Philippine Institute for Development Studies

Inflation and Economic Growth

in the Philippines
Josef T. Yap

DISCUSSION PAPER SERIES NO. 96-11

The PIDS Discussion Paper Series

constitutes studies that are preliminary and
subject to further revisions. They are be-
ing circulated in a limited number of cop-
ies only for purposes of soliciting com-
ments and suggestions for further refine-
ments. The studies under the Series are
unedited and unreviewed.
The views and opinions expressed
are those of the author(s) and do not neces-
sarily reflect those of the Institute.
Not for quotation without permission
from the author(s) and the Institute.

September 1996

For comments, suggestions or further inquiries please contact:

The Research Information Staff, Philippine Institute for Development Studies
3rd Floor, NEDA sa Makati Building, 106 Amorsolo Street, Legaspi Village, Makati City, Philippines
Tel Nos: 8924059 and 8935705; Fax No: 8939589; E-mail: publications@pidsnet.pids.gov.ph
Or visit our website at http://www.pids.gov.ph
 INFLATION AND ECONOMIC GROWTH IN THE PHILIPPINES'

Jo_'ef T. Yap2

I. Introduction

"Inflation is dead" has been a popular adage for the past five years but it seems deep-
seated problems have a way to resurrect themselves. This cavalier declaration by people like
Lester Thurow and Roger Bootie is inaccurate madmay be based on flawed theories or superficial
empirical evidence.

In the Philippine context at least, inflation is still regarded as a clear and present danger
to macroeconomic stability. After being maintained at single digit levels for the past four years,
1992-95, inflation reached double digit levels in the past nine months, averaging 11 percent over
this period (see Table 1). The Philippines remains to be one of the high inflation countries in
Asia.

What complicates the analysis of the relation between inflation and output growth are
exogenous supply shocks in the past two decades and the manner in which economic managers
responded to these shocks. Policies to stabilize the economy resulted in episodes of stagflation.
Inflation subsequently declines but only after a lag while contractionary effects on output persist
and eventually produce supply bottlenecks that contribute anew to inflationary pressure.
Measures to dampen aggregate demand are again implemented but the vicious cycle is only
repeated. This is another angle to the familiar boom-bust cycle.

This paper will describe the Philippine experience with inflation and economic growth.
Section II explains the importance of controlling inflation by looking at _he relationship of output
and inflation. In Section IIIa brief review of the determinants of inflation in the Philippines will
be presented. This will be the basis of the discussion on policy regimes in the Philippines in
Section IV. Examination of pertinent data will show that heavy dependence on external finance
and myopic macroeconomic management has led to a "fundamental problem of reconciling
policies for economic growth and development with policies for moderating inflation." The
adverse effects of stabilization measures will be summarized in Section V. Brief concluding
remarks are given in the last section.

1Paperpreparedforthejoint studyon "TheFundamentalProblemof ReconcilingPoliciesfor Economic

Growthand Developmentwith Policiesfor ModeratingInflation"underthe EDAPProgramme.Presentedatthe
NationalCouncilofAppliedEconomicResearch,New Delhi,25-26July 1996.

_ResearchFellow,PhilippineInstitutefor DevelopmentStudies(PIDS).Theauthorwishesto acknowledgethe

researchassistanceprovidedbyMa.TeresaS. Duefias.Theusualdisclaimerapplies.
II. The Costs of Inflation

The importance of sound money is one of two facets of the "Washington consensus," a
term coined after the world's de facto capital. Here, the ideas were given a sense of formality by
various multilateral institutions and think tanks including the International Monetary Fund and
the World Bank. Sound money is of course synonymous to low inflation. The other facet of the
"Washington consensus" is free markets and its concomitant programs of liberalization,
deregulation and privatization.

In his critique of the "Washington consensus," Krugman (1995) states that estimates of
the cost of inflation--defined as the overall reduction in real income--are "so low that they are
embarrassing." He is quick to add, however, that bringing down inflation is tmdoubtedly a good
thing and that very high inflation rates seriously disrupt the functioning of a market economy.
He warns though that some of the methods used to achieve disinflation in developing countries
have serious repercussions.

The direct costs of inflation could be examined at different levels. At the individual level,
inflation exacts a toll on those with relatively fixed incomes. Relatedly, inflation favors debtors
at the expense of creditors. Welfare analysis should be applied to determine the net impact of
inflation.

At the fiml level, the effects of inflation are usually couched in temas of menu costs. The
relevant contributions are those of Rotemberg (1982, 1983), Naish (1986), Dmaziger (1988) and
Benabou and Konieezny (1994). Inflation affects output when firms have to incur costs as they
adjust to the new price level (e.g. changing their price list for customers).

At the macroeconomic level, studies have been more quantitative in nature. There has
been recent cross-country evidence supporting the view that long-run growth is adversely
affected by inflation. An oft-cited reference is that of Fischer (1993). The framework that is
used is derived from endogenous growth theory which tries to determine the causes of difference
in growth rates in different countries. The negative effect of inflation on output stems from the
resulting macroeconomic instability which makes it more difficult for economic agents to plan
efficiently thus reducing investment. .

In the Philippine the direct costs of inflation have been measured by estimating its impact
on output and its components. The welfare costs and distributional effects of the inflation tax
have been largely ignored. In the PIDS Annual Macroeconometric Model (Reyes and Yap,
1993a), for example, a rise in sectoral prices and the general price level results in a decline in
demand for the relevant sectoral output. This explains why the impact of a peso depreciation on
total output is contractionary, particularly in the industry and service sectors.

Inflation as a proxy for macroeconomic stability also has a _gative impact on real fixed
investment in the PIDS model. Thus, controlling inflation will result in higher capital formation
and expand the future productive capacity of the economy.

Lira (1996) has a similar investment equation, which is one of the four equations he
estimates to trace the relationship between output growth and inflation. The first equation shows
the logarithm of the consumer price index (LCPI) to be positively related to the moving average
 3

of money supply (LMAMS), a moving average of wages (LMAW), a moving average of the
price of imported inputs (LMAPM), and the past level of CPI (LCPI[-I ]). The variables are all
in logarithms. On the other hand LCPI is negatively related to the real interest rate (TBCPI) and,
rather surprisingly, to the rate of capacity utilization (CAPU). In equation form we have

+ + + + -

LCPI -- f(LMAMS, LMAW, LMAPM, LCPI[-1], TBCPI, CAPU) (1)

The second equation shows the wage rate as proxied by the compensation of non-
agriculture workers (LNQSE, which is a component of W in equation [1]) to be negatively
related to the unemployment rate (UNEMPR) and positively related to a moving average of the
rate of capacity utilization (MACAPU) and the past level of inflation which captures indexation
(LCPI[-1]). In equation form we have

- + +

LNQSE = f(UNEMPR, MACAPU, LCPI[-1 ]) (2)

Meanwhile, the unemployment rate is found to be negatively related to a moving average

of the capacity utilization rate and the lagged unemployment rate, thus:

UNEMPR = f(MACAPU, UNEMPR[- 1]) (3)

Finally, we have the investmem equation where the logaritlm_ of gross domestic capital
formation (LGDCF) is positively related to a moving average of GDP in logarithmic form and
negatively to inflation. The specified equation is:

LGDCF = f(LNMAGDP, INFL) (4)

All these equations are estimated using quarterly data for the period 1981-1994.

