                 A MODEL FOR THE MALTESE ECONOMY


              Marein van Schaaijk and Lino Briguglio





1. INTRODUCTION

The object of this paper is  to apply a macroeconomic model to the
Maltese economy. The  model, called  MACMIC, was  developed by Van
Schaaijk (1991)  and  has already  been  applied to  Suriname. The
model has a overall general framework,  but needs to be adapted to
the take account of the special conditions existing in the economy
which is  being modelled.  For this  reason, the  paper contains a
description  of  the  structure  and  performance  of  the Maltese
economy.

The study is  divided into five  parts. Section 2,  gives an brief
account of  the  structural and  cyclical  changes in  the Maltese
economy since 1960.

The third  section describes  the  MALTA model  and applies  it to
Maltese data covering 1972 to 1990. The equations of the Model are
described individually and estimated.

In Section 4 we carry out a number of simulation exercises to test
the model  and  to forecast  the  future path  of  some endogenous
variables. We  also  introduce  a number  of  exogenous  shocks to
predict what would happen as a result.

Section  5  concludes  the  paper   with  an  agenda  for  further
development of the model  as a tool  for analysing, simulating and
forecasting economic changes in small economies.
2. THE MALTESE ECONOMY


2.1 Malta in an International Context


The Maltese  economy is  one of  the  smallest in  the world.   In
1990,its GNP was approximately US$  21,000 million and the Maltese
population was approximately   350  thousand, with a  land area of
320quare kilometres.

In terms of GNP per capita, however,  Malta is not on the low side
by international  comparisons.  Although usually  classified  as a
developing country, its average 1990  per capita GNP, amounting to
approximately  US$   6000,  is   higher  than   most  third  world
countries,and the World Bank classifies the Maltese economy  as an
uppermiddle income  one. The  UNDP  Human Development  Index ranks
Malta as number 29 in the world.


2.2 A Brief Historical Outline

The most  important    change  that has  occurred  in  the Maltese
economy during between 1960 and 1990 was related to  the expansion
of the  manufacturing  sector  and the  tourist  industry  and the
phasing out the dependency on British military expenditure.

Before 1964, Malta  was a British  colony, and its  economy was to
avery large  extent tied  to the  interest of  Britain. During the
fifties, it became manifestly  obvious that changes in the British
defence policies  were  going  to result  in  massive  rundowns of
employment with the  British services  in Malta, and  the need was
felt to  implement a  co-ordinated  development plan  to diversify
Malta'seconomy.


Between 1959 and  1989, six  development plans were  launched.  An
important objective  of the  plans  was to  increase the  share of
manufacturing, tourism  and agriculture  in the  Domestic Product.
Since Malta's internal market  is very small, industrial expansion
had to  be  sustained  through increased  reliance  on  the export
market,and  therefore   the  importance   of  competitiveness  for
attaining the plan  objectives  was  always stressed.   All  plans
insisted on the need  for adapting of attitudes  and of methods of
production to the changing structure of the Maltese economy.

A satisfactory measure  of success  has been  achieved, during the
past thirty years. This  is evidenced by the  growth in the number
of  the  gainfully    occupied   persons,  the  expansion  of  the
manufacturingsector, the  rise  in  real national  income  and the
phasing out of dependence on British military services.
2.3 Cyclical and Structural Changes

Cyclical Changes

During  the  period  under  consideration,  the  Maltese   economy
experienced a  cyclical pattern  of  change, with  relatively slow
growth rates of GDP during the first half of the three decades and
relatively fast growth  rates during  the second half  of the same
decades.

The fluctuations  in   the  Maltese economy  are reflected  in the
pattern of  change of  real  GDP and  employment, relative  to the
trend over a thirty year period, shown in Figure 1.


Figure 1. The Growth Pattern of the Maltese Real GDP
It can be seen  that the economy performed  at its best during the
second half of   each  decade.   The   most rapid  increase in GDP
occurred  between  1975  and  1979  and  the  fastest  increase in
employment occurred between 1965 and 1969.

Probably  the  most  important  factors  which  accounted  for the
fastgrowth rates of  output and employment  during the sub-periods
just referred to, were the  so-called construction boom during the
second half  of the  sixties,  and the  rapid increase  in tourism
during the  seventies.    These  two  types  of  expenditures have
relatively  low  import  contents  and  therefore  high multiplier
effects on domestic value added and employment.



