Docstoc

Foot and mouth

Document Sample
Foot and mouth Powered By Docstoc
					FOOT AND MOUTH DISEASE




            Foot and mouth
      Evaluating alternative options for controlling disease outbreaks

           Ali Abdalla, Liangyue Cao, Anna Heaney, Milly Lubulwa and Stephen Beare


                                               Introduction
                                               The economic and social costs of a foot and
                                               mouth disease (FMD) outbreak in Australia
                                               could be considerable. Recent experience in
                                               the United Kingdom and Europe highlights
                                               the devastating impact of an outbreak, with
                                               consequences including the loss of interna-
                                               tional market access and disruption to the
                                               domestic market for the affected livestock
                                               industries, production losses, the slaugh-
                                               tering of thousands of animals, compensa-
                                               tion payments to producers and restrictions
                                               on livestock movements across many
                                               European countries.
 It is important to evaluate the cost             The full economic consequences of an
  effectiveness of a range of control          outbreak of FMD in Australia are uncertain
  strategies, including vaccination,           but the beef and dairy industries are likely
                                               to be the worst affected. The gross value of
   for dealing with an outbreak of             Australian farm production of livestock
        foot and mouth disease.                products was estimated to be around $17.6
                                               billion in 2001-02 (ABARE 2002). In the same
                   x                           year, exports of beef and veal, live cattle and
                                               dairy products earned around $7.9 billion
  The cost of lost market access is            and accounted for half the value of exports
   far and away greater than the               of Australia’s livestock products. With such
    direct costs of controlling an             a high value of exports at risk, closure of
                                               international markets to Australian product
       outbreak of the disease.                could, in itself, lead to substantial economic
                                               costs.
                   x                              The threat posed by an FMD outbreak has
                                               given rise to further consideration of appro-
 While a vaccination strategy may              priate policy responses to disease incursions
  increase the time before market              in Australia. In November 2001, Common-
 access can be regained, it greatly            wealth, state and territory officials and
                                               industry leaders met to discuss policy issues
  reduces the risk and associated
                                               relating to preparedness and response capa-
  costs of a large scale outbreak.             bilities for dealing with both FMD and
                                               bovine spongiform encephalopathy (BSE or
                   x                           ‘mad cow’ disease). Among the recommen-
                                               dations arising from this forum was the need
                                               to consider a range of FMD control strate-
                                               gies, including vaccination, and to under-
                                               take an economic analysis to evaluate
                                             Australian Commodities vol. 9 no. 1, March quarter 2002
226
                                                                           FOOT AND MOUTH DISEASE

conditions under which vaccination would                  of domestic production excluded from trade
become a preferred control option.                        and the period of time before disease free
   Vaccination is most likely to be an effec-             status was restored.
tive option when there is a high probability                  To minimise the volume of exports
that the disease will spread rapidly through              affected, the principle of zoning (defined by
livestock herds and available resources will              various Office International Des Epizooties
be insufficient to stamp out the infection                codes) allows a country to demarcate
immediately. However, vaccination may not                 regions into infected and disease free areas.
be appropriate in all circumstances as it can             Zoning is likely to be an integral component
impose costs through, for example, extend-                of Australia’s FMD response, as access to
ing the delay associated with regaining                   export markets can be retained for product
access to export markets.                                 from designated disease free areas. Zoning
   In this paper, estimates are presented of              could, therefore, reduce the costs associated
direct and market access costs for vaccina-               with market closure considerably (figure A).
tion and two other FMD control strategies                     State and territory boundaries would pro-
for a simulated outbreak in an intensive live-            vide appropriate zoning limits initially
stock region in northern Victoria, an area                because they may be more easily accepted
where vaccination is most likely to be a                  internationally as distinct geographic boun-
favorable control option.                                 daries (Commonwealth of Australia 2001).
                                                          However, it may prove difficult and expen-
                                                          sive, in terms of monitoring costs, to gain
Costs of an FMD outbreak                                  acceptance of FMD free status by Australia’s
FMD is a highly contagious viral infection                trading partners.
of domestic and wild cattle, sheep, goats and                 Importantly, the market access costs of an
pigs that is spread by direct contact between             FMD outbreak escalate over time — for
infected and susceptible animals, and indi-               example, the costs associated with in-
rectly by contaminated fomites, windborne                 creasing periods of lost market access, as
spread and dissemination of the virus by                  shown in figure A, increase at an increasing
animal products. The major method of                      rate.
spread is through animal movements                            If an outbreak is controlled quickly, live-
(Hassall and Associates 1993). The costs                  stock producers may be able to retain most
associated with an outbreak of FMD fall into              of their stock inventories until market access
two broad categories — operational (or                    is regained. However, if market access is lost
direct) costs, which are those designed to                for an extended period, breeding stock may
avoid, eliminate or reduce the impacts of the             need to be reduced. International competi-
disease and production losses, and forgone                tors may respond by expanding their herds,
revenue, from the loss of international trade
in livestock and livestock products.
   Operational costs include those associated
with decontamination of infected properties,                A     Market access costs of an
                                                                       FMD outbreak
slaughter and disposal of infected animals,
and the cost of both professional and non-
                                                            3.0
professional labor involved in administering
control strategies. However, it is clear that                                       No zoning
                                                            2.5
the bulk of the economic cost of an FMD
                                                            2.0
outbreak would stem from the potential loss
of market access. In particular, suspension                 1.5
of exports to the high returning Pacific Rim
                                                            1.0                                 Zoning
markets would have substantial impacts on
the livestock industries and the Australian                 0.5
economy, as all exports of cloven hoofed ani-
mals and their products would cease for an                  $b
undetermined period of time. The extent of                         6       9     12       15       18
the impact would depend on the proportion                                       Months

Australian Commodities vol. 9 no. 1, March quarter 2002                                              227
FOOT AND MOUTH DISEASE