The first three equations combined will yield a short-rma Phillips curve. Higher capacity
utilization resulting from higher growth will lead to lower unemployment, higher wages and
higher inflation. In practice the negative effect on the price level exerted by higher capacity
utilization does not offset the impact of the higher wage rate.

Lira suspects that the negative coefficient of the variable CAPU in equation (i) is due to
simultaneity, and is a reflection of the relationship between inflation and investment in equation
(4). Note that investment and capacity utilization are directly relate&

In general, when the direction of flow is from higher growth (originating from the supply
side), a trade-offexists between ilfflation and unemployment. But an increase in the inflation rate
brings down output growth via equation (4) and eventually leads to an increase in the
unemployment rate. The relationship between inflation and unemployment is asymmetric and
depends on whether the change is inflation-driven or output-driven.
 4

A simple regression of the real growth rate of per capita GDP (g) on inflation (INFL) over
the period 1970-1995 yields:

g = 3.909 - .241 INFL + .572 g,

(5.20) (4.72)

Adj. R2= .652 D.W. = 1.84

ADF Test Stat = -3.08 MacKinnon 5% critical value = -1.96

Breusch-G0dfi'ey LM test (4 lags): N*R 2 = 5.11 Prob = .27

The figures in parentheses are the relevant t statistics. The results indicate a significant negative
relation between inflation and per capita income growth although this may be caused by the
episodes of stagflation.

To further investigate the relationship between output and inflation, we estimate a model
similar to the one developed by Robert Lucas and Robert Barro and applied by Hanson (1980).
A description of the model and the empirical results are presented in the Appendix.

The results show that only unanticipated inflation or monetary growth has a positive
effect on the growth rate. Eight percentage points of unexpected inflation produce about one
extra percentage point of growth. The results also show that an increase in expected inflation due
to supply related developments (which means that money supply growth remains unchanged)
leads to a decline in output growth rate. But the effect is small, supporting the observation made
by Krugman.

I11. Determinants of Inflation in the Philippines

Knowledge of the factors that significantly affect inflation will enable economic
managers totarget appropriate variables in their effort to maintain price changes at a moderate
level. Most econometric inflation models for developing economies are based on mark-up over
cost equations since external factors are readily incorporated. The latter is essential for open
economies like the Phiiippines.

The first equation of the study of Lira (1996) shows that cost-push and demand-pull
factors are important in the determination of the price level. The price of imported goods has the
highest elasticity, followed by wages and money supply.

Reyes (1996), using a monthly inflation model, has similar findings. "I'he price index for
imports of non-fuels has the highest elasticity, followed by the liquidity variable and wage. The
study of Reyes is an extension of the model of Mariano (1985). In the latter study, the variables
with the largest elasticities are the liquidity variable, the import price index for non-fuels and the
wage variable. The equation of Mariano shows, however, that food prices are also a significant
determinant of inflation. The models of Lim, Reyes and Mariano use the price level and not
inflation per se as the dependent variable.

An earlier study is that ofR. Bautista (1983) who estimated an inflation equation based
on the annual growth of CPI. His results show that a large part of inflation for the period 1965-
1982 can be attributed to foreign price increases and the depreciation of the peso. He does not
find the growth of money supply and changes in the wage rate to be significant.

Meanwhile, a further refinement to the studies of inflation is the empirical model of C.

Bautista (1991). By estimating a macroeconomic vector autoregression model (VAR) and
applying variance decomposition, Bautista shows that the forecast error variance of inflation due
to exchange rate movements is higher than that of money supply growth. This lends support to
the BOP view of inflation which maintains that the exchange rate is the main cause of price
changes. The empirical results do not show, however, that the fiscal view of inflation, which
contends that the main determinant is money supply growth fuelled by fscal deficits, is
unimportant.

An interesting study is that of Lim (1987) which is based on the possibility that the
working capital cost-push effect may offset the monetarist effect so that inflation may even rise
after a reduction in money supply. The increase in inflation is caused by the rise in the interest
rate which raises the cost of borrowing for working capital. The empirical results indicate a
positive relation between relevant interest rates and inflation. Lim then concludes that "the
simple quantity theorjy of money is oversimplified and hides the fu!l impact of monetarist
prescriptions to inflation and that it neglects the transmission mechanism of credit madmonetary
cutback which may entail a drastic fall in income, investments, personal consumption
expenditure and most likely, government spending."

These studies show that adherence to the orthodox IMF prescriptions do not bring about
the desired results. Inflation control is more effective through exchange rate stabilization and
less import dependence. Moreover, contractionary monetary policy by raising interest rates and
increasing the cost of working capital only exacerbates inflation.

IV. Policy Regimes: Stabilization and Output Growth

The economic structure of the Philippines has changed very little over the past two and
a half decades and portrays much of a Latin American prototype economy: a dualistic structure
in which a small proportion of the labor force is employed in a capital-intensive and highly
protected manufacturing sector, while the larger share of the population is employed in typically
low-productivity agriculture and urban informal services.

Over the past twenty-five years there has been a continuous struggle to maintain
macroeconomic stability with growth being alternately limited by a foreign exchange constraint,
a savings constraint and a fiscal constraint. The period 1970-79 is crucial since it was at this time
that the Philippines accumulated a huge external debt burden which dictated the structure of
economic policy in subsequent years. This is the reason this.period is included in the analysis.
 The following is a narrative of policy regimes from 1970-1995 based on Vos and Yap
(1996). The basic data are shown on Table 2 while a summary is given in Table 3. The general
theme is that economic managers, consistent with the econometric evidence, sought to control
inflation by containing exchange rate movements and restricting money supply.

U0balanced growth in the 1970s

Stagnation in the process of structural transformation might conceal the economic growth
and changes that did take place in the 1970s and 1980s. The economy grew at an average rage
of over 5 percent during the 1970s. This reasonable growth performance was largely fuelled by
a protectionist incentive structure, a spillover of the import substitution program in the 1950s,
and the massive support from public investment in infrastructure and energy financed largely by
foreign borrowing during the 1975-1979 period.

Based on external shocks affecting the economy, economic performance in the 1970s can
be meaningfully divided into two sub-periods: 1970-74 and 1975-79.

19 70- 74 (Consolidation phase)

This period is characterized by adjustment to the balance-of_payments and short-term

external debt crisis which was built up during the late 1960s. An authoritarian regime is
established which implements a macroeconomic stabilization program. Recovery is supported
by a boom in commodity prices for Philippine exports towards the end of this period, leading to
high growth rates by historical standards. Inflation accelerates to an annual average of 18.8
percent and is caused by cost-push factors (a major devaluation and the first oil price shock).