Structural Changes

The patterns  of change  just described  have been  accompanied by
changes in the composition of GDP.   Table 1  presents data on the
contribution  of  major   sectors  during   the  sub-period  under
consideration. It  can be  seen from  this table  that the fastest
growing sector was manufacturing, which  accounted for just 17% in
the  early  sixties   and  increased  to   about  33%  during  the
lateseventies.

Table 1. Sectoral Net Output  as % GDP at Factor Cost.

Averages for Six Five-Yearly Sub-Periods
________________________________________________________________
                60-64    65-69     70-74   75-79   89-84  85-90
________________________________________________________________
Manufacturing    16.6     20.8      24.3    32.7    30.5   28.0
Cons. & Quar.     4.4      4.4       4.2     2.9     5.0    4.1
Market Services  33.7     33.3      30.4    30.9    34.3   34.2
Agr. & Fishing    7.3      7.2       7.2     5.3     4.1    4.1
Public Sector*   17.9     19.8      21.8    19.8    18.7   21.8
British Forces   15.1      9.5       6.3     2.3      -      -
Property Income   5.0      5.0       5.8     6.1     7.4    7.7

Total           100.0    100.0     100.0   100.0   100.0  100.0
_______________________________________________________________
* This covers public administration and public corporations.
Source: National Accounts of the Maltese Islands.


There  was  a  small  decline  in  the  percentage  share  of  the
manufacturing sector during  the eighties. A  further breakdown of
this sector would  indicate that   it  changed structurally during
the past thirty  years, with  the textile,  clothing and machinery
industries expanding their  relative shares and  becoming the most
important export  oriented  industries  during  the  seventies and
remaining so during the eighties.

The British  military establishments  on  the other  hand, reduced
their share of GDP from an average  of 15% in the early sixties to
zero during the eighties. As   already explained, this was in line
with the development  strategy adopted in  the Maltese development
plans.

The changes in  the shares of  other economic sectors  were not as
dramatic as  those  pertaining  to manufacturing  and  the British
bases. During  the thirty  year period, construction and quarrying
contributed an  average  of 4.2%,  market  services an  average of
32.7% and the public sector an average of 19.5% of GDP.  The share
of agriculture and fishing  averaged  6.1%  but tended to decrease
duringthe 25 year  period.



2.4  Changes in Total Final Expenditure


The most important factor  that emerges from  an analysis of total
final expenditure in Malta is  that the Maltese economy has become
increasingly open  up  to the  seventies,  and remained  very open
during the eighties. It  is now one of  the most open economies in
the world.
The following table shows the percentage of exports and imports in
total final expenditure between 1960 and 1990.


Table 2. Exports and Imports as % of Total Final Expenditure.
              Averages  Six Five-Yearly Sub-Period
_________________________________________________________________
    PERIOD     60-64    64-69    70-74    75-79    80-84    85-89
_________________________________________________________________
C+I+G (*) %    67.10    68.52    67.80    56.43    59.73    59.21
Exports %      32.90    31.48    32.20    43.57    40.27    40.79

T.F.E,(*) %   100.00   100.00   100.00   100.00   100.00   100.00

G.D.P. %       59.43    57.07    54.08    51.14    53.26    53.23
Imports %      40.57    42.93    45.92    48.86    46.74    46.77
_________________________________________________________________
*  C  +  I   +  G    represent   domestic  expenditures,  made  up
ofconsumption, investment and government expenditures.
T.F.E. stands for total final expenditure.
Source:  Annual Abstract of Statistics

Aggregate expenditures in  Malta have tended  to follow a cyclical
pattern of change similar to the  one described above for GDP.  In
general, these  expenditures increased  rapidly during  the second
half of the  sixties, of  the seventies,  and of  the eighties and
increased slowly  or even  declines during  the other sub-periods.
The  type  of  expenditure  that  fluctuated  most  was investment
expenditure.  Exports  of  goods  and  services,  have  tended  to
increase rapidly    during the  second  half each  decade,  and to
decrease during  the  early  sixties  and the  first  half  of the
eighties.

Imports, measured in real  terms, tended to grow  at a slower rate
than total final  expenditure during  the seventies  and the first
half of the eighties.  This  means that the real import content of
total final expenditure  tended to decrease  and the real domestic
value added content  tended to increase  during these two decades.
This tendency  has  been brought  about  by the  policy  of import
substitution and import controls which  have been resorted to with
increased intensity since the seventies and up to 1986. During the
second  half  of  the  eighties,  a  trade  liberalisation  policy
reversed this tendency.

Measured at constant  prices, the  Imports/Total Final Expenditure
ratio showed a  more marked  tendency to decrease  than at current
prices, during the 19701986 period.  The discrepancy  between the
ratio in real  terms and the  same ratio at  current prices arises
because the Maltese overall terms of trade tended to deteriorate.
Like other small countries Malta depends to a very large extent on
imports for  its  consumer and  capital  goods and  its industrial
supplies. It also depends on export  markets to meet its very high
import bill and to  produce on a  sufficiently large scale because
its domestic market is very small.

The overall dependence on  exports and imports,  shown in Table 3,
is further aggravated by the  fact that the Maltese economy relies
on a  very narrow  range of  exports. Malta's  exports are heavily
concentrated on clothing  and electronic components  (about 60% of
merchandise exports and about 30%  of total exports) and on travel
and transportation (about  80% of  services exports  and about 40%
oftotal exports).

Table 3     Main Economic Indicators  Realised Figures
------------------------------------------------------------------
                                    87       88       89       90
         annual percentage changes:
 export price                        3.       5.       2.       3.
 consumption price                   2.       1.       0.       3.
 import price                        2.       1.       3.       4.
 wage per capita in enterprises     -2.       3.       3.       9.
 money                              10.       9.      10.      11.
 volume of exports                  13.       6.      11.      12.
 volume of private consumption       1.       9.       9.       5.
 volume of investment enterprises   17.      17.      -6.       7.
 volume of imports                  12.      11.      11.      13.
 jobs in enterprises                11.       2.       3.       1.
 real disposable wages +benefits     7.       7.       2.       5.
         levels:
 price-costrate exportsector        .99     1.03     1.04     1.04
 unemployment rate                   6.       5.       5.       4.
   in percentage of moneyvalue in year t-1:
 surplus on current account          0.       4.      -2.      -0.
 financial deficit of government     4.       0.       4.       3.
 -----------------------------------------------------------------
3. A MACROECONOMIC MODEL FOR MALTA


3.1 The Model

The present  study utilises  a model  called MALTA  to analyse and
forecast the Maltese economy. It is a macroeconomic model, tailor-
made for small  economies, developed  by Van  Schaaijk (1991). Its
main advantage is that it is  simple and can be run with variables
readily available in the National Accounts.

It consists of a macro-block and a micro-block. The macro-block is
made up  of  equations  explaining the  endogenous  variables, and
utilises  aggregate  variables.  The  micro-block  deals  with the
exportsector, and is  made up of  price, production and investment
equations  for  the  main  export  products.  The  micro-block  is
relatively easier to construct for  small economies than for large
economies, since  in  many  small economies  around  a  very large
percentage (sometimes exceeding 90%) of exports are furnished by a
very narrow range of goods and services. This is the case with the
Maltese economy,where,  as we  have shown,  around 75%  of exports
consist  of  two   manufactured  products,   namely  clothing  and
electronic  components,and   two   services,  namely   travel  and
transportation.

The model makes  use of  70 primary  variables, which  can then be
used to generate  more variables. The  primary variables are first
entered into a  primary input  file of a  spreadsheet programme to
test for the overall consistency. The data handling is perhaps the
most time consuming  exercise in estimating  a model. However, Van
Schaaijk (1991) has constructed  a book-keeping spreadsheet model,
called MACROABC to enter the primary variables and to derive other
variables from the primary ones. The  main purpose of the model is
to test  for  consistency  of  the  variables  in  a macroeconomic
system.   The 70  variables  included in  the Malta  primary input
spreadsheet are shownin Appendix 1.