imposing greater costs on domestic pro-               stamped out. Strategic ring vaccination
ducers if future market share is lost.                around the infected area could be con-
   In turn, the extent and duration of the            sidered where there is a risk that an out-
outbreak are influenced by the strategy               break could rapidly escalate. In the
implemented to eradicate the disease. The             second week of control, it is assumed that
duration of closure to trading markets is also        four separate focal points have been iden-
highly dependent on the control strategy              tified and the decision to use vaccine is
and is governed by international rules spec-          taken.
ifying the criteria for regaining freedom of          An epidemiological model was used to
disease status. Choosing an FMD manage-            simulate the likely size and duration of out-
ment strategy, therefore, involves several         breaks based on these three control strate-
tradeoffs as the strategy employed will have       gies. The model is described in more detail
implications for the extent of the operational     in Garner and Lack (1995). The outcomes
costs, duration of the outbreak and, perhaps       from the epidemiological model were used
most importantly, duration of market access        to estimate costs of control, compensation
restrictions.                                      payments and periods of market closure for
                                                   the three strategies. Model parameters were
Evaluating FMD manage-                             taken from previous studies by Hassall and
                                                   Associates (1993) and Garner, Allen and
ment strategies                                    Short (1997).
The costs and benefits of vaccination were            The major outputs for each control
compared with those of two other FMD               strategy are duration of the outbreak in
management strategies in order to evaluate         weeks, number of infected herds, number of
its cost effectiveness in the event of an out-     dangerous contact herds and the number of
break in northern Victoria. In the scenario        herds destroyed and/or vaccinated to
presented here, the disease is recognised as       achieve eradication, by animal type. The
a multifocal outbreak two weeks after its          model determines the end of the outbreak
introduction.                                      when there is no incubating or infectious
                                                   herds left in the population. This is the min-
Three control strategies evaluated                 imum period, because in reality authorities
• Stamping out infected herds (SO) — this          would not know that the outbreak had
  strategy involves slaughtering herds that        ended until some weeks had passed with no
  have been identified as being infected, to       outbreaks.
  eliminate the source of further spread of           Each of the control strategies has both
  the disease.                                     human and operational resource implica-
• Stamping out infected herds and dan-             tions. Resource constraints are therefore
  gerous contact herds (SODC) — this               imposed in the model. For example, if vac-
  strategy involves the slaughtering of            cination is used, the number of herds able to
  infected herds and of herds that come in         be slaughtered in a week is halved because
  contact with infected animals. Dangerous         resources are diverted away from slaugh-
  contact herds are at high risk of becoming       tering. Also, there is no dangerous contact
  infected and although they may not show          slaughter when vaccination is used because
  signs of infection, they may be suspected        those resources are used for dealing with the
  of incubating the disease. This strategy is      infected herds. That is, the number of herds
  employed to reduce subsequent disease            that can be treated in total is increased, as
  instances.                                       vaccination requires less time than slaugh-
• Stamping out infected herds and ring vac-        tering.
  cination (SORV) — this strategy involves            Market closure costs vary depending on
  slaughtering infected herds and vacci-           the control strategy implemented. Under
  nating herds in surrounding properties           stamping out (SO) and dangerous contact
  within a specified distance. Vaccination is      strategies (SODC), a period of three months
  used to create a buffer ring in an attempt       must elapse after the last case of infection is
  to contain the disease within an area after      eradicated before market access can be
  infected herds have been identified and          regained. If vaccination is used, the vacci-
228                                              Australian Commodities vol. 9 no. 1, March quarter 2002
                                                                           FOOT AND MOUTH DISEASE

nated animals must be slaughtered and a                   dependent on the time taken to detect the
further period of three months must pass                  outbreak.
before market access can be regained.                        The earlier an outbreak is detected, the
   For the purposes of this analysis, it is               greater the probability the disease can be
assumed that it would take one month to                   controlled quickly. For example, under
slaughter the vaccinated animals. Slaughter               stamping out, the probability that the dis-
of vaccinated animals would be undertaken                 ease will escape containment is estimated to
through abattoirs. While it is possible that              be 0.19 if the disease is detected within two
producers could get a market return from                  weeks of introduction. In comparison, the
the slaughtered animals by exporting to                   stamping out dangerous contact strategy
countries that accept product from vacci-                 carries a 0.09 probability that the disease will
nated animals or to FMD declared areas, it                escape if detected within two weeks. Under
is more likely that there is no market for vac-           these strategies, the FMD virus escapes con-
cinated animals.                                          tainment when human and operational
   It is assumed in this analysis that there is           resources are overextended. The probability
no market for vaccinated animals and that                 that the disease will escape under a vacci-
the costs of slaughter and disposal would                 nation policy is extremely small.
have to be met. All properties where infected                The range of possible outcomes from the
and dangerous contact herds are slaughtered               model simulations of the time needed to
and 10 per cent of properties where animals               eradicate the disease plus the additional
are vaccinated are decontaminated. All costs              time needed to attain disease free status rep-
in this analysis are presented in net present             resent possible durations of market closure
value terms in 2000-01 dollars.                           under each strategy. These are shown as
   Livestock production in northern Victoria              probability distributions in figure B.
is primarily a high intensity, pasture based                 Under the SO and SODC strategies, there
industry, consisting of dairying, irrigated               are two general outcomes from an outbreak.
and dryland sheep and beef grazing, pigs                  The outbreak could be relatively small
and poultry. There are large areas of irri-               (affecting less than 1000 herds) and con-
gated pasture with very high stocking rates.              trolled quickly or it could be large and last
Likely agents for the spread of the FMD                   for an extended time.
virus include milk tankers and respiratory                   The slaughtering of dangerous contact
aerosols that would assist airborne spread                herds increases the likelihood that the dis-
under conditions of high livestock density,               ease is controlled quickly and reduces the
at least over short distances.                            probability of a large outbreak when com-
   Environmental conditions, relatively high              pared with the stamping out policy.
stocking densities and high rates of stock                   Large outbreaks occur when the rates of
disposal through saleyards and associated                 disease spread significantly outstrip avail-
stock movements in this region indicate that              able resources, enabling the disease to accel-
the potential spread of FMD could be rela-                erate out of control. In this situation, the
tively high.                                              model results show that a high proportion
                                                          of livestock would be slaughtered in an
Implications of alternative                               attempt to eradicate the disease. In the sce-
                                                          nario presented here, more than 90 per cent
strategies                                                of livestock would be slaughtered under SO
The extent and duration of an FMD out-                    and SODC strategies in northern Victoria.
break depend on the control strategy imple-                  In comparison, the average proportion of
mented, which in turn determines the time                 livestock slaughtered under the SO and
before market access can be regained.                     SODC strategies when small outbreaks
However, the outcome of a control strategy                occur was simulated to be 3.2 per cent and
is not fixed. There is a probability that the             1.7 per cent respectively. The average pro-
disease will be controlled quickly and there              portion slaughtered under the SORV
is a probability that the disease will escape             strategy was estimated to be 1.8 per cent.
containment. These probabilities vary with                   The use of ring vaccination greatly
different control strategies and are highly               reduces the likelihood of a large outbreak.
Australian Commodities vol. 9 no. 1, March quarter 2002                                              229
FOOT AND MOUTH DISEASE

             Frequency distribution of            As a consequence the costs of livestock
 B        likely duration of market closure       slaughter are correspondingly smaller.

 a Stamping out strategy                          Costs of an FMD
  frequency
  12
                                                  outbreak
                                                  The variability in operational requirements
  10                                              and duration of market access restrictions
                                                  under each control strategy has a significant
      8                                           bearing on the overall cost of an FMD out-
                                                  break. In order to compare the control strate-
      6                                           gies, operational costs and market access
                                                  costs were estimated for different durations
      4
                                                  of market closure for the three strategies.
      2
                                                     A bioeconomic model for livestock pro-
                                                  duction and trade was used to estimate the
      0                                           trade costs resulting from market closure.
                                                  This model has been described in more
          2   4   6   8 10 12 14 16 18 20
                       Months                     detail in Cao, Klijn and Gleeson (2002).
                                                  Using the previously stated assumptions
 b Stamping out dangerous contact strategy        about the time it takes to regain disease free
  frequency                                       status under different strategies, control
                                                  costs that correspond to market closure
  30
                                                  durations of six, twelve and eighteen
                                                  months have been estimated for each
                                                  strategy. Costs for outbreaks between these
                                                  durations were interpolated.
  20
                                                     Two scenarios were considered for each
                                                  duration of market closure. In a no zoning
                                                  scenario, all Australian exports to FMD free
  10                                              countries are suspended for the entire period
                                                  of market closure. In a zoning scenario, the
                                                  state of Victoria was identified as an FMD
      0                                           affected zone. The rest of Australia was des-
          2   4   6   8 10 12 14 16 18 20         ignated as an FMD free zone. It was assum-
                       Months                     ed that access to international markets
                                                  would remain closed for a period of six
 c Stamping out and ring vaccination strategy     months while the zone was established, after
  frequency                                       which trade from the FMD free zone would
  50                                              continue as normal.
                                                     The distribution of costs for the control
                                                  strategies, with and without zoning, is
  40
                                                  shown in figure C. The distribution of costs
                                                  reflects the underlying distribution of the
  30
                                                  likely time to control the outbreak and the
                                                  corresponding control and loss of market
  20                                              access costs associated with the duration of
                                                  the outbreak. The expected cost of an out-
  10                                              break is shown in figure D.
                                                     With a stamping out (SO) policy alone,
  0                                               the probability of controlling the outbreak
          2   4   6   8 10 12 14 16 18 20         quickly is relatively low. At the same time,
                       Months                     the probability of the disease escaping con-
                                                  tainment is relatively high and, as a conse-
230                                             Australian Commodities vol. 9 no. 1, March quarter 2002
                                                                            FOOT AND MOUTH DISEASE


  C         Costs of controlling a disease outbreak – with and without trading

                                             Stamping out
   frequency           No zoning                          frequency           Zoning
   12                                                     12

   10                                                     10

    8                                                      8

    6                                                      6

    4                                                      4

    2                                                      2

    0                                                      0
        0    411     883 1355 2727 4099                        0   625   670 714 1566 2419
                       Cost ($million)                                      Cost ($million)

                                 Stamping out dangerous contact
   frequency           No zoning                          frequency           Zoning


  30                                                      30



  20                                                      20



   10                                                     10



    0                                                      0
        0    414     889    1365 2730      4096                0   627   676   725 1570        2415
                       Cost ($million)                                      Cost ($million)

                                Stamping out and ring vaccination
   frequency           No zoning                          frequency           Zoning
  50                                                      50

  40                                                      40

  30                                                      30

  20                                                      20

   10                                                     10

    0                                                     0

        0      292      651    1009      1367                  0      612     651    689       727
                       Cost ($million)                                       Cost ($million)


Australian Commodities vol. 9 no. 1, March quarter 2002                                               231
FOOT AND MOUTH DISEASE

quence, the expected costs of this control             Overall, the use of vaccination leads to the
option are high. The distribution of costs is       same level of expected costs as the SODC
widely spread and there is a high degree of         policy without zoning. Expected costs are
variability in the cost of an outbreak under        lower with ring vaccination under zoning
the stamping out option. Zoning substan-            as the probability of regaining access before
tially reduces the level and variation in costs     or soon after the assumed time to establish
under the stamping out policy, reflecting a         zoning is reduced.
greater likelihood that the outbreak will per-         In comparison with the SORV control
sist after the zone is assumed to be estab-         strategy, SODC increases the probability of
lished.                                             quickly regaining export markets. This is
   A stamping out policy combined with the          because regulations governing the use of
slaughtering of dangerous contacts (SODC)           vaccines require a three month period to
increases the likelihood of controlling the         elapse after the slaughter of the last vacci-
outbreak early and reduces the probability          nated animal before market access can be
that the disease will escape containment.           regained. However, there is an increased risk
The distribution of expected costs reflects         under the SODC strategy of a high cost out-
these changes in the likely duration of the         break if the disease escapes containment.
outbreak and the greater probability of a rel-
atively short loss of market access. However,
there is still a substantial risk that the dis-     Concluding remarks
ease will escape containment. It is clear that      The findings of the analysis presented here
the impact of zoning is smaller under a             suggest that adopting a stamping out policy
SODC strategy than for stamping out. This           alone in intensive livestock production areas
reflects the greater likelihood that the out-       is unlikely to be cost effective. This is
break will be controlled before or soon after       because it is unlikely that the disease will be
zoning can be established.                          controlled quickly and there is a significant
   A stamping out policy combined with              probability that the disease will escape con-
ring vaccination (SORV) significantly               tainment.
reduces the likelihood that the disease will           Adopting a stamping out policy with the
escape containment under the assumptions            elimination of dangerous contacts increases
used in the epidemiological model. At the           the likelihood that the disease will be con-
same time, it eliminates the possibility of a       trolled quickly and it is less likely to escape
comparatively short loss of market access.          containment. Nevertheless, the probability
The distribution of expected costs reflects         that the disease will spread and impose large
the loss of both the high risk and high return      costs under this control strategy cannot be
options.                                            ignored.



  D     Expected cost of FMD outbreak, by strategy

                   No zoning                                                 Zoning
2.0                                                  2.0
                                Direct costs
 1.5                            Access costs          1.5

 1.0                                                  1.0

0.5                                                  0.5

 $b                                                   $b
         SO          SODC            SORV                      SO            SODC            SORV
                     Strategy                                                Strategy

232                                               Australian Commodities vol. 9 no. 1, March quarter 2002
                                                                             FOOT AND MOUTH DISEASE

   The use of ring vaccination reduces the                strategy. However, having established that
likelihood that market access will be                     vaccination can be an effective FMD control
regained quickly but greatly reduces the                  strategy, there is a need to evaluate a broader
probability that the disease will escape con-             range of conditions to determine when the
tainment.                                                 use of vaccination is likely to be appropriate.
   When compared with controlling the out-
break through stamping out, the elimination
of dangerous contacts and ring vaccination                References
reduce the expected cost of an outbreak by                ABARE 2002, ‘Statistical tables’, Australian
about the same order of magnitude.                         Commodities, vol. 9, no. 1, March quarter.
However, ring vaccination reduces the risk                Cao, L., Klijn, N. and Gleeson, T. 2002, Modeling
of a large scale outbreak.                                 the effects of a temporary loss of export mar-
                                                           kets in the case of a foot and mouth outbreak
   The resources committed to establishing
                                                           in Australia: preliminary results on costs to
FMD free zones is clearly dependent on the                 Australian beef producers and consumers, 46th
expected time it will take to control the out-             Annual Conference of the Australian Agri-
break and hence the control strategies                     cultural and Resource Economics Society,
adopted. It should be noted that the costs of              Canberra, 12–15 February.
establishing and monitoring zones has not                 Commonwealth of Australia 2001, Ausvetplan
been taken into account.                                   2001 – Disease Strategy: Foot-and-Mouth Disease,
   The benefits of zoning increase substan-                Agriculture and Resource Management
tially the longer it takes to control the out-             Council of Australia and New Zealand,
break. However, this may also reduce the                   Canberra, May.
                                                          Garner, M.G. and Lack, M.B. 1995, ‘An evalua-
likelihood that zones can be established and
                                                           tion of alternate control strategies for foot-and-
cost effectively maintained.                               mouth disease in Australia: a regional
   Clearly these results reflect a number of               approach’, Preventive Veterinary Medicine, vol.
key assumptions made in constructing the                   23, pp. 9–32.
disease scenarios. For example, the benefits              Garner, M.G., Allen, R.T. and Short, C. 1997, Foot-
of ring vaccination may be greater if the time             and-Mouth Disease Vaccination: A Discussion
taken to detect the outbreak was increased.                Paper on Its Use to Control Outbreaks in Australia,
This would place greater stress on available               Bureau of Resource Sciences, Canberra.
resources and increase the likelihood that                Hassall and Associates 1993, Impacts of exotic
the disease would spread.                                  animal diseases on regional economies within
                                                           Australia, Unpublished report to the Exotic
   Conditions under which the spread of the
                                                           Animal Disease Preparedness Consultative
disease was slower might favor the imple-                  Council, Canberra.
mentation of the dangerous contacts control




Australian Commodities vol. 9 no. 1, March quarter 2002                                                 233
         Foot-and-Mouth Disease Outbreak:
                   Modelling Economic Implications for Australia1

                                                  Siobhan Dent
                         Department of Primary Industries, Queensland
This paper estimates the extent of the economic impacts that a Foot-and-Mouth
Disease (FMD) outbreak would have in Australia. Due to Queensland s reliance on
the beef cattle and, to a lesser extent, sheep industries, the net effects of an FMD
outbreak in Queensland would be far greater than that for the rest of Australia.
The livestock farming (mainly beef cattle, sheep, dairy and pig) and meat product
manufacturing (mainly beef, sheep meat and pork) industries are not the only
industries that would be adversely affected by an FMD outbreak. Over half of the
losses in jobs and industry output would occur in non-agricultural industries.
Due to the economic conditions created by an FMD outbreak, several industries
primarily associated with mineral exports could actually benefit as a result of the
outbreak.

Dedication
This report is dedicated to Russ Reynolds who passed away 22 October 2001.
Russ was General Manager of the Policy Analysis Unit in the Department of Primary
Industries, Queensland. He had much enthusiasm for generating information to
inform policy debates and was instrumental in this FMD study. Russ was heavily
involved from the conception of the project through to the analysis, interpretation
and presentation of the results.

Acknowledgments
The author would like to acknowledge the input from Mark O Sullivan and John
Switala.
This study also drew on expertise from various business groups within the
Department of Primary Industries, Queensland, particularly Animal and Plant Health
Services (Peter Black) and the Agency for Food and Fibre Science (John Hargreaves
and Bob Armstrong).
The Office of Economic and Statistical Research, Queensland Treasury, provided
valuable comments on the assumptions imposed on the Monash Multi Regional
Forecasting (MMRF) model, and the interpretation of the results from the model.
Greg Watts, in particular, extensively reviewed this report from a technical aspect
and provided many useful comments.
The Department also wishes to acknowledge Philip Adams from the Centre of Policy
Studies, Monash University, for advice and producing the simulation results reported
in this report.



1
 In this study it was assumed that the beef cattle, sheep, pig and dairy industries, and associated products, would
be at risk from a FMD outbreak.


                                                        1
Introduction
There has not been an outbreak of FMD in Australia for 130 years. However, the
issue has resurfaced in the aftermath of the recent outbreak in Great Britain.
The majority of Australian beef exports (around 90%) are sold to FMD-free
countries, predominantly around the Pacific Rim. These markets accept meat exports
only from FMD-free countries and, as a consequence, suppliers receive a price
premium. An FMD outbreak in Australia would result in the immediate closure of
these FMD-free export markets to Australian sourced live cloven foot animals, as
well as to beef, sheep meat and pork products.2
Our knowledge of the likely economic impacts of an FMD outbreak in Australia has,
to date, been limited to the direct/onfarm effects. The recent outbreak in Great
Britain illustrated that, although the impacts of an FMD outbreak are significant on
the livestock farming and meat processing industries, they are minor in comparison
to the flow-on effects to other sectors in the economy.
The objective of this study is to ascertain the likely impacts that an outbreak would
have on the directly affected at risk 3 livestock farming and meat processing
industries, and the likely flow-on effects to other industries and the economy as a
whole.

Methodology
The Centre of Policy Studies at Monash University was contracted by the
Queensland Department of Primary Industries to undertake the economic modelling.
The Queensland Department of Primary Industries identified the shock to be
modelled, specified the input to the model, and with the help of Queensland Treasury
and a technical reference panel undertook the analysis and reporting.
Specifically, the Monash Multi Regional Forecasting (MMRF) model was used. The
MMRF model is a multi-sector dynamic model of the Australian economy covering
the six States and two territories. It models each State in its own right and is therefore
ideally suited to determining the impact of region-specific economic shocks. Since
the MMRF model is dynamic, it is able to produce sequences of annual solutions
connected by dynamic relationships.
A hypothetical FMD outbreak
A number of possible outbreak scenarios were modelled but, after reviewing
previous studies, it was decided that the following worst-case scenario would be
reported in this paper, namely a major FMD outbreak that occurred close to
Brisbane without the implementation of zoning4. The assumed outbreak would result
in the loss of access to FMD-free export markets for six years for all Australian
sourced live cloven foot animals, as well as to beef, sheep meat and pork products.


2
  It is expected that an FMD outbreak may also result in the closure of FMD free markets to minimally processed
dairy products.
3
  In this study it was assumed that the beef cattle, sheep, pig and dairy industries, and associated products, would
be at risk from a FMD outbreak.
4
  The magnitude of the impact of export market closures on the Australian agricultural industry and economy
would be influenced by whether effective zones around the infected regions were put in place. The
implementation of such zones could allow the uninfected regions of Australia to retain their exports to FMD-free
countries and could therefore lessen the impact on the Australian agricultural industry and economy.


                                                         2
The assumed producer response
The effects of an FMD outbreak on Australian at risk livestock industries are
discussed below and are based on previous work by Lembit, M.J. and Fisher, B.S.
(1992), who examined the likely direct/onfarm effects of a FMD outbreak for
broadacre agriculture.
Phase 1 (Year 1): The announcement of an outbreak would result in the immediate
closure of all FMD-free export markets to Australian at risk livestock and
livestock products. These export market restrictions would effectively reduce the
demand for Australian at risk livestock and livestock products (see figure 1 an
inward shift of the demand curve for livestock and livestock products), which would
then push down the domestic price of livestock and livestock products (P to P1).
Phase 2 (Years 2 and 3): In the initial year of the outbreak, livestock producers
would hold back their animals from slaughter in the hope that prices would improve
(inward shift of the supply curve). However, due to the collapse in demand in Phase
1, as low prices prevailed, producers would increase their livestock turnoff and
consequently the supply of livestock and livestock products would increase (an
outward shift in the supply curve for livestock and livestock products). Greater
supply would cause the domestic price of livestock and livestock products to fall
further.
Phase 3 (Years 4 to 6): The increased slaughterings in Phase 2 would lead to a
significant decline in herd size and consequently slaughterings would fall. This
would cause the supply of livestock and livestock products to decline (an inward
shift of the supply curve for livestock and livestock products), which would place
upward pressure on the domestic price. The higher price of livestock would
encourage producers to withhold stock for breeding. This would result in a further
reduction in the supply of livestock and livestock products, and higher domestic
prices.
Phase 4 (Year 7): Once Australia was declared FMD-free, market access to FMD-
free countries would be regained. Demand for Australian livestock and livestock
products would be restored (an outward shift of the demand curve for livestock and
livestock products), and the domestic price of livestock and livestock products would
increase. Severe supply constraints would result in a further inward shift of the
supply curve for Australian livestock products, which would place further upward
pressure on the domestic price of livestock and livestock products.
Phase 5 (Years 8 to 15): Livestock producers would respond to the prevailing high
domestic prices by increasing herd sizes. However, there would be a biological lag
time between when the decision was made to increase the herd and when the increase
in the supply of livestock was realised. Therefore, once livestock numbers grew,
producers would increase their slaughterings and the supply of livestock and
livestock products would increase (an outward shift in the supply curve of livestock
and livestock products). As supply continued to rise, the domestic price would begin
to fall, and approach its original position (other things being equal).




                                          3
Figure 1: At Risk livestock and livestock product industries response to an FMD
                            outbreak Phase 1
                                       Price
                                                                                 S



                                             P
                                             P1

                                                                                      D
                                                                                D1
                                                 0
                                                                Q1 Q                       Quantity

To model the hypothetical FMD outbreak, the likely impact the outbreak would have
on the quantity of livestock and livestock products exported from each State5 was
entered into the model. The assumed deviations in export volumes are shown in table
1 and represent the percentage change in export volumes from what would have
occurred without an FMD outbreak (base case).
Table 1: Changes in the export volume of livestock and livestock products
         (percentage deviations from base case)
                                2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

     NSW                        -18.       3.7       9.4 -17. -30. -34. -33. -30. -25. -21. -17. -12.4 -7.5 -3.3 0.0

     Vic                         -4.   -0.6 0.3 -7.             -8.4 -9.        -7.1 -6.3 -5.            -4.   -3.   -2.6 -1.6 -0.8 0.0

     Qld                        -28.       6.1 14.4 -22. -41. -48. -49.2-45.4-38. -31. -25. -18.2-10.                            -4.6 0.0

     SA                          9.8       0.9       0.7 -4.    -5.       -3.   -0.8 0.2        0.0 -0.2 -0.4 -0.5 -0.5 -0.7 0.0

     WA                          1.8       0.7       1.9 -8. -11. -10.          -8.       -6.5 -5.       -4.   -4.   -3.   -2.   -1.3 0.0

     Tas                        -17.       3.4       8.1 -14. -25. -30. -29. -27. -23. -19. -15. -11.                      -6.7 -2.8 0.0


Results
Effects on the national and Queensland economy
    Figure 2: Real GDP and national employment
                       0.4                                     Real GDP Employment
                       0.3
      Deviation from




                       0.2
      base case (%)




                       0.1
                         0
                       -0.1 0    1     2     3        4   5     6     7     8    9        10   11   12    13   14    15

                       -0.2
                       -0.3
                       -0.4
                       -0.5                                   Year after the outbre




5
 The export volume assumptions were based on previous work completed by Lembit, M.J. and Fisher, B.S.
(1992).


                                                                       4
National effects
In the first year of the outbreak, it was assumed that overseas demand for Australian
livestock and livestock products would fall. The reduction in export demand was
assumed to result in a reduction in export volumes of livestock (-9.7%) and livestock
products (-14.6%). These reductions were predicted to result in a direct fall in real
(adjusted for inflation) Gross Domestic Product (GDP) of 0.2% below the base case.
However, the model forecasts that this direct effect on real GDP would be mitigated
by a predicted fall in domestic prices.6 Overall, real GDP was predicted to fall by
0.05% ($240 million) below the base case (see figure 2). The model assumed that the
demand for labour would adjust in response to the external shocks, and employment
was projected to decline (-0.06% or 5300 people) in the first year of the outbreak as
the real cost of labour increased7.
In Year 2 and Year 3 of the outbreak, it was assumed the quantity of livestock and
livestock products exported would increase. This was achieved by an increase in the
activity (real value added) of Australia s livestock farming and livestock products
industries through the liquidation of national herd to discounted markets. Directly, it
was estimated that the projected increase in activity of these industries in Year 2
would lead to a 0.1% increase in real GDP above the base case. The model projected
that overall real GDP would increase by 0.2% ($1230 million) above the base case in
Year 2, while national employment was projected to increase to 0.1% (10 100
people) above the base case8
High levels of slaughterings in Year 2 and Year 3 would reduce the size of the
national herd. The assumed supply contraction in the livestock farming and livestock
products industries was projected to lead to a fall in real GDP between Year 4 and
Year 6 of the outbreak. Real wages would take time to adjust to the external shock,
and employment was projected to fall below the base case. In Year 6, real GDP was
projected to be 0.3% ($1430 million) below the base case and employment was
projected to be 0.2% (16 700 people) below the base case.
In Year 7, market access would be regained to FMD-free markets, and demand for
Australian livestock and livestock products, it was assumed, would return to its
original level. This was projected to lead to an increase in the price of livestock and
livestock product exports, and an improvement in the terms of trade as export prices
improved relative to import prices. An improvement in the terms of trade, all other
things held equal, would be expected to increase real GDP (through an increase in
income). However, the assumed slaughtering-out of livestock in Year 2 and Year 3,
and the biological lag time involved when rebuilding herds, meant that the supply of


6
 The fall in demand for at risk Australian livestock and livestock products was projected to translate into lower
domestic incomes, which were projected to lead to a reduction in general domestic demand and in turn to lower
domestic prices.
7
  The projected initial rise in the real cost of labour would arise from a combination of wage bargainers
attempting to maintain real wage outcomes and the projected terms of trade deterioration. A terms of trade
deterioration would increase the price of expenditure (e.g. CPI) relative to the price of output (e.g. GDP deflator).
Thus, with nominal wages assumed to be sticky in the short term and effectively indexed to the CPI, terms of
trade deterioration would lead to an increase in the price of labour relative to the price of output.
8
  National employment was projected to increase due to several factors including: (1) increased activity in the
livestock farming and livestock products industries; (2) an easing of wage bargaining pressures due to the
previous year s fall in employment and increases in domestic output prices to CPI; and (3) a fall in the price of
labour relative to the rental price a capital.



                                                          5
livestock and livestock products would not be sufficient to satisfy the resumed
demand.
The continued output contractions in the livestock farming and livestock products
industries alone would be expected to lead to a direct decline in real GDP of 0.14%
in Year 7. The model predicted that overall real GDP would decline by 0.45%
($2400 million) below the base case.9 The Department of Primary Industries,
Queensland, estimated that a major FMD outbreak near Brisbane would incur control
costs of approximately $500 million, which are only minor in comparison to the
forecast $2400 million loss in real GDP in Year 7 alone. The model also predicted
that, in Year 7, activity of the national livestock farming industry would be
approximately $700 million below the base case, while activity of the national
livestock product industry was forecast to be approximately $200 million below the
base case. These losses, when combined, would be less than 40% of the projected
loss in real GDP ($2400 million).
In Year 7, national employment was projected to be -0.2% (21 800 people) below the
base case.
Over time, the size of the national livestock herd would increase, and the activity of
livestock farming and livestock product industries would increase and approach the
base case. Real GDP was projected to increase as the activity of these industries
increased, and was forecast to return to the base case in Year 13. National
employment was projected to return to the base case in year 1010.
Queensland effects
Figure 3: Queensland real GSP and employment
     Deviation from base case




                                3.0                             Real GSP Employment

                                2.0

                                1.0

                                0.0
             (%)




                                       0   1   2   3   4   5     6   7       8   9   10   11   12   13   14   15
                                -1.0

                                -2.0

                                -3.0

                                -4.0                           Year after the outbre

The large size of the Queensland livestock farming and livestock product industries
(approximately 50% of the national beef cattle herd is in Queensland), along with the
large assumed changes in the export volumes (and subsequent decline in prices) of
these industries, meant that the effect of the outbreak would be greater (in percentage
terms) in Queensland than in the rest of Australia. Queensland real Gross State

9
  Activity from livestock farming and livestock products was projected to decline by a greater proportion than
their use of factor inputs. Reflecting that the livestock farming industry would still have to utilise labour and
capital to build up their herds and the livestock products industry would withhold labour and capital for use when
the supply of live animals resumed. The withholding of resources by these industries would mean that resources
would not be freed up for use by other industries, restricting their output growth, which would therefore
exacerbate the decline in real GDP.
10
   The more moderate outcome for employment would be caused by a reduction in the real wage rate. It is an
assumption in the model that, over time, real wages adjust so that the effect of an external shock on employment
is zero in the long term. Therefore, by assumption, the negative labour market effects of the FMD outbreak would
be reductions in real wages rather than reductions in employment.



                                                                         6
Product (GSP) was forecast to reach its lowest point relative to base case in Year 7 (-
2.8% or $2340 million below the base case). In the same year, employment in
Queensland was projected to be 2.2% (33 900 people) below the base case.
Industry effects — Australia wide and specific to Queenlsand
Livestock farming and livestock product industries
In 1999—2000, beef cattle constituted 52% of Queensland s livestock herd and 16%
of the national livestock herd. It was assumed that the closure of FMD-free-export
markets would have a greater impact on beef cattle and beef product exports than on
other livestock and livestock product exports. Therefore, the assumed impact on
Queensland livestock farming11 export volumes would be more pronounced than the
impact on national livestock farming export volumes (see figure 4).
Figure 4: Livestock farming export volumes
                                                           20.0                                                AustraliaQueensland
                                Deviation from basecase




                                                           10.0
                                                            0.0
                                                                  0       1       2       3       4       5        6       7       8       9    10    11    12    13    14    15
                                                          -10.0
                                        (%)




                                                          -20.0
                                                          -30.0

                                                          -40.0
                                                          -50.0

                                                          -60.0                                               Year after the outbre

As per the discussion on page 3, in the initial year of the outbreak, the volume of live
animal exports would decline due to the closure of FMD-free markets. The closure of
these markets would place downward pressure on livestock prices. It was assumed
that as low prices continued in Year 2 and Year 3, producers would increase their
turnoff and the quantity of live animal exports to FMD endemic markets would
increase. By Year 4, livestock herds would be diminished and the volume of live
animal exports would decline as producers withheld stock for breeding. Market
access would be restored in Year 7, but assumed constraints on the supply of
livestock would mean that quantity of livestock exported would not begin to return to
the base case until Year 8. It was assumed export volumes would reach the base case
in Year 15.
 Figure 5: Livestock farming (beef cattle, sheep, dairy, pigs and poultry) activity
     Deviation from base case




                                                          40.0                                                AustraliaQueensland
                                                          30.0
                                                          20.0
                                                          10.0
                                         (%)




                                                          0.0
                                                      -10.0 0         1       2       3       4       5        6       7       8       9       10    11    12    13    14    15

                                                      -20.0
                                                      -30.0
                                                      -40.0
                                                      -50.0                                               Year after the outbre



11
  The livestock farming industry mainly includes the farming of beef cattle, sheep, milk cattle, poultry, pigs and
commercial fishing.


                                                                                                                           7
Output from the national and Queensland livestock farming industries (mainly beef
cattle, sheep, dairy, pigs and poultry) was projected to deviate from the base case in a
similar way to the assumed export volume deviations. However, projected declines in
domestic meat product prices in the first four years after the outbreak would lead to
an increase in the domestic demand for livestock farming output. Therefore, the
negative deviations in activity were forecast to be more moderate than the assumed
export volume deviations. Queensland livestock farming activity was projected to be
39.5% ($770 million) below the base case in Year 7, while National livestock
farming activity was projected to be 8.3% ($700 million) below the basecase.
Figure 6: Livestock farming (beef cattle, sheep, dairy, pigs and poultry) employment
                 Deviation from base case




                                             10.0                                                      AustraliaQueensland

                                              5.0

                                              0.0
                                                      0       1       2       3       4        5        6       7       8       9    10    11    12    13    14    15
                                             -5.0
                         (%)




                                             -10.0

                                             -15.0

                                             -20.0

                                             -25.0                                                 Year after the outbre

Initially, the FMD outbreak was projected to cause employment in the livestock
farming industry to decline as price and output reductions increased the real cost of
employing labour. As livestock turnoff increased in the second and third years after
the outbreak, employment was projected to increase. However, the severe supply
contractions between the fourth and seventh year were projected to lead to a
significant decline in employment. In Year 7, employment in Queensland livestock
farming was projected to be 21.6% (13 000 people12) below the base case, while
employment in the national industry was projected to be 0.7% (1900 people) below
the base case.
The deviations in Queensland s livestock product 13 export volumes were assumed to
be identical to the changes in the State s livestock export volumes (see figure 4).
However, the national changes in livestock product export volumes were projected to
be greater than the national changes in livestock export volumes (not shown). This
reflects the concentration of livestock product industries in States that were assumed
to be more affected by the outbreak (e.g. Queensland).
Figure 7: Livestock product (predominantly meat processing) activity
 Deviation from base case




                                     20.0                                                      Australia Queensland

                                     10.0

                                            0.0
                                                  0       1       2       3       4       5        6        7       8       9   10    11    12    13    14    15
         (%)




                                 -10.0

                                 -20.0

                                 -30.0

                                 -40.0                                                        Year after the outbre
12
   Changes in the people employed in each industry were estimated by multiplying the employment in the
industry in 1999—2000 by the projected percentage impact on industry employment by the MMRF model.
13
   The livestock product industry includes meat/meat product manufacturing, and dairy product manufacturing.



                                                                                                                8
Similarly to the livestock farming industry, the negative deviations in the livestock
product industry s activity were forecast to be more moderate than the assumed
deviations in export volumes. In Year 7, Queensland livestock product industry
activity was projected to be 33.6% ($184 million) below the base case, while the
national industry s activity was projected to be 9.5% ($200 million) below the
basecase.
Figure 8: Livestock product (predominantly meat processing) employment
     Deviation from base case




                                15.0                              AustraliaQueensland
                                10.0
                                 5.0
                                 0.0
                                -5.0 0      1    2   3   4   5     6    7   8   9       10    11    12    13   14    15
             (%)




                                -10.0
                                -15.0
                                -20.0
                                -25.0
                                -30.0
                                -35.0                            Year after the outbre

Employment in the Queensland livestock product industry would follow a similar
trend to employment in the Queensland livestock farming industry. In Year 7,
employment in the Queensland livestock product industry was projected to decline to
31%, or 4200 people, below the base, while employment in the national industry was
projected to be 8.2%, or 5100 people below the basecase.
 Export-orientated industries
Previously, discussion has focused on the negative impacts that an FMD outbreak
would have on the livestock farming and livestock product industries. However, as a
result of the projected general increase in Australia s international competitiveness
(arising from a currency depreciation) and resource reallocation within the economy
toward more profitable industries, certain industries would benefit. Such industries
include those that are highly exposed to international trade and have few direct
linkages with the adversely affected livestock and food industries. Examples of
export orientated Queensland industries are black coal mining,14 aluminium/alumina
and magnesium mining,15 and air transport.16 Output from these industries is
projected to generally increase above base-case levels, with the extra output being
internationally exported (see figure 9).
Figure 9: Export volumes - Queensland export-orientated industries
     Deviation from base case




                                8.0                                                   Air
                                                Black coal Aluminium/alumina & magnesium transpor
                                7.0
                                6.0
                                5.0
                                4.0
                                3.0
             (%)




                                2.0
                                1.0
                                 0.0
                                -1.0    0   1    2   3   4   5      6   7   8       9    10    11    12    13       14    15
                                -2.0
                                -3.0                             Year after the outbre
14
   The black coal industry includes the mining of black coal (thermal and metallurgical).
15
   The aluminium/alumina and magnesium industry mainly includes alumina production, aluminium smelting,
and aluminium rolling, drawing and extruding.
16
   The air transport industry mainly includes international and domestic air transport.



                                                                        9
Export volumes of each of these industries were projected to reach their highest point
relative to base case values in Year 7. Air Transport17 would have the largest
percentage increase in export volumes from base case levels.
Figure 10: Employment Queensland export-orientated industries
     Deviation from base case



                                5.0                                                  Air
                                               Black coal Aluminium/Alumina & magnesium transpor

                                4.0

                                3.0

                                2.0
             (%)




                                1.0

                                0.0
                                       0   1    2   3   4   5      6   7    8   9   10   11    12    13    14    15
                                -1.0

                                -2.0                             Year after the outbre
Employment levels were projected to change with output and exports in response to
changes in the real cost of employing labour (see figure 10). Employment in the
Queensland black coal and aluminium/alumina and magnesium industries was
projected to peak relative to base case, seven years after the outbreak.
Service industries
Service industries constitute around 70% of the Australian and Queensland
economies (in terms of output value). Due to the large size of these industries, it is
valuable to gain an understanding of the possible impact an FMD outbreak might
have on them.
Figure 11: Activity - Queensland service industries
                                       Trade hotels Public Servic                                 Construction
                                                                       Financial & business service
     Deviation from base case




                                3.0

                                2.0

                                1.0
             (%)




                                0.0
                                       0   1    2   3   4    5     6    7   8   9   10    11    12    13    14    15
                                -1.0

                                -2.0

                                -3.0                             Year after the outbre


The activity of major Queensland service industries was projected to slightly decline
in the first year of the outbreak and then increase above base case levels in the
second and third years of the outbreak.18 The activity of these industries was
projected to decline between the fourth and seventh years of the outbreak and was
then projected to recover to base case levels by Year 15.


17
  The projected increase in the export volumes and employment in air transport reflects the increase in
Australia s international competitiveness and resource reallocation within the economy. This simulation did not
examine the impact of a change in consumer tastes away from travelling to a country with an FMD outbreak, as
was the case in Great Britain.
18
   The projected increase in output above the base case in Year 2 and Year 3 reflects the assumption that
producers would increase their turnoff of livestock over this period in response to prevailing low prices, which in
turn would increase the output of meatworks.


                                                                       10
The Queensland construction19 industry would have the largest percentage output
deviations from base case values. Queensland construction s output troughed in the
eighth year after the outbreak at 2.8% (or $203.4 million) below base case values.20
Figure 12: Employment - Queensland service industries
                                       Trade/Hotels Public Servic     Financial/Business services Construction
     Deviation from base case




                                3.0

                                2.0

                                1.0

                                0.0
             (%)




                                       0   1    2   3   4   5     6    7   8   9   10   11   12   13   14   15
                                -1.0

                                -2.0

                                -3.0

                                -4.0                            Year after the outbre


Employment in Queensland s major service industries was projected to move in the
same direction as the deviations in output. Seven years after the outbreak, the
deviations in employment in trade and hotels,21 financial and business services,22 and
construction were greater than the deviations in output.23

Conclusions
Models are an accepted way to analyse complex issues, providing their limitations
are understood.
The modelling in this paper highlights that more than half of the impacts of a major
FMD outbreak would occur in industries (e.g. construction, financial business
service, retail trade and hotels) other than the at risk livestock farming and
livestock product industries. In Year 7 alone, real GDP was projected to be $2,400
million below the base case, while employment was projected to be 22 000 people
below the basecase. These losses are far greater than the projected Year 7 losses in
the national livestock farming industry ($700 million and 1900 people) and the
national livestock product (mainly meat processing) industry ($200 million and 5100
people).

19
   The construction industry mainly includes the construction of houses, other residential buildings, non-
residential buildings, roads and bridges.
20
  The greatest impact on construction output was projected to occur in Year 8 because the assumed withholding
of livestock between Years 4 and Year 7 would lead to a decline in the rates of return on capital. This would lead
to a projected decline in investment to Year 7, and because it was assumed that a one-year lag would exist
between changes in investment decisions and changes in capital stock, construction output was projected to fall to
Year 8.
21
  The trade/hotels industry includes wholesale trade, retail trade, accommodations, pubs, taverns and bars, cafes
and restaurants, and clubs.
22
  The financial/business services industry includes financial institutions, insurance services, legal and accounting
services, employment services, computer services, marketing and business management services.
23
   The projected declines in output from these industries would lead to reductions in the demand for capital and
labour. The slow adjustment of capital stock for each industry accompanied by the rapid adjustment of wages
(relative to the change in capital stock) and decreasing growth (relative to base case) in the labour supply
(partially offset by labour shedding in the livestock farming and livestock product industries) implies that the
rental price of capital would decline relative to the wage rate. Businesses would therefore substitute capital for
labour, explaining the greater projected decline in employment than output in Year 7.



                                                                      11
The impact of the outbreak on Queensland, a major beef cattle state, was
proportionally more severe than the national impact. Queensland employment was
projected to be 33 900 people below the basecase and real GSP projected to be $2
340 million below the basecase in Year 7 alone.
The Department of Primary Industries, Queensland estimated that a major FMD
outbreak in Brisbane would incur total control costs of approximately $500 million.
The costs of control appear significant, but are only minor when compared to the
likely loss to the national economy.
The figures above are based on the assumption that a major FMD outbreak would
result in the closure of FMD-free export markets to Australia for a period of six
years. However, continuing low prices would prompt producers to increase
slaughterings in an effort to maintain incomes. The increase in slaughterings would
lead to a reduction in the size of the national livestock herd to such an extent that,
when market access was regained in Year 7, there would be an insufficient supply of
livestock and livestock products to satisfy the restored demand. It was assumed that
it would take some time to rebuild the national herd, and it would not be until fifteen
years after the outbreak that export volumes would return to the base case level.
Our knowledge of the likely economic impacts of an outbreak of FMD in Australia
has, to date, been limited to the direct/on-farm effects. However, this study highlights
that not only the livestock farming and livestock product industries would be affected
by an outbreak of FMD. A number of export-orientated industries, which have few
direct linkages with the adversely affected livestock and food industries, would
benefit from the projected general increase in Australia s international
competitiveness, and the reallocation of resources within the economy toward more
profitable industries.


A detailed technical report is available by contacting the author:

Mrs Siobhan Dent
Project Officer
Business Strategy Unit
Rural Industry Business Services Group
Department of Primary Industries, Queensland
E-mail: siobhan.dent@dpi.qld.gov.au


References
Lembit, M.J. and Fisher, B.S. (1992) The economic implications of an outbreak of
      Foot-and-Mouth Disease for broadacre agriculture. ABARE paper presented
      at the National Symposium on Foot-and-Mouth Disease, Canberra, 8—10
      September 1992.




                                          12
    The drivers of Australia s productivity
                    surge∗

                                Gary Banks
                     Chairman, Productivity Commission


Key points
•   Australia s productivity growth surged to a record high in the 1990s — more than
    double the rate achieved over the 1980s. Australia s productivity surge was also
    very strong by international standards.
•   A new set of service industries — especially Wholesale trade and Finance &
    insurance — made major contributions to the 1990s productivity acceleration.
•   Australia was comparatively quick in adopting information and communications
    technologies (ICTs) in the 1990s and their use has featured in the productivity
    accelerations of the new service industry contributors.
•   Microeconomic reforms were pivotal in Australia s improved productivity
    performance, by sharpening incentives for businesses to be more productive and
    providing them with greater flexibility to adjust to a more competitive
    environment. Microeconomic reforms encouraged and assisted the uptake of ICTs
    and the transformation of industries in ways that tap new productivity potential.
•   In looking to the future, further productivity gains are possible from continued
    ICT uptake and business transformation, and Australia is well placed to benefit
    from e-commerce.
•   Policy will continue to play an important role — particularly in relation to labour
    market flexibility and the development of human capital (in the widest sense).


Introduction

Productivity is not only the key to the performance of firms and industries; it is
fundamental to the living standards of the general community. There is now general
recognition that Australia experienced a transformation in its productivity record
through the 1990s. The causes and industry origins of the productivity surge are
perhaps less well known or accepted. A stream of recent Productivity Commission
research has provided some useful insights, not only reaffirming the role of
microeconomic reform, but also revealing the mechanisms by which it has operated.


∗ Presented at Outlook 2002, Canberra, 7 March.


                                                                                    1
This paper draws heavily on that research to describe trends in Australia s
productivity; the role of different industries in the recent acceleration; how ICT has
played a part and the linkages to micro reform. But first, a few points on the concept
and measurement of productivity itself.


Measuring productivity

Productivity is, technically, the ratio of output produced to input used. Considered
more broadly, productivity measures capture the ability of a nation to harness its
physical and human resources to generate output (and income).

Productivity improvement can have connotations of minimising the use of inputs
for example, adopting production processes that eliminate waste or unnecessary costs.
But, equally and importantly, it can be thought of as maximising output using
resources in the production of goods and services that add most value.

Productivity is measured in two main ways:
•   labour productivity the ratio of output produced per unit of labour used; and
•   multifactor productivity the ratio of output produced per combined input of
    labour and capital (buildings, machinery and equipment, etc),

There are pluses and minuses with each of these measures and some care is needed in
interpreting them.

Labour productivity can be relatively easy to measure, whether it be at the level of a
firm, industry or national economy requiring only that output and the number of
employees or hours of work be quantified. (Measurement of output in some
industries, particularly services, is difficult.) Labour productivity reflects how well
resources are used in generating output, but there are some dangers in interpreting
labour productivity as an indicator of worker efficiency. There are so many factors
outside the control of workers the amount of capital available, changes in
technology, management expertise that affect output and therefore labour
productivity.

Multifactor productivity (or MFP) is conceptually a better measure of efficiency of
resource use than labour productivity, since it includes the two major elements
labour and capital on the input side. Improvements in efficiency are a major
contributor to improvements in labour productivity and growth in per capita incomes.
Indeed, the Commission has calculated that MFP growth accounted for about half the
labour productivity and average income growth from the mid-1960s to the end-1990s
(Parham, Barnes, Roberts and Kennett 2000).




                                                                                     2
In practice, obtaining a reