, Slowdowns and Meltdowns: Postwar Growth Evidence from 74 Countries, Review of Economics and Statistics, pp. 56171.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Benhabib, J. and M.M. Spiegel, 1994, “The role of human capital in economic development: evidence from cross-country data.” Journal of Monetary Economics, Vol. 34, pp. 14373.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Borensztein, E., J. De Gregorio and J-W. Lee, 1998, “How does Foreign Direct Investment Affect Economic Growth, Journal of International Economics, Vol.45, pp. 11535.

    • Search Google Scholar
    • Export Citation
  • Campbell, John Y., and Pierre Perron, 1991, Pitfalls and Opportunities: What Macroeconomists Should Know About Unit Roots, NBER Macroeconomic Annual.

  • Coe, David T. And C. John McDermott, 1996, “Does the gap model work in Asia.” IMF WP/96/96.

  • Collins, Susan M. And Barry P. Bosworth, 1996, “Economic growth in East Asia: Accumulation versus Assimilation.” Brookings Papers on Economic Activity, No. 2., pp. 135191.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Crafts, Nicholas, 1998, “East Asian growth before and after the crisis.” IMF WP/98/137.

  • De Masi, Paula R., 1997, “IMF estimates of potential output: Theory and practice.” IMF WP/97/177.

  • Dickey, David A., and Wayne A. Fuller, 1981, Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root, Econometrica, Vol. 49, No. 4 (July).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dolan, Richard E., ed. 1993. Philippines: a country study. Washington, DC: Library of Congress.

  • Dowling, Malcolm, and Peter M. Summers, 1998, “Total Factor Productivity and Economic Growth—Issues for Asia”, The Economic Record, Vol. 74, No. 225 (June), 17085.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, Stanley, 1993, “The Role of Macroeconomic Factors in Growth,” Journal of Monetary Economics, vol. 32 (December), pp. 485512.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, Stanley, Ratna Sahay and Carlos Vegh, 1999. “Modern Hyper- and High Inflations,” Draft, January 1999.

  • Greene, William H., 1993, Econometric Analysis (New York: Macmillan).

  • Gwartney, J., R. Lawson and W. Block, 1996, Economic Freedom of the World.

  • Hill, Hal, 1995, The Indonesian economy since 1966. Cambridge: Cambridge University Press.

  • International Monetary Fund, 1995, Philippines: Background Paper (SM/95/253).

    • Crossref
    • Export Citation
  • International Monetary Fund, 1998, World Economic Outlook (October).

  • Jomo, K.S., 1990, Growth and structural change in the Malaysian economy. London: Macmillan.

  • Krongkaew, Medhi, ed., 1995, Thailand’s industrialization and its consequences, Houndmills, Basingstoke, Hants: Macmillan.

  • Krueger, Anne, 1998, “Why Trade Liberalization is Good for Growth,” The Economic Journal, Vol. 108 (September), pp. 151322.

  • Krugman, Paul, James Aim, Susan Collins, and Eli Remolona, 1992, Transforming the Philippine Economy, A NEDA/UNDP Publication.

  • Krugman, Paul, 1994, “Myth of Asia’s miracle,” Foreign Affairs, Vol. 73 (Nov.–Dec), pp. 6278.

  • Leipziger, D., ed. 1997, Lessons from East Asia. Ann Arbor: University of Michigan Press.

  • Levine, Ross, and David Renelt, 1992, A Sensitivity Analysis of Cross-Country Growth Regressions, American Economic Review, 82, 4 (September), pp. 942963.

    • Search Google Scholar
    • Export Citation
  • Ng, Serena, and Pierre Perron, 1995, Unit Root Tests in ARMA Models with Data Dependent Methods for the Selection of the Truncation Lag, Journal of the American Statistical Association 90, pp. 268281.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Perron, Pierre, 1994, Further Evidence on Breaking Trend Functions in Macroeconomic Variables, mimeo, University of Montreal.

  • Sachs, Jeffrey and Andrew Warner, 1995, Economic Reform and the Process of Global Integration, Brooking Papers of Economic Activity, 1995:1

  • Sachs, Jeffrey and others, 1998, Promotion of Broad-Based Economic Growth in the Philippines, A NEDA/UNDP Publication.

  • Sarel, Michael, 1997, “Growth and Productivity in ASEAN Countries.” IMF WP 97/97.

  • Sarel, Michael, 1997, Growth in East Asia. “What We Can and What We Cannot Infer From It,” IMF WP/95/98.