Once the test for consistency is satisfied, the variables are used
to  estimate   a  set   of  equations   representing  behaviourial
relationship for  the  endogenous  variables  and  identities. The
exportand investment variables used in the macroblock are derived
from the equations of the microblock.

The choice  of  equations  and their  specification  to  an extent
depend  on  the  country  for  which  the  model  is applied.  The
equations specified  for  Malta  are described  in  the  next sub-
section.

Using  a  number  of  behavioural  relations  and  identities, and
plugging in the  exogenous variables  and the  initial (first year
values) of the endogenous  variables, the model  is solved using a
special version  of  SIMPC, which  is  an iterative  programme for
solving a  simultaneous  system  of  equations,  developed  by Don
Econometrics.

A listing of  the programme,  adapted for  the Malta macroeconomic
model, is given in Appendix 3.

The model  can  be  utilised  for    (1)  analysing  past economic
changes, which would help the  policy makers to understand how the
economy works  and  the  relative  sensitivity  of  the endogenous
variables to  external  shocks  (2)  forecasting  and simulations,
which could help the  policymakers to asses  the impact of planned
exogenous changes and (3)  provisional up-dating, which would help
researchers to estimate  variables for  recent years,  or for some
missing years, when published statistics are yet available in this
regard.


3.2 The Specification of the Model

In the case of Malta, fifteen behaviourial equations were utilised
three for the microblock and twelve for the macroblock.

The Micro-block

The micro-block equations should  have represented a disaggregated
export  sector,  but  we  did  not  have  the  time  in  procuring
consistent  data  for  this  purpose  and  analyse  that.  We have
therefore decided  to use  one aggregate  equation as  a temporary
measure, hoping to  remedy this  shortcoming in  later versions of
the model.

Equation 1  of  the  microblock  states  that  export  prices are
determined by prices of imports and wage rates.

Equation 2  is  based on  the  assumption that  export performance
depends on  export competitiveness.  The latter  variable has been
represented by the difference between export and import prices.

Equation 3 states that industrial  investment in the export sector
depends on output.

The specification of the micro-block and the results of estimation
using 1972-1990 data, are given in Appendix 2.


The Macro-block

The macro-block consists of twelve equations explaining changes in
consumption,  imports,  private  sector  employment, unemployment,
wagerates, investment prices, consumer prices, money and taxes.

Only a  brief  description of  the  equations is  given  here. The
specifications of the equations  and the estimation results, using
1972-1990 data, are presented in Appendix 2.

Equation 1 is  a consumption  function, allowing  for the separate
effects of  wage  income  and profits  on  changes  in consumption
expenditure.

Equation 2 deals with changes in imports. It allows for the effect
of prices changes (and indirectly of exchange rate changes) and of
final output changes on imports.
Figures 2
Equation  3  explains  the  factors  affecting  changes  in labour
demand. This is a function of changes in final output (weighted by
labourintensities)  and   the   changes   in   unemployment  rate.
Unemploymentrates are introduced to  capture the effects of excess
demand.

Equation 4  states that  the changes  in unemployment  rates are a
function of labour  demand minus labour  supply, while equation 5,
which is  a version  of an  augmented Phillips  curve, states that
changes in wage  rates are  determined by  changes in  the rate of
unemployment and  changes in  consumer prices.  Equations 6  and 7
explain changes  in investment  and consumer  prices respectively.
Investment prices are  assumed to depend  on importprices and wage
rates, with  a  heavy  54% weighting  on  import  prices. Consumer
prices are assumed to depend on import prices and wages(with a 27%
weighting on import prices), and on expenditure tax.

Equation 8 deals with the growth  of money, defined as currency in
circulation plus bank  deposits. This is  assumed to be influenced
by budget surpluses (or deficits)  by government lagged one period
and by balance of payments surpluses (or deficits).

Equations 9  to  12  deal  with taxes.    Equation  9  states that
importduties depends  on import  value,  while equation  10 states
that  expenditure   taxes  depends   on  consumption  expenditure.
Equation 11 and equation 12 deal  with direct taxes on persons and
corporations,  with  personal  income  and  corporate  profits are
explanatory variables.

The endogenous and exogenous variables utilised in these equations
are described in Appendix 2.


Some Comments of the Results

A few comments  on the  preliminary nature  of the  results are in
order here.