19 75- 79 (Period of debt-led growth)

This period begins with adjustment to the first oil price shock and falling prices of major
Philippine export commodities, reflected in the deteriorating terms-of-trade starting in 1975.
The supply-led external finance boom provides the required adjustment finance and allows an
investment expansion led by import substituting industries and public infrastructure and energy
programs. Real GNP growth reaches 6.1 percent per annum.

Manufacturing activity has the highest growth rate. The period's exceptional growth is
also fuelled by an export-oriented program, which operates under a protectionist incentive
structure and support from the expansion in public infrastructure. Encouraged by successful
changes elsewhere in the region and sector support of multilateral agencTes (the World Bank, in
particular), specific government programs are set up to promote nontraditional manufactured
exports. The program is successful to the extent that the share of manufactured _xports increases
from seven percent in 1970 to 75 percent in 1992. Deeper analysis shows that this export success
is largely illusory : for instance, exports of semi-conductors contribute more than 10 percent of
total export earnings, but less than 0.5 percent of total value added and around 0.1 percent of
employment. Such stylized facts also apply to garments exports. Due to the high share of
 imported inputs in both sectors, backwa::d linkages with the rest of the economy are small,
limiting the diffusion of the sectoral growth dynamics.

Agriculture growth, though below the perfornaance of the economy, is not unsatisfactory
compared with the perfomaance of the sector in neighboring countries, reaching an all-time high
often percent growth in 1976. Agricultural performance is aided by governnlent support in the
form of rural infrastructure and improved input supplies (fertilizers, credits and high-yielding
seeds) in a number of important subsectors (especially rice) and regions.

The period also witnesses the aggressive expmlsion of the public sector mainly through
newly created public enterprises. The consolidated public sector deficit reaches an all time high
of 9.3 percent of GNP in 1976.The massive expansion of externally financed public investment
is conditioned in part by a program of energy-saving, effective import substitution and, in part,
by a progrmn of political consolidation by the Marcos regime. Some analysts contend that public
enterprises were used as a conduit of foreign loans which would otherwise have not been
contracted.

GNP per capita almost doubles from US$ 336 in 1974 to US$ 586 in 1979; but this
growth does not seem to trickle down in equal shares to all population groups. For the
agricultural sector, relative prices (internal terms of trade) appear favorable at the start of the
period, but decline towards the end of the decade. Data on real wages are scarce and have to be
taken with caution, but available information on the real wage index for unskilled workers
suggests a declining trend since the beginning of the decade. Falling terms of trade for
agriculture, declining real wages, and employment growth thlli_g behind output growth
(particularly in industry) underlie the shift in income distribution toward urban profit incomes
during this period of moderate to rapid GNP growth. Since the industrialization process is
concentrated in the domestic market and basic consumer goods, this shift in income distribution
signals an unbalanced sectoral growth path likely to run against demand constraints similar to
industrialization experiences in Latin America.

Crisis and Adjustment in the 1980s

Adjustment to another set of external and internal shocks mark the three subsequent sub-
periods: 1980-82, 1983-85, 1986-90.

1980-1982 (Pre-crisis stage)

This "pre-crisis' period is characterized by coinciding internal a_d external shocks.

External shocks are, in seque_ ce, the second oil price shock, the steep increase in world interest
rates, recession in the industrialized countries, and an additional (short-term) loan-gupply shock.
The domestic shock results from the external finance boom of the previous period and the
subsequent excessive financial leverage of public and private firms. A World Bank-supported
financial liberalization program pushes up domestic interest rates and provokes liquidity
problems for many firms and banks, eventually leading to an internal financial crisis in 1981.
The goverrtlnent's attempt to selectively bail out favored, distressed firms leads to a severe loss
of private investment confidence and massive private capital outflows.

The economy slows down during this period to m: average growth rate of 3.6 percent; real
GNP growth rate drops to 1.9 percent in 1982. The downward trend occurs despite the rise in
public sector deficit from 3.4 percent of GNP in 1979 to 7.3 percent in 1981 in an attempt to
counteract the impact of the external shocks.

1983-85 (Crisis stage)

A major balance of payments crisis emerges as terms of trade continue to fall and
available external finance contracts, a consequence of the crisis in the international financial
system following the default of major debtor countries. Adjustments to the crisis involve a
moratorium on debt repayments, cuts in public expenditure, and a massive reduction in import
volumes. These lead to a fall in private investment rates. The economy plunges into a major
recession.

The growth of money supply (M3) falls to 12.3 percent in 1983-85 from around 18
percent in the preceding period. Reductions in public expenditures are partly offset by increased
transfers required to keep the struggling public enterprises in business. The CPSD falls from 6.2
percent to 3.2 percent ok"GNP. Industrial capacity utilization drops severely as the allocation of
foreign exchange impinges heavily on import-substituting industries, exacerbated by a high
interest rate policy which attempts to arrest private capital outflows ('capital flight') and
counteract the inflationary effects of devaluation. Real GDP falls by 7.3 percent in 1984.

The manufacturing sector suffers most, showing an 11.5 percent fall in output in 1984
and 15.8 percent in 1985. GNP per capita recedes to the levels reached in the late 1970s, as it
declines from US$ 723 in 1982 to US$ 547 in 1985. The agricultural terms of trade increase,
however, as successive devaluations push up average agricultural prices.

Next to the strong, mainly expenditure-reducing domestic adjustment, there are attempts
to strengthen the traded goods sector. A series of devaluations, along with contractionary fiscal
and monetary policies, result in a depreciation of the real effective exchange rate (REER), which
is 17 percent in 1983 and 4.6 percent in 1984. This gives a boost to export growth and, along
with the severe cuts in imports, creates a current account surplus in 1986.

Nevertheless, the crisis hastens the collapse of the political regime, as all economic
sectors are affected and the domestic coalition supporting the Marcos regime sees its financial
resources (domestic and foreign) dwindling.

1986-89 (Period of Difficult Recovery)

This period starts offwith the internal shock of the collapse of the Marcos regime and the
installation of a new government embarking on a program o£ quick recovery by pump-priming
 9

current public expenditures while cutting public investment. External conditions are slightly less
adverse than in the previous two periods, while debt rescheduling and new sources of external
funding (official bilateral and multilateral flows) provide adjustment finance. Private and public
demand expansion mid enhanced import capacity stimulate a recovery of private investment. The
sharp reduction in political uncertainty and worldwide acclaim for the peaceful transition also
provide a boost.

Other external developments that aid the recovery are the fall in world prices of crude oil
in late 1987, recovery of coconut prices, and the increase in the US sugar quota to the Philippines
in 1986. Growth rebounds to over 5 percent. Because of the low inflation rate, the REER
depreciates sharply_in 1986 but fluctuates in a narrow band the rest of the period.

Unfortunately, the new government does not take advantage of its international popularity
and immediately agrees to honor external debt obligations. This leads to a net capital outflow
of $7.7 billion between 1986 and 1991.

This is one reason the euphoria is short-lived. The economy reaches full capacity in 1988
and supply bottlenecks emerge. This is reflected in an acceleration of the inflation rate from 3.8
percent in 1987 to 10.6 percent in 1989. Another domestic shock in the forn_ of the most serious
coup attempt in 1989 shakes business confidence. Structural economic problems remain, thereby
fuelling the sense of political instability. GNP per capita recovers to US$ 700 in 1989, though
it is still below the level reached at the beginning of the decade.

Recession and Recovery in the 1990.8

1990-92 (Recession and Fiscal Constrain 0

Another balan e of payments crisis ensues towards the latter part of 1990, hastened by
the rise in oil prices foliowing the Iraq invasion of Kuwait. Economic recovery toward the end
of 1991 is predicted, but this does not materialize because of a severe energy crisis that can be
traced to a fiscal constraint with public investment in infrastructure cut back to well below two
percent ofG?,_P for most of the 1980s. To make matters worse, the biggest volcanic eruption this
century occurs in mid-1991. GNP growth in this period averages a mere 1.8 percent, but staying
below one percent in 1991-92.

The Aquino administration pursues trade liberalization measures with more vigor. AFTA
takes center stage in 1992. A key policy measure is the liberalization of all regulations governing
foreign exchange transactions, which takes effect in the last quarter of 1_I92.

The peso depreciates sharply in 1990 but recovers lost ground in 1992 after real interest
rates rise again. The net result is an appreciation of the REER.

A new government is elected in 1992, promising drastic measures to address the energy
crisis but it continues to paralyze economic activity throughout most of the year. In the island of
 10

Luzon, where 80 percent of industrial output is produced, power outages are a daily experience.
Economic perfonnance is a far cry from the 6 percent growth target the Philippine govermnent
had set out for this period. This target is based on assumptions about adequate external finance,
which has not been as forthcoming as expected, but obviously this is but one condition which
cannot compensate for all the structural problems of the Philippine economy.

1993-95 (Economie Reco very an d RefornO

The energy crisis is finally resolved spurring economic recovery. The economic plattbma
of the Ramos administration crystallizes. Trade liberalization, deregulation and privatization
programs are implemented. Among the affected sectors are telecommunications and shipping.
Profitable state enterprises and prime gc vernment real estate are sold to the private sector. The
high points are the approval of the GATT accord in late 1994 and the issuance of an executive
order in July 1995 further reducing tariffs and setting a timetable for the eventual realization of
uniform rates.

With a more liberal environment, capital flows pour into the country. While these are
mainly speculative, the foreign exchmage constraint is relaxed as reserves reach a historical high.
The peso appreciates sharply in 1994 but depreciates as the Central Bank relaxes monetary
policy. On the whole, the REER continues to appreciate in this period. Meanwhile, proceeds
from the privatization program ease the fiscal constraint. What impedes more rapid investment
is the decline in domestic savings relative to GNP.

The administra';on proudly announces that per capita GNP surpa rses the $1,000 tlu'eshold
in 1995, three years ahead of schedule. However, no mention is made of the paradox that real
GNP growth has been constantly below target in the past five years which only means that the
currency is overvalued--a combination of inflation above target and an exchange rate below
target. With tax effort still below the ASEAN norm, the question is whether growth will be
sustained or the Philippines will go the way of the Tequila syndrome.

V. The Costs of Disinflation

Evidently the Philippines is one country which has been committed to the Washington
consensus. When B(_P crises ensue, the response is to devalue the currency and apply
contractionary fiscal and monetary policies. Thereafter, efforts are focused on stabilizing the
exchange rate to ease inflationary pressure.

Despite claims by the Philippine Central Bank to the contrary, the exc"hangerate has been
a primary target of monetary policy as shown by Gochoco (1992). By comparing the volatilities
of money aggregates and the exchange rate, she confirms that the defense of the Cxchange rate
has been the overriding objective of monetary policy. Relatedly, Reyes and Yap (1993b) find
no causality between money and prices during the period 1981-1992.
 11

While the inflationary consequences of a devaluation is a major factor, another

consideration for a stable exchange rate is the performance of the industrial sector which remains
relatively inward-oriented. A more compelling reason though in the period aJ:er 1985 is the
internal transfer problem wherein the government owes the bulk of the external debt (about 80
percent) but has to source the foreign exchange from the private sector. A strong peso, of course,
mitigates the costs of repaying the external debt.

The consequence of tiffs exchange rate policy is an overvalued currency which has been
well documented (Medalla, et al. 1995). This can be also be observed by comparing the
movement in the REER of the Philippines with other countries in the region, some of which are
our closest competitors (Table 4). During the period 1970-993 Malaysia and Thailand had an
average real effective exchange rate depreciation of 2.7 percent while that of Indonesia was 4.6
percent. In the case of the Philippines, the average was only one percent. This is rather
surprising considering that the other countries had sounder fundamentals in terms of external
balances, particularly in the last ten years.

An overvalued currency has several adverse effects. Table 5 shows that the ratio of
tradeables to nontradeables value added has been falling since 1979. The price ratio oftradeables
to nontradeables has also declined since 1987. As Montes and Lim observe: "The continuing
shift in relative prices that makes real estate, financial speculation and retail trade (economic
activity in new shopping malls) attractive vis-_-vis the industrial and tradeable sectors is a
disturbing trend."

Another effect nfan overvalued currency is a widening trade deficit (Table I), which
reached 12 percent in i994 and 1995, the highest level at least for the past twenty-five years.
The economy's propensity to import has been increasing (Table 6) while export performance has
not kept pace. The ratio of exports to GDP even declined in 1992. In the past decade,
remittances from overseas workers prevented the problem of a widening trade deficit from
turning into a full-blown crisis. It should be apparent that dependence on OCW remittances
cannot continue for long and measures must be implemented to increase exports (with the
required backward linkages) and reduce import dependence.

The conservative debt management strategy of the Philippine government also reduced
the flexibility in fiscal management. The task of maintaining the public sector deficit at
manageable levels (to restrain inflation, of course) was made more difficult because total national
government expenditures began to increase, with the expansion largely driven by interest
payments on national government debt. From a six percent share of total national govermnent
spending in 1980, this item peaked at 35 percent in 1988 before settling at 31 percent in 1991
(Table 7). While the yield and elasticity of tax revenues increased, it was not enough to prevent
the development of unsustainable budgetary trends.

Table 7 shows that as a share of total national government outlays, Maintefiance and
Operating expenditures (MOE) along with infrastructure spending were the categories that
experienced the largest cuts. As a result of these reductions, public investment on a per capita
basis declined by almost 50 percent between 1981 and 1991 (Table 8). Meanwhile, total public
 12

sector expenditure on a per capita basis fell by slightly more than 31 percent during the same
period.

The rather myopic fiscal policy caused the disruptive power outages in the early 1990s.
Moreover, infrastructure has been deteriorating rapidly and new projects have not kept pace with
economic expansion and population growth. This has created supply bottlenecks in many
sectors of the economy. Lira (1995) presents diagrams showing that per capita length of roads
and per capita length of bridges have been falling in the 1980s mid early 1990s. The severe
shortage in the supply of rice in the latter part of 1995 is related to this dearth in infrastructure.
While weather conditions are the proximate cause, the deficiency in in'igation facilities is the root
cause behind the struct_ral decline in our rice-producing capability.

