Munich Personal RePEc Archive
GDP Forecast for Australia
Saraogi, Ravi
Madras School of Economics, Chennai, Tamil Nadu, India
May 2008
Online at
https://mpra.ub.uni-muenchen.de/22797/
GDP For ecast for Austr alia
How high will the Kangar oos jump?
Ravi Saraogi
5/ 26/ 2008
Australia: An Introduction
The Commonw ealth of Australia is a country in the southern hemisphere comprising the mainland of
the w orld's smallest continent, t he major island of Tasmania, and a num ber of other islands in the
Indian and Pacific Oceans.N4 The neighbouring countries are Indonesia, East Timor, and Papua New
Guinea to the north, the Solom on Islands, Vanuatu, and New Caledonia t o the north-east, and New
Zealand to the south-east. Australia is the only country that is also a continent.
Australia has a prosperous, West ern-st yle mixed economy, w ith a per capita GDP slightly higher t han
that of the UK, Germany, and France in t erms of purchasing pow er parit y. The countr y w as ranked
third in the United Nations' 2007 Human Development Index and sixth in The Economist w orldw ide
quality-of-life index 2005. The absence of an export -orient ed manufact uring indust ry has been
considered a key w eakness of the Australian economy. M ore recently, rising prices for Australia's
com modit y exports and increasing tourism have m ade this criticism less relevant. Nevertheless,
Australia has the w orld's fourth largest current account deficit in absolute t erms (in relative term s it
is more than 7% of GDP). This is considered problematic by some economist s, especially as it has
coincided w ith t he high t erms of t rade and low interest rates that make the cost of servicing the
foreign debt low .
The Haw ke Governm ent start ed the process of economic reform by floating the Australian dollar in
1983, and partially deregulating the financial syst em. The How ard governm ent continued the
process of microeconomic reform, including a partial deregulation of the labour market and the
privatisation of stat e-ow ned businesses, most notably in the t elecom munications industry. The
indirect tax system w as substantially reform ed in July 2000 w it h t he introduction of a 10% Goods and
Services Tax (GST), w hich has slightly reduced the heavy reliance on personal and company incom e
tax that characterises Australia's t ax system.
At January 2007, there w ere 10,033,480 people em ployed, w ith an unemploym ent rat e of 4.6%.
Over the past decade, inflation has typically been 2–3% and the base interest rat e 5–6%. The service
sect or of the economy, including tourism, educat ion and financial services, const itut es 69% of GDP.
Agriculture and natural resources constitut e 3% and 5% of GDP but contribut e substantially to export
performance. Australia's largest export market s include Japan, China, the US, South Korea and New
Zealand.
Australia: Fact Sheet
Australian Economy: A Graphical Tour
Source: Reser ve Bank of Austr alia
The M odel
We are estimating a Vector Auto Regressive (VAR) m odel to forecast the quart erly grow th rat es in
the GDP (at constant 2000 USD) of Australia1, w hich has som e obvious benefits over a pure
simultaneous equation syst em. In the case of simult aneous equation syst ems, some variables are
treat ed as endogenous and som e as exogenous or predet ermined (exogenous plus lagged
endogenous). Before w e estimat e such models, w e have to make sure that the equations in the
syst em are ident ified (either exact ly or over-). This identification is often achieved by assuming that
some of the predet ermined variables are present only in some equat ions. This decision is often
subjective and has been severely criticized by Christopher Sims2. According to Sims, if t here is true
simultaneity among a set of variables, they should all be treated on an equal footing; there should
not be any a priori distinction betw een endogenous and exogenous variables. It is in this spiri t that
Sims developed his VAR model w hich w e use for our purpose.
All the variables in a VAR model are endogenous and w e can apply the usual OLS procedure3. The
model w hich w e w ill be using is,
Yt
= α + β H
CIt -i+ γ P
CIt -i + η BIt -i+ θ ESIt -i+ εt
w here, Yt = Quarterly grow t h rate in GDP (at const ant 2000 US$)
α = Intercept term
HCIt -i = Human capital index wit h i t h
time period lag
PCI t -i = Physical capital index with i t h
time period lag
BIt -i = Banking index w it h i t h
time period lag
ESIt -i = Ext ernal sector index w it h i t h
time period lag
β, γ, η, θ =
Regression coefficientsεt
= Error termIn the above model, w e estimat e human capital index, physical capital index, banking index and
ext ernal sector index w ith the help of 16 variables4 as described below -
Human Capit al Index
HCIt = a (PG) + b (DR) + c (Edu)