First of all the micro-block for Malta is not yet complete. We did
not yet  found the  time for  modelling the  export sector  in its
disaggregated form.  Instead  we  used three  equations  for total
exports of goods and services, export prices and investment.
These  equations   were   chosen   more   for   their  forecasting
capabilities than for their theoretical underpinnings. For example
the investment equation is investment to output and not to changes
in output, as predicted in  the accelerator theory. The reason for
this is that an accelerator type of equation  produced implausible
results and  introduced a  very complicated  lagged structure into
the model.

The export equation is also a very  simple one, and as can be seen
from  the  R2,  it  leaves  much  to  be  desired. An  alternative
specification, which  would require  a larger  data set  than that
currently available, would allow for  the effects of weighted real
exchange rates and weighted GDP's of Malta's client countries.
Figures 2
Figures 3
The micro-block  will be  developed further  is future experiments
with the model.

The   macro-block   also    requires   further   development.   In
particular,the labour demand equation needs to be improved and the
corporate  sector  needs   to  be  more   adequately  modelled  to
endogenise retained income (corporate saving).

There is  the need  to improve  the specification  of some  of the
equations, since  some  of  the  estimated  coefficients  were not
statistically significant at the 95% level.

Because of these shortcomings, we must stress that the exercise is
to be considered as a preliminary one.
4. SOME SIMULATION EXERCISES


4.1 Historical Simulations

We have simulated the Maltese  Economy using the exogenous and the
pre-determined variables only,  with the values  of the endogenous
variables being  produced by  the model  itself. These simulations
appeared to  be  by and  large  successful,  in that,  with  a few
exceptions, the equations generated  paths of endogenous variables
very similar to  the actual  ones. The cyclical  pattern of growth
described above was clearly captured.

We have also performed a  historical simulation for just two years
namely 1989  and  1990.   The  exercise  involved  using exogenous
variables for 1989 and 1990,  and the pre-determined values of the
endogenous variables for 1988.

The predicted values of some of the endogenous variables are shown
in Table  3  in  section  2. The  yearly  differences  between the
simulated and actual values for 1989 and 1990 are given in Table4.



Table 4. Some Aggregate Variables Predicted by the Model

  Main Economic Indicators   Historical Simulation

 yearly differences simulation compared with realised figures
 ----------------------------------------------------
                                     89       90
         annual percentage changes:
 export price                          -1.       0.
 consumption price                      3.       1.
 import price                           0.       0.
 wage per capita in enterprises        -1.      -3.
 money                                 -6.      -6.
 volume of exports                      2.      -3.
 volume of private consumption         -9.      -2.
 volume of investment enterprises      19.       6.
 volume of imports                     -3.      -4.
 jobs in enterprises                    0.       3.
 real disposable wages +benefits       -5.      -3.
         levels:
 price-costrate exportsector        -.02     -.01
 unemployment rate                    1.       1.
         in percentage of moneyvalue in year t-1:
 surplus on current account             1.       2.
 financial deficit of government       -8.      -7.
 -----------------------------------------------------


It can be seen that the  model underpredicts the actual values of
most of  the  variables. There  was  a  large margin  of  error in
predicted   investment,   but   this   was   expected   given  the
unsatisfactory specification of the investment function.
4.2 Forecasting the Endogenous Variables


The model was also used  to forecast some endogenous variables for
1991-1995. The exercise required information about the future path
of the  exogenous  variables. These  were  predicted on  the basis
oftime-trend, estimated  from the  1985-1990  values, when  such a
trend was detected, and on the  basis of a random number generator
when the exogenous variable fluctuated randomly around a mean.

The results of the forecasting exercise  are shown in Table 5. The
simulated  figures  would  seem  to  suggest  that  there  will be
positive growth rates  on all  the endogenous  variables, but with
growing deficits on current account of the balance of payments.