Tight monetary policy also has potential negative effects as seen from the study of Lim
(1987). Aside from their inflationary impact, high interest rates have also taken their toll on
private investment. As seen from Table 2, private investment as a share of GNP has not reached
the 25 percent threshold recommended by the World Bamk. Moreover, periods of low money
growth are associated with low GDP growth and relatively high inflation (Table 1), supporting
the hypothesis of the working capital cost-push effect.

VI. Conclusion

The Machiavellian approach to controlling inflation has not been effective in the
Philippines. Certainly, Krugman's warning about the serious repercussions of some methods of
disinflation is quite relevant especially when juxtaposed against the potential costs of inflation.

Financial managers responded appropriately during the crisis periods but were rather
short-sighted during the recovery phases paz-ticularly during the period 1986-95. For one thing,
a more aggressive debt management strategy shouid have been adopted. But even with the
"honor flaydebt" commandment in place, monetary, fiscal and exchange rate policy should not
have been formulated in a manner that sacrifices long-term growth.

Definitely, an inflation of 10-15 percent could have been tolerated as long as the source
was productivity-enhancing government expenditure. A steady depreciation to maintain a
realistic real exchange rate could have been accommodated in this scenario.

As it stands now, economic managers have to contend with volatile capital flows and as
a result, the wisdom of liberalizing the capital account ahead of the trade sector is being
questioned. To prevent further appreciation of the real effective exchange rate in the presence
of these capital flows, it has been proposed that instead of targeting monetary ag_regates, the
Central Bank targets interest rates and an appropriate real exchange rate (De Dios, 1995).
Nevertheless, the key to macroeconomic stability is strong fiscal performance via a higher tax
effort. This is what preoccupies the government at present and the outcome of the tax reform
program still hangs in the balance.
 13

Finally, it should be mentioned that one prominent source of inflationary pressure that
did not figure in the discussion is the oligopolistic structure of the economy. This may well be
the major impediment to economic development but this is a subject best left to other studies.

References

Bautista, C. C. (1991), "Sources and Variability of Inflation in an Open Economy." University

of the Philippines School of Economics Discussion Paper 9115.

Bautista, R. M. (1983), "Determinants of Inflation in the Philippines." University of the

Philippines School of Economics Discussion Paper 8309.

Benabou, R. and J. D. Konieczny (1994), "On Inflation and Output with Costly Price Changes:
A Simple Unifying Result," American Economic Review 84 (1), March.

Danziger, L. (1988), "Costs of Price Adjustment and the Welfare Economics of Inflation and
Disinflation," American Economic Review 78 (4), September.

De Dios, E. S. (1995), "On Recent Financial Flows: Causes and Consequences" in Fabella and
Sakai (eds.)

Fabella, R. S. and H. Sakai, eds. (1995) Towards Sustahwd Growdt. Tokyo: Institute of
Developing Economies.

Fischer, S. (1993), "Macroeconomic Factors in Growth." Paper presented at the World Bank
Conference on "How do National Policies Affect Long-run Growth?".

Gochoco, M. S. (1992), "Targets, Instruments, and Monetary Policy in an Open Economy: A

GARCH Application." Working Paper NO. PB92-04 Federal Bank of San Francisco,
Center for Pacific Basin Monetary and Economic Studies.

Hanson, J. A. (1980), "The Short-Run Relation Between Growth and Inflation in Latin America:
A Quasi-Rational or Consistent Expectations Approach" American Economic Review,
70 (5), December.

Krugman, P. (1995), "Dutch Tulips and Emerging Markets" Foreign .._ffairs, 74 (4),
July/August.

Lim, J. Y. (1987), "The New Structuralist Critique of the Monetarist Theory of Infl_ion: The
Case of the Philippines," Journal of Development Economics 25.

Lim, J. Y. (1995), "Structural Shifts in Key Macro Parameters: Implications on the Sustainability
of Growth" in Fabella and Sakai (eds.).
 14

Lim, J. Y. (1996), "On the Question of a Tradeoff Between Sustainable Growth and Price
Stability" Manuscript.

Mariano, R. S. (1985), Forecasting Monthly bsflation in the Philippines. PID_7Monograph

Series No. 10.

Medalla, E. M., G. R. Tecson, R. M. Bautista, J. H. Power and Associates (1995), Catching

Up with Asia's Tigers Volume II. Makati: Philippine Institute for Development Studies.

Montes, M. F. and J. Y. Lim (1996), "Macroeconomic Volatility, Investment Anemia and

Environmental Struggles in the Philippines," World Development, 25 (2).

Naish, H. F. (1986) "Price Adjustment Costs and the Output-Inflation Trade-off," Economica
53 (210), May.

Reyes, C. M. (1996), "An Empirical Analysis of the Detemlinants of Inflation in the Philippines"
Manuscript.

Reyes, C. M. and J. T. Yap (1993a), "Re-estimation of the PIDS Macroeconometric Model"

Manuscript.

Reyes, C. M. and J. T. Yap (1993b), "Money and Prices in the Philippines, 1981-1992: A
Cointegration Analysis," Journal of Philippine Development, Volume XX, No. 1.

Rotemberg, J. (1982), "Monopolistic Price Adjustment and Aggregate Output," Review of

Econonzic Studies, 49 (October).

Rotemberg, J. (1983), "Aggregate Consequences of Fixed Costs of Price Adjustment," American

Economic Review 73 (3), June.

Vos, R. and J. T. Yap (1996) The Philippine Economy: EastAsia's Stray Cat?--Structure,
Finance and Adjustment. London: Macmillan.
 Table 1 15
Selected Macroeconomic Variables
1970-1995
• im i I i i i i ,i

Year Inflation GDP Money Exchange Trade 91-day External

(%) growth supply Rate Balance/ T-bill Debt
rate growth (P/S) GDP (%) (m US$)
(%) (%) (%)
I I I III I I I I I I II

1970 15,00 3.76 4.92 6.02 -0.43 13.14 2297

1971 21,59 5.43 11.71 6.43 -0.67 11,95 2393
1972 8,41 5.45 13.16 6.67 -1.55 11.92 2732
1973 1638 8.92 52.15 6.76 2.88 9.43 2886
1974 34,07 3.56 17.44 6,79 -3.03 10.05 3755
1975 7.18 5.56 36.21 7.25 -7.82 10,34 4939
1976 8.76 8.81 24.26 7.44 -6.20 10.19 6768
1977 9.95 5.60 22.37 7.40 -3.89 10.9 8069
1978 7.33 5.17 18.01 7.37 -5.76 10.89 10694
1979 17.67 5.64 10.65 7.38 -5.61 12.25 13352
1980 18.09 5.15 18.20 7.51 -5.97 12.14 17252
1981 13.29 3.42 21.08 7.90 -6.24 12.61 20893
1982 10.20 3.62 16.06 8.54 -7.12 13.81 24677
1.983 9.95 1.87 19.90 11.11 -7.47 14.17 24816
1984 50.32 -7.32 7.20 16,70 -2.16 30.53 25418
1985 23.11 -7,31 9.87 18.61 -1.57 26.81 26252
1986 0.80 3.42 7.31 20.39 -0,68 14.43 28256
1987 3.72 4.31 12.12 20.57 -3.06 11.39 28649
1988 8.81 6.75 22,57 21.09 -2.86 14.06 27915
1989 12.20 6.21 27.98 21.74 -6,10 19.33 27616
1990 14.17 3.04 18.36 24.31 -9.07 23.4 28549
1991 18.66 -0.58 15.49 27.48 -7.07 21.35 29956
1992 8.95 0.34 11.04 25.51 -8.86 16.12 30934
1993 7.61 2.12 24.55 27.12 -11.44 12.25 34282
1994 9.03 4.42 26.58 26.42 -12.24 13,62 37698
1995 8.08 4.84 25.32 25.71 -12.09 11.35 40595 /a

Memo Item:
Average inflation for September 1995 to May 1996 is 11.2%..
/a - as of June 1995.
i i _.

Sources: Philippine Statistical Yearbook, various years.

Bangko Sentral ng Pilipinas.
 Table 2
SelectedEconomicIndicators of the Philippines, 1970-1995 16

Indicator 1970-74 1975-79 1980-82 1983-85 1986-89 1990-92 1993-95

I I II II "-- I II

I. Income
Per Capi';aRe_! Income,pesos 9831 11589 12669 _"328 10871 11461 11525
Real GNP (% growth) 5.7 6.1 36 -4.9 5.5 2.3 4.1
Real GDP 5.4 6.2 4.1 -4.3 5,2 0.9 3.8
Agriculture 2,8 4.5 2.8 -2.1 33 0,7 1.9
Industry 8.0 7.9 4 -8.9 5.8 =0.2 4.9
Services 5.0 5.4 4,9 -1.1 5.6 2.0 3.9

Investment( %shareto GNP), real

Private 14.89 18.1 20.49 16.75 15,09 17,03 18,41
Public 1.97 5.90 5.89 6.35 2,96 3.91 3,97

Real GDP (% shares)

Agriculture 27,4 24.5 23.3 23_2 23,9 22_6 22_2
Industry 35,3 39.6 40.7 38.0 35.2 34.9 34.8
Services 37,4 35.9 36 38.8 40.9 42,5 42.9

I1. ExternalSector
Degreeof Openness(% of GNP) _a 34.3 37.3 38_3 36_1 39,0 45,3 54,1
CurrentBalance/GNP(%) 0.7 ' -5.3 -6.8 -4,1 -0,6 -3.1 =4.0
BOP/GNP(%) 1,6 -1.2 -2.4 0.7 2.0 2.4 1.1

Termsof Trade (index) _b 97,6 72.9 70.8 82.1 85.7 69.4 64.6
Real EffectiveExchangeRate(index) _c 98,9 96.8 102.5 86.4 68.8 71.2 79.3

III. Public Sector

PublicSector Deficit/GNP(%) _d n.a. 5.7 6.2 3.2 2,6 2.2 0_3

IV. MonetarySector
MoneySupply(M3) (growthrate) 19,9 22.3 18,4 12.3 17,5 15,0 25_5
GrossInternationalReserves, mS 743 1767 2480 937 2207 3894 6814

V, LaborSector
OpenUnemploymentRate(%) 5.6 7,1 8,9 11_2 10,4 8,6 6_6
UnderemploymentRate(%) 13.4 11.8 26.3 30_8 24,8 10.0 9,5
RealWage Index(unskilledurban) 227.3 157.1 119.2 105.2 110.6 105.3 126.0

SectoralEmployment(% shares)
Agriculture 52 52.1 51.6 50.0 47.6 44.0 45.0
Industry 15.8 15.3 14.7 14.5 14.7 15.7 15.5
Services 32.2 32.5 33.8 35.5 37.7 38.8 39.5

VI. Prices
inflationRate(%) \f 18.8 9,9 13.4 26_8 5.9 13.9 8.2
GDPdeflator(growthrate) 17,4 10,0 11_4 28_4 7.3 12.5 8.0
WholesalePriceindex(growthrate) 25,7 8,9 14.6 33_7 7.8 9.4 3.4
InternalTermsof Trade(% change) _ 7.0 -3.1 -7.2 1_4 -2.2 -13,1 22,3

VII Population(rail,, end of period) 41.1 47 50.8 54.7 60.1 63.7 68.6
Populationgrowthrate(%) 2.8 2.7 2.6 2.5 2.4 2.8 2.5

VIII Income Distribution

PovertyIncidence n.a. n.a. n.a. 44.2 40.2 39.2 35.7
GiniRatio n.a. n.a. n.a. 0.4466 0,4446 0,4680 0,4507
No,of familiesbelowpoverythreshold(m) n.a. n.a. n.a. 4.355 4,230 4.781 4,559

IX GNP per capita (US$, end of period) 336 556 723 5_." 698 "" 1090
Realpesosof 1995 ( endof period) 10465 12329 12868 10461 11638 1lC',. 1.1433

Notes:

a/ Definedas [(Imports+Exports)/GNP] in constantprices.

bl Relativepriceof unitpriceindexof merchandiseexportsoverunitpriceofimports.
c/ Tradeweightedrealexhchangerate;monthlyaverage,December1980=100,
d/ Includesgeneralgovenrment,state*monitored corporationsandCentralBank.
e/ Definedas occupiedworkersworkinglessthan40 hoursperweek,
f/ Pricesbasedon 1986=100,
gl Rateof implicitGDPdeflatorof agriculture
to thatofnon-agriculture.

Sources:
NationalIncomeAccounts,variousyears.
PhilippineStatisticalYearbook,variousyears,
FamilyIncomeandExpenditureSurvey,variousyears.
BangkoSentralng Pilipinas.
 17
Table 3
Major Economic Shocks and Policy Responses in the Philippines
1970 - 1995
t

1970-74 1975-79 1980-82 1983-85 1986-90 1990-92 1993-95 IB

li4_,, I I III I I

F;:ternal Foreign Cheap Oil Sto_Jpage N_'_ti_.,_d Gulf Surge in speculative '
Shock debt foreign pdca of foreign debt war capital inflows
crisis credit . shock debt rescheduling
inflow inflow Sharp rise in OCW
Commodity Increased Recession remittances
price boom foreign abroad Resumed
investment multilateraland
Oil price Highwodd bilateral loan
shock Deteriorating interest rates inflows
terms of
trade Restricted Net resource
foreign credit outflow due to
debt payment

Domestic Rise of strong Dewey Dee Assassination Takeover of Volcanic Mid-term elections
Shock nationalist financial of Aquino Aquino gov't, eruption,
movement/ crisis energy crisis, Rice and sugar
Martial Law Coup attempt Presidential crisis
declaration election
Concern over
criminality

Monetars Tight due to Expansionary, Highly Restrictiveand Expansionary Tight, lower real Established an
Policy inflationary subsidized expansionary deflationary in eaflieryears; interest rate independent
pressures from credit to and counter- high interest tight with high due to inflation Central Bank
devaluation priority areas cyclical, rates interest rates in surprisebut
financial later years recovery in 1992 More relaxed
liberation policy

Fiscal Tight due to Expansionary, Counter Contractionary, Expansionary Tight, cutback Large scale
Policy inflationary particularlyon cyclical concentrated on in earlier years operation and privatization
pressures from government debt service but became expenditures
devaluation investment and bailout of tight with on capital NG surplus
goverment ._s concentration outlays
corporations on domestic
borrowingsand
tax reforms

Trade and Devaluation, Export Removal of Suspensionof Trade Forex GAFF accord
Industry export promotion, QR's trade lib, liberalization, liberalization,
Policy promotion protection of taxation of focus onAFTA, Timetable for
import tradeables, . sharply uniform tariff
substitute, rationing of appreciating
failed plan for foreign exchange peso, emphasis
heavy devaluation on energy
industrialization projects

Combined High growth, High growth, Slow growth, Deep economic Economic Recession, high Economic recovery
Effects high inflation shift to non- inflation recession, recovery up inflation in 1991
traditional high inflation to 1989, Large increase in
exports increasing Peso exports
current account appreciation
deficits _-!.,ctuationsin
9

Private Favorable, Favorable, Unfavorable, Collapse of Renewed Renewed Investment still

Response high investments high savings reduced savings, business confidence _ confidence due disappointing but
and investments capital flight but confidence initially, erratic to peaceful signs of greater
continued behavior in later transition of confidence
investments spurred years power, sharp
by gov't pump- decline due to
priming energy crisis

Note: Summary of key events is based on Table 3.1 of M. Lamberte, J. Lim, R_Vos, J. Yap, E. Tan, and S. Zingapan.
PhilippineExternal Finance. Domestic Mobilizationand Development in the 1970s and 1980s, PIDS: Makati, 1990.
 18
Table 4
Comparison of Real Effective Exchange Rate
Selected Countries
(1980=100)

Years Philippines Taiwan Malaysia Thailand Indonesia

IIIll_Jlll L I I Jill Ill Ilml Jl

1970 96.35 91.01 132.54 130.11 115.85

1971 100,32 86.09 129.70 122.04 101.95
1972 93.45 78.81 130.51 107.41 8933
1973 93.22 77.92 133.02 104.67 97.39
1974 111.20 105.27 137.40 116.67 121.05
1975 102.20 100.20 129.35 109.05 117.82
1976 100.27 96.74 118.74 108.99 129.30
1977 97.63 92.10 112.29 102.47 128.27
1978 88.10 85.94 103.04 88.36 110.41
1979 95.94 90.99 104.89 92.53 91.50
1980 100.95 99.14 103.81 101.49 98.34
1981 101.32 107.11 102.51 104.23 99.50
1982 105.18 102.73 110.66 109.91 114.76
1983 86.91 96.41 111.82 109.51 85.70
1984 82.93 94.25 113.04 100.97 83.26
1985 89.41 89.27 104.29 87.15 86.28
1986 71.45 82.15 82.77 74.31 63.84
1987 67.11 91.32 77.04 69.50 41.46
1988 66.04 95.82 70.46 68.19 40.02
1989 70.71 107.51 70.35 67.76 41.32
1990 69.20 105.01 68.70 70.82 41.35
1991 68.18 103.24 65.83 70.45 40.01
1992 76.37 109.34 70.71 69.38 38.81
1993 74.65 103.91 70.58 68.85 38.89

1970-74 98.9 87.8 132.6 116.2 105.2

1975-79 96.8 93.2 113.7 100.3 115.5
1980-82 102.5 103.0 105.7 " 105.2 104.2
1983-85 86.4 93.3 109.7 99.2 85.1
1986-89 68.8 94.2 75.2 69.9 46.7
1990-92 71.3 105.9 .... 68.4 70.2 40.1

Source: Bangko 8entral ng Pilipinas.

 19
Table 5
Comparison of Tradeable and
Non-tradeable Sectors

Ratio of Ratioof
Value added Price Index
Year (T/NT) (TINT)

1970 64.166 0.909

1971 63.082 0.891
1972 64.095 0.877
1973 70.446 0.914
1974 66.274 0.947
1975 63.599 0.962
1976 62.575 0.947
1977 63.911 0.963
1978 65.723 0.984
1979 65.997 1.007
1980 63.371 0.978
1981 58.686 0.960
1982 54.204 0.941
1983 51.610 0.963
1984 55.955 1.070
1985 55.158 1.000
1986 53.471 0.995
1987 54.658 1.024
1988 54.057 0.998
1989 50.904 0.977
1990 48.846 0.957
1991 48.767 0.969
1992 46.886 0.948
mllll I III

Source: J.A.R. Tan III (1995): "Modelling

Real and Financial Se_or Intera_ion in a
Credit Rationing Economy: Implication on
Central Bank Policy", Dissertation UP
School of Economics.
 2O

Table 6
Analysis of Trade Performance

1991 1992 1993 1994 1995 1Q'96

" II Jl immlll| ...........

1) Ratio of Exports to GDP 19.4 18,5 20,9 21.0 23.5 24.5

2) Ratio of Imports to GDP 26,5 27,4 32.4 33.3 35.8 38,4

3) Ratio of Imports of Capital

Goods to Fixed Investment 32.4 36,3 43.4 45.3 48.5 n.a

4) Ratio of Imports of Consumer Goods

to Total Personal Consumption 3.0 3.1 3.8 4.4 5.1 n.a

5) Ratio of Imports of Raw Materials,

Intermediate Inputs and Fuel to GDP 16.8 16.6 18.1 18.2 19.7 n.a
imll ii

Note • A similar table appears on page 19 of Lim (1995).

Source of Basic Data: Bangko Sentral ng Pilipinas and National Statistical Coordination Board.
 Table 7
Total Government Expenditure
Percentage Share

CURRENT OP_ERATING.EXPENDITURE CAPITAL OUTLAYS TOTAL

Maintenance Interest Personal Others Total Infrastructure Others Total GOVERNMENT
Year and Payments Services EXPENDITURE
Operating
(MOE) (INTPAY) (PERSERV) (OTHERS) (TCOE) (INFRA) (OTHFRS) (TCAPOUT)

1976 39.52 3.65 29.72 4.79 77.68 13.25 9.08 22.32 20338
1977 41.72 3.94 27.61 4.56 77.83 10.36 11.81 22.17 22766
1978 35.72 4.37 28,53 5.33 73.96 14.03 12.01 26.04 26002
1979 30.62 6.36 28.44 5.75 71.16 I4.76 14.08 28.84 28959
1980 28.68 6.13 24.92 5.74 65.47 19.62 14.91 34.53 37443
1981 23.89 5.15 22.55 4.38 55.97 2I. 15 22.88 44.03 47150
1982 26.16 7.06 21.13 8.64 63.00 13.44 23.56 37.00 50392
1983 23.57 9.86 27.39 7.32 68.13 13.71 18.16 31.87 50670
1984 19.85 16.65 26.97 5.13 68.59 10.02 21.39 31.41 82503
1985 16.83 18.68 29.20 5.78 70.48 7.02 22.50 29.52 78424
1986 15.76 22.67 26.21 5.55 70.19 8.18 21.63 29.81 95349
1987 16.73 32.71 28.83 6.37 84.64 6.47 8,89 I5.36 112830
1988 14.96 35.10 31.22 5.66 86.94 6.43 6.G3 13.06 130652
1989 15.93 32.51 30.52 6.98 85.93 7.78 6.29 14.07 168312
1990 14,10 33.19 29.02 7.97 84.29 6.21 9.50 15.71 214248
1991 14.99 30.93 29.98 6.63 82.52 7.29 10.19 17.48 241508

Source: Department of Budget and Management.

 22
Table 8
Public and Private Investment
On per capita basis (pesos in 1985 prices)
(1980-1991)
ij III ,,m, ,,, , ' II I =t

Year Public Public Private

Investment Expenditure Investment
II II I I II IIIIli

1980 722.33 1760.83 2171.76

1981 1036.57 2019.38 2150.14
1982 914.74 1943.19 2367.29
1983 1121.47 2079.19 2370.55
1984 965.91 1845.05 1574.38
1985 552.81 1348.89 1096.29
1986 442.29 1222.04 1168.38
1987 415.98 1214.36 1246.83
1988 418.68 1269.21 1435.69
1989 524.61 1402.40 1730.37
1990 579.46 1517.00 1796.71
1991 525.39 1386.66 1513.98

Public Investment = Public Construction + Public Durable Eqpt.

Private Investment = Private Construciton + Private Durable Eqpt.
Public Expenditure = Pubic Investment + Public Consumption
i i li i

Source: National Income Accounts.

 APPENDIX
A Simple Model of Output Growth and Inflation

I. The Basic Model

Consider the Lucas-Phelps aggregate supply function:

Yt = ao + al(Pt - PtE)+ a2Yt_l+ a_time (1)

where:

Y, - log of aggregate output at time t

Pt - log of price level at time t
PtE- E(log of Pt[Information available at time t).

E is the expectations operator and at the moment we neglect error tenns.

Instead of specifying an aggregate demand function, it is assumed that money demand

adjusts until aggregate demand equals aggregate supply. Hanson (1980) notes that it should be
further assumed that monetary disturbances do not spill over to the world economy because of
variations in the effective exchange rate. A common demand for money equation derived from
the quantity theory and popularized by Harberger is as follows:

P, = Mt - blYt + b2time (2)

where M1is the log of money demand at time t. The relevant opportunity cost of holding money
is omitted for the sake of simplicity. Subsequent empirical results show that this simplification
is valid. The time variable is added in (2) to capture trend elements in money demand,
particularly those affecting velocity such as the development and growing sophistication of
financial institutions, credit facilities, and the like.

We first take expected values of (1) and (2), obtaining:

YtE= a0+ a2Yt.l+ a3time (3)

PtE= Mt E - blYtE+ b2time (4)

Note that in (1) the expected price forecast is zero so that the second term disappears in (3).

Substituting (3) into (4) yields:

PtE= M,E - a0bl - a2blYt.1+ (b2- a3bl)time (5)

Substituting (5) and (2) into (1) yields:

Yt = a0+ a_*(M, - M,E)+ a2Yl._+ a3time (6)

 24

where a_* -- al/(1 + a_b0. (6) is the equation estimated by Barro,

The equation of interest to Hanson is derived by simply substituting (2) into (1):

Yt = ao* + a_*(Mt -PtE)+ a2*Yt.i + c3time (7)

where ao*= ao/(1 + a_bl), a2* = a2/(1 + alb_), and ca = al*b2 + a3/(1 + alb_).

By taking time derivatives we obtain the equation to be estimated for this study, thus:

DYt = ao* + al*(DMt - DPtE)+ a2*DY,.t + c3time + th (8)

where ut is a white noise error term. Hanson estimated (7) mad(8) for various Latin American
countries.

ii. Empirical Results

In order to estimate (8), a proxy for the unobservable DPtz must be obtained. We use the
extrapolative predictorwhere the infomlation set utilized is limited to lagged values of the actual
variable, which in this case is DPr The estimated equation using OLS is:

DPt = .107 + 1.01 DPt_I+ .137 DPt.2 - .504 DPt.3

(5.21) (.52) (2.07)

- .484 DPt.4 + .590 DPt.5 + .020 DPI.6

(2.03) (2.44) (.08)

- .268 DP,.7 - .001 time -.502 DYt.i

(1.55) (2.45) (2.01)

+ .215 DYt.2 + .059 DYt.3 - .005 DYt. 4 (9)

(.80) (.22) (.02)

Adj. R2-- .910 D.W. = 1.98 1983.4 - 1994.4

ADF Test Stat = -2.49 MacKinnon 5% critical value = -1.95

Breusch-Godfrey LM test (4 lags): N*R 2 = 2.98 Prob = .56

The figures in parentheses are the relevant t values. The estimated value of DPt or DPt is then
used as a proxy for DP. Note that we include the time Variable and lagged values of DY to avoid
inconsistency in the second stage of estimation.

Equation (8) is estimated using OLS as:

DY_= .003 + .125 (DMt - DP_r) + .610 DYt.I + .010 DYt.2 "
 25

(2.51) (3.41) (.05)

+ .144 DY,.3 - .170 DYt_4

(.78) (1.18)

Adj. R2= .769 D.W. = 2.10 1983.4 - 1994.4

ADF Test Stat -- -1.96 MacKinnon 5% critical value = -1.95

Breusch-Godfrey LM test (4 lags): N*R 2= 3.01 Prob = .55

The presence of DMt in Equation (8) leads to simultaneity bias and hence the estimates
using OLS are not consistent. Using IV estimation, we arrive at the following results:

DYt = .003 + .123 (DMt- DPtE)+ .614 DYt.I + .010 DYt.2

(2.32) (3.35) (.05)

+. 144 DY,.3 " •172 DYt.4

(.78) (1.18)

Adj. R2= .769 D.W. = 2.10 1983.4 - 1994.4

ADF Test Stat = -1.97 MacKinnon 5% critical value = -1.95

Unfortunately, the LM test is not applicable when OLS is not used. But since the
estimates using OLS and IV are quite similar, then the LM test for the OLS estimation can be
used as an approximation. The results support the hypothesis of no serial correlation up to 4 lags.
Also the unit root test indicates that the residuals are stationary and hence the regression is a
valid one (i.e. it is not spurious).

The time variable was omitted because it turned out to be insignificant. Including a cost
of money variable (in this case the 91-day Treasury bill rate) did not improve the fit of the model.
This justifies the exclusion of this variable from the original model.

The estimate ofa_* is .12, which is quite close to the values obtained by Hanson for the
Latin American countries he studied. This value means that approximately eight percentage
points in unanticipated inflation or monetary growth leads to a one percentage point increase in
the growth rate.

The work of Hanson was criticized for failing to account for fiscal and external variables
in the determination of DPt. The issue boils down to an empirical question and since we obtained
satisfactory results in both equations (9) and (8'), the model is considered to be adequate.