1
Sample dat a is from 1960:1 t o 2006:4 and t he forecast is up t o 2050:4
2
C. A. Sims, “Macroeconomics and Reality,” Econometrica, vol. 48, 1980, pp. 1–48.
3
Gujarat i, “ Time Series Econom et rics: Forecast ing” , Basic Economet rics, pp 853
4
Where,
a,b,c are w eights ( a + b + c =1 and a=b=c)
PG= Quart erly population grow t h rates in per cent
DR= Quarterly dependency ratio
Edu= Quart erly govt. expenditure on education as a per cent of quarterly GDP (at current US$)
Physical Capital Index
PCIt = d (GFCFt -1) + e (FDIt -1) + f (M Cap)
Where,
d,e,f are w eights ( d + e + f = 1 and d=e=f)
GFCFt -1 = Quarterly gross fixed capit al formation as a per cent of quarterly GDP (at current US$) with
one t im e period lag
FDIt -1 = Quart erly net foreign direct invest m ent as a per cent of quart erly GDP (at current US$) w ith
one t im e period lag
M Cap = Quarterly stock market capitalization as a per cent of quarterly GDP (at current US$)
Banking Index
BIt = g (I) + h (R) + i (L) + j (CR)
g, h, i, j are w eights ( g + h + i + j = 1 and g=h=i =j)
I = Quart erly prime lending rate
RR = Quarterly bank liquid reserves t o bank assets ratio (ratio)
L= Quarterly banks non performing loans to total loans (rat io)
CR = Quarterly bank capital to asset ratios
External Sect or Index
ESIt =k (REER) + l (CA) + m (T) + n (F)
Where,
k, l, m, n are w eights (k + l + m + n = 1 and k=l=m =n)
REER= Quart erly Real Effect ive Exchange Rate (REER)
CA= Quarterly capit al account balance as a per cent of quarterly GDP (at current US$)
T= Quart erly trade balance in goods and services as a per cent of quart erly GDP (at current US$)
The Variables 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
60 65 70 75 80 85 90 95 00 05 PG .48 .50 .52 .54 .56 .58 .60 .62 .64
60 65 70 75 80 85 90 95 00 05 DR 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
60 65 70 75 80 85 90 95 00 05 EDU 18 20 22 24 26 28 30 32
60 65 70 75 80 85 90 95 00 05 GFCF
S.N o Variable N ame
Variable
Symbol U nits Frequency Actual Series From
1 GDP Growt h Rat e (in const ant 2000 US$) Y % Quarter ly 1960:1 t o 2006:4 2 Populat ion Growt h Rat e PG % Quarter ly 1960:1 t o 2006:4 3 Dependency Rat io DR Ratio Quarter ly 1960:1 t o 2006:4 4 Educat ion on Expendit ure (% of curr ent GDP) Edu % Quarter ly 1973:3 t o 1994:4
5
Gross Fixed Capit al Form ation (% of current
GDP) GFCF % Quarter ly 1971:1 t o 2005:4
6
Net Foreign Direct Investm ent (% of current
GDP) FDI % Quarter ly 1960:1 t o 2006:4 7 M arket Capit alizat ion (% of current GDP) M cap % Quarter ly 1988:1 t o 2006:4 8 Prim e Lending Rat e I % Quarter ly 1975:1 t o 2006:4 9 Bank Liquid Reserves t o Bank Asset s Ratio RR Ratio Quarter ly 1960:1 t o 2006:4 10 Banks Non Perf orming Loans t o Tot al Loans L Ratio Quarter ly 2000:1 t o 2006:4 11 Bank Capit al t o Asset Ratio CR Ratio Quarter ly 2000:2 t o 2006:4 12 Real Eff ect ive Exchange Rat e REER Index Quarter ly 1980:2 t o 2006:4 13 Capit al Account Balance (% of current GDP) CA % Quarter ly 1961:1 t o 2006:4
14
Trade Balance in Goods and Services (% of
-2 -1 0 1 2 3 4 5
60 65 70 75 80 85 90 95 00 05 FDI 20 40 60 80 100 120 140 160 180
60 65 70 75 80 85 90 95 00 05 MCAP 6 8 10 12 14 16 18 20 22
60 65 70 75 80 85 90 95 00 05 I 0 4 8 12 16 20 24
60 65 70 75 80 85 90 95 00 05 RR .1 .2 .3 .4 .5 .6 .7
60 65 70 75 80 85 90 95 00 05
L 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
60 65 70 75 80 85 90 95 00 05
CR 90 100 110 120 130 140 150
60 65 70 75 80 85 90 95 00 05
REER -0.2 0.0 0.2 0.4 0.6 0.8 1.0
60 65 70 75 80 85 90 95 00 05
The follow ing graphs are for the initial indices that w ere construct ed w ith the use of appropriat e
w eights (equal w eights have been given to each variable w ithin an index). As a visual examination
w ould suggest, apart from the series for GDP and perhaps HCI, none of the series appear t o be
stationary. -4 -3 -2 -1 0 1 2 3 4
60 65 70 75 80 85 90 95 00 05 T 0 2 4 6 8 10 12
60 65 70 75 80 85 90 95 00 05 F 2 4 6 8 10 12 14 16 18
65 70 75 80 85 90 95 00 05 BI -10 0 10 20 30 40
65 70 75 80 85 90 95 00 05 ESI -2 -1 0 1 2 3
65 70 75 80 85 90 95 00 05 GDP 0 10 20 30 40 50 60
65 70 75 80 85 90 95 00 05 PCI 0.4 0.8 1.2 1.6 2.0 2.4
65 70 75 80 85 90 95 00 05 HCI
BI = Banking Index
ESI = External Sect or Index
GDP = GDP Grow th Rat e
HCI = Human Capital Index
For furt her information about the stationarity of the series, the augment ed dickey fuller t est w as applied to each of the five series to and appropriate t ransformation w as used to convert non stationary data into stat ionary data.
S.No. Data Series Stationary Transformation Used for St ationarity
1 GDP Yes Already stationary
2 HCI Yes Already stationary
3 PCI No First difference transformation
4 ESI No First difference transformation
5 BI No First difference transformation
BI= Banking Index (Non St ationary) DBI = First difference of BI (St ationary) ESI=Ext ernal Sector Index (Non Stationary) DESI=First difference of ESI (St ationary) PCI=Physical Capital Index (Non Stationary) DPCI=First difference of PCI (Stationary)
2 4 6 8 10 12 14 16 18
65 70 75 80 85 90 95 00 05 BI -3 -2 -1 0 1 2 3 4
65 70 75 80 85 90 95 00 05 DBI -10 0 10 20 30 40
65 70 75 80 85 90 95 00 05 ESI -10 0 10 20 30 40
65 70 75 80 85 90 95 00 05 DESI 0 10 20 30 40 50 60
65 70 75 80 85 90 95 00 05 PCI -8 -4 0 4 8 12
After making the data stationary, w e estimat ed an unrestrict ed VAR type w ith 3 lag intervals. The lag interval w as select ed based on t he follow ing criterion-
VAR Lag Order Selection Criteria
Endogenous variables: DBI DESI DPCI GDP HCI Exogenous variables: C
Sample: 1960:3 2006:4 Included observations: 183
Lag LogL LR FPE AIC SC HQ
0 -1107.402 NA 0.131074 12.15739 12.24509 12.19294
1 -673.0682 840.1859 0.001495 7.683805 8.209951 7.897078
2 -484.3979 354.6589 0.000250 5.895059 6.859659 6.286058
3 -368.0067 212.4297* 9.23E-05* 4.896249* 6.299303* 5.464975* * indicates lag order selected by the criterion
LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error
AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion
The VAR model has been outlined below -
Eq1: dbi = F( dbi, desi, dpci, gdp, hci ) Eq2: desi = F( dbi, desi, dpci, gdp, hci ) Eq3: dpci = F( dbi, desi, dpci, gdp, hci ) Eq4: gdp = F( dbi, desi, dpci, gdp, hci ) Eq5: hci = F( dbi, desi, dpci, gdp, hci )
The quart erly estimat es w ere obtained up to 2050:4 with the follow ing results
As w e can see, the grow th rat e in quarterly GDP (at const ant 2000 US$) stabilises in the range of 0.5
to 1 per cent. The above model w as also estimat ed w ith the quart erly grow th rate of GDP at current
prices w ith the follow ing results- .
-2 -1 0 1 2 3
70 80 90 00 10 20 30 40 50 GDP
0.0 0.5 1.0 1.5 2.0 2.5 3.0
70 80 90 00 10 20 30 40 50 GDP (Baseline Mean)
The grow th rat e for quarterly GDP at current prices stabilizes at around 1.9 per cent w ith minor
fluctuations on either side. The above result how ever should be accept ed w ith the several caveats.
The data has not been checked for problem s of autocorrelation, multicolinearity and
het eroscedast icity. The accuracy of t he estimat e may have been compromised due to part
availability of some data series and the indexation t echnique adopt ed in such cases. The w eights
used for indexation w ere assumed t o be equal. This can how ever be challenged on t heoretical
grounds.
Appendix A – Variable Description5
Populat ion growth (quarterly %): Quarterly Annual population grow t h rate. Population is based on
the de facto definition of population, w hich count s all resident s regardless of legal stat us or
citizenship--except for refugees not permanently set tled in the count ry of asylum, w hich are
generally considered part of the population of the count ry of origin.
Dependency Ratio (dependents t o w orking-age population): Dependency ratio is t he ratio of
dependents--people younger than 15 or older than 64--to the w orking-age population--t hose ages
15-64. For example, 0.7 means there are 7 dependents for every 10 w orking-age people.
Gross fixed capital formation (% of GDP at current US$): Gross fixed capital formation (formerly
gross domestic fixed investm ent) includes land improvem ents (fences, ditches, drains, and so on);
plant , machinery, and equipm ent purchases; and the construction of roads, railw ays, and t he like,
including schools, offices, hospitals, private resident ial dw ellings, and comm ercial and industrial
5
Dat a for t he below ment ioned variables was collect ed for the t ime per iod 1960:1 t o 2006:4 from
w ww.economy.com (M oody’s Economy)
-8 -4 0 4 8 12
70 80 90 00 10 20 30 40 50
Actual GDP at Current Prices Growth Rate (Quarterly) up to 2006:4
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
70 80 90 00 10 20 30 40 50
buildings. According to the 1993 SNA, net acquisitions of valuables are also considered capital
formation.
Net Foreign direct investment, net inflow s (% of GDP at current US$): Foreign direct investm ent are
the net inflow s of investm ent to acquire a lasting management int erest (10 percent or more of
voting stock) in an enterprise operating in an econom y other than that of the investor. It is t he sum
of equit y capital, reinvestm ent of earnings, other long-term capital, and short -term capit al as show n
in the balance of paym ents. This series show s net inflow s in the reporting econom y and is divided by
GDP.
M arket capitalization of listed companies (% of GDP at current US$): M arket capitalization (also
know n as market value) is the share price times the number of shares outstanding. List ed domestic
companies are the domestically incorporat ed companies listed on the count ry's stock exchanges at
the end of the year. Listed companies do not include investm ent companies, mutual funds, or other
collective invest ment vehicles.
Prime lending rate (in %): Prim e lending int erest rat e is the rate charged by banks on loans to prime
customers.
Bank liquid reserves t o bank assets ratio: Rat io of bank liquid reserves to bank assets is the ratio of
dom estic currency holdings and deposits w it h the monetary authorities t o claims on other
governments, nonfinancial public ent erprises, the private sector, and other banking insti tutions.
Bank nonperforming loans to t otal loans (ratio): Bank nonperforming loans to total gross loans are
the value of nonperforming loans divided by the total value of the loan portfolio (including
nonperforming loans before the deduct ion of specific loan-loss provisions). The loan amount
recorded as nonperforming should be t he gross value of the loan as recorded on the balance sheet,
not just the amount t hat is overdue.
Bank capital to asset s (ratio): Bank capital to assets is the ratio of bank capital and reserves to total
assets. Capital and reserves include funds contribut ed by ow ners, ret ained earnings, general and
special reserves, provisions, and valuation adjustm ent s. Capital includes tier 1 capital (paid -up shares
and com mon st ock), w hich is a common feature in all count ries' banking syst ems, and t otal
regulatory capital, w hich includes several specified types of subordinat ed debt instrum ents that
need not be repaid if the funds are required to maint ain minimum capit al levels (these comprise tier
Real effective exchange rate index (2000 = 100): Real effective exchange rat e is t he nominal
effective exchange rat e (a measure of the value of a currency against a w eight ed average of several
foreign currencies) divided by a price deflator or index of costs.
Net capital account (% of GDP at current US$): Net capital account includes government debt
forgiveness, investm ent grants in cash or in kind by a governm ent entit y, and taxes on capital
transfers. Also included are migrants' capital transfers and debt forgiveness and invest m ent grants
by nongovernm ental entities. Data are in current U.S. dollars.
Net trade in goods and services (% of GDP at current US$): Net trade in goods and services is
derived by offset ting imports of goods and services against exports of goods and services. Exports
and imports of goods and services comprise all t ransactions involving a change of ow nership of
goods and services bet w een residents of one country and the rest of the w orld. Data are in current
U.S. dollars.
Forex reserves (excluding gold, % of GDP at current US$): Total reserves com prise holdings of
monetary gold, special draw ing rights, reserves of IM F members held by the IM F, and holdings of
foreign exchange under the control of monetary authorities.
Expenditure on Education (% of GDP at current US$): To capture the skill base, governm ent