  • Vamvakidis, Athanasios, 1999, Regional Trade Agreements Versus Broad Liberalization: Which Path Leads to Faster Growth? Time Series Evidence, IMF Staff Papers, 46 (March), pp. 4268.

    • Search Google Scholar
    • Export Citation
  • Vogelsang, Timothy, 1997, Wald-Type Tests for Detecting Shifts in the Trend Function of a Dynamic Time Series, Econometric Theory 13, 818849.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Williamson, J., 1969, “Dimensions of postwar Philippines economic progress.” Quarterly Journal of Economics, Vol. 83 No. 1, pp. 93109.

  • World Bank, 1999, World Development Indicators (CD-Rom).

  • Young, Alwyn, 1992, “Tale of two cities: factor accumulation and technical change in Hong Kong and Singapore.” in NBER Macroeconomics Annual.

  • Young, Alwyn, 1995, “Tyranny of numbers: confronting the statistical realities of the East Asian growth experience.” Quarterly Journal of Economics Vol. 110 (Aug.): pp. 64180.

    • Crossref
    • Search Google Scholar
    • Export Citation
1

Prepared by Piyabha Kongsamut and Athanasios Vamvakidis.

2

See Young (1992 and 1995), Krugman (1994).

3

Krueger (1998) provides a lucid discussion of the shortcomings of an import-substitution strategy.

4

For a detailed economic history of the Philippines, see Dolan (1991) and Leipziger (1997).

5

As suggested in IMF (1995), the poor record of productivity growth in the Philippines likely reflects “the misallocation of investment. A large share of public investment during the 1970’s and early 1980’s went toward high-risk, capital-intensive projects, some of which subsequently failed. Meanwhile, the allocation of private investment was distorted by protectionist measures and specific fiscal incentives favoring capital- and import-intensive industries.”

6

For example, two of the major export products (sugar and coconuts) were put under the control of marketing boards which nominally were set up to help farmers but actually enriched a few businessmen. Also, price controls existed for many products.

7

Some other explanations for the pick-up in TFP growth in East Asian countries are advanced in Dowling and Summers (1998). They point out that the rise in TFP coincided with “a huge wave of foreign direct investment into developing Asia. During this period the multinationals brought with them not only capital but also management and organizational skills which could be reflected by higher estimates of TFP. Another hypothesis which is consistent with these facts is that TFP growth in Asia coincided with liberalization of trade within Asia and an increase in the pace of world trade growth after 1985 and on into the 1990’s” (p. 181).

8

Sachs and Warner (1995) suggest that the Philippines has been an open economy since 1988. Ben-David and Papell (1997) estimated the trend break years as 1982 using the export share and 1979 using the import share. However, the end-point of their sample did not permit 1989 or 1992 to be considered as trend breaks (as explained in Annex II.1).

9

The methodology follows Fischer (1993) and Barro and Sala-i-Martin (1995).

10

Since the effect of different factors may vary for different countries, the estimates provide an order of magnitude, rather than a precise measure of the impact of a particular policy on growth in the Philippines.

11

Four variables are chosen, based on the availability of data and their inclusion in earlier studies: the trade share (PPP adjusted), the import duty ratio, the export duty ratio and an openness index constructed by Sachs and Warner (1995). The latter is equal to the number of years a country has been “open” in the period 1970–95.

12

Since growth regressions control for differences in investment shares, the estimate of the trade variables will not measure this effect.

13

Borensztein, De Gregorio, and J-W. Lee (1998) find that FDI contributes to economic growth only when the host country has a minimum threshold stock of human capital.

15

Growth regressions are usually estimated using a cross section of countries, for 5 or 10 or even 20 year averages. The sample would be too small and variation too little to estimate a growth regression using annual data for only one country.

16

Philippine trade over GDP has increased considerably more than the PPP adjusted share, as Table 3 shows.

Philippines: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Philippines: Real GDP per Person in 1950, 1973, 1996

    (Constant prices based on PPP exchange rates -- 1990 international dollars)

  • View in gallery

    Philippines: GDP per Capita, 1970-98

  • View in gallery

    Philippines

    GDP GROWTH BY SECTOR, 1980–98

    (Changes in percent of previous year’s GDP)

  • View in gallery

    Philippines: Exports and Imports, 1960-98

    (In percent of GDP)

  • View in gallery

    Investment and Saving, 1980-97

    (In percent of GNP)