Table 5. Forecasting Results for 1991-1995

              Main Economic Indicators    Base Line
------------------------------------------------------------------
                           91       92       93       94       95
         annual percentage changes:
 export price               5.       4.       3.       4.       4.
 consumption price          5.       4.       4.       4.       4.
 import price               3.       3.       3.       3.       3.
 wages in enterprises       8.      10.       7.       7.       7.
 money                      2.       4.       4.       3.       3.
 volume of exports         -0.       4.       7.       6.       5.
 vol. private consumption   4.       5.       5.       5.       5.
 vol.investment enterprises 9.      10.      11.      10.      10.
 volume of imports          2.       5.       7.       6.       6.
 jobs in enterprises        8.       2.       2.       3.       3.
 real disposable wages      6.       6.       5.       5.       5.
         levels:
 price-costrate export    1.04     1.03     1.03     1.03     1.03
 unemployment rate          2.       1.       0.      -1.      -2.
   in percentage of moneyvalue in year t-1:
 surplus on current account-5.      -7.      -8.     -10.     -12.
 fin. deficit government   -2.       1.       2.       3.       4.
 ----------------------------------------------------------------


We also attempted  to generate exogenous  shocks into the economy,
to see  what would  happen to  the endogenous  variables. Two such
shocks are reported here.

The first  is  an exogenous  impulse  in the  form of  an  8% wage
increase changes in  1993. The  resulting changes  compared to the
figures shownin Table  5 are  given in  Table 6.  The results show
that, as expected,exports and employment would decrease.
Table  6.  The  Effect  of   an  Exogenous  8%  increase  in  Wage
Rates(Percentage Differences compared with base line)
------------------------------------------------------------
                                   93       94       95

 export price                       2.       4.       5.
 consumption price                  1.       2.       2.
 import price                       0.       0.       0.
 wage costs per unit of output     11.      20.      27.
 money                             -0.      -1.      -3.
 value of exports                  -5.      -8.     -10
 value of private consumption       1.       3.       5.
 value investments enterprises      1.       4.       8.
 value imports                     -3.      -4.      -4.
 number of jobs in enterprises      0.      -3.      -5.
-------------------------------------------------------------


the second is  an exogenous 5%  increase in the  Maltese prices of
exports during the  years 1991  and 1992.   The resulting changes,
compared to the figures shown in Table 5, are shown in Table 7. It
can be seen that this would negatively effect exports, followed by
decrease in consumption, investment and employment. The reason for
this is  of  course  that  the  domestic  value  added  of exports
constitute a very large proportion of GDPin the small open economy
of Malta.


Tale  7.  The  Effect  of  an  Exogenous  5%  increase  in  Export
Prices(Percentage Differences)

   leveldifference in percentage compared with base line
------------------------------------------------------------------
                           91       92       93       94       95

 export price               7.      10.       7.       6.       7.
 consumption price          1.      -0.      -2.      -2.      -2.
 import price               0.       0.       0.       0.       0.
 wage costs p.unit output  11.       2.     -15.     -17.     -15.
 money                     -1.      -1.      -1.       0.       2.
 value of exports         -17.     -23.     -18.     -17.     -18.
 ,, private consumption    -1.      -3.      -5.      -6.      -6.
 ,, investments enterprises 0.      -2.      -6.     -10.     -12.
 value imports            -12.     -17.     -14.     -15.     -16.
 number of jobs enterprises 0.     -10.     -17.     -17.     -16.
-----------------------------------------------------------------
5. CONCLUSION


In  this  study  we  have  attempted  to  do  two  things,  namely
describing a real small economy  and attempting to present a model
which is suitable for such a small economy. We have shown that the
Maltese economy is very small  and very export oriented. The model
captures this  dependence  on  foreign trade,  and  the simulation
exercise have clearly shown that a change in the Maltese prices of
exports would have  severe negative repercussions  on the economy.
This is  of  course  also  true of  any  other  variable affecting
exports, such  as  changes  in economic  conditions  abroad, which
cannot be influences by Malta and are not included in the model.

It has  been stressed  that  the results  should be  considered as
preliminary ones, since there  is much work to  be done to improve
the model  and to  render  it more  useful  for policy  makers and
researchers. As  stated above,  the major  weakness of  this first
version of the Malta Model is the micro-block, since it is yet too
aggregative and  contains  probably  mis-specifies  the investment
equation. We need  also to  allow for  non-price factors affecting
export demand such as demand conditions outside Malta.

We hope to  rectify these  shortcomings in future  versions of the
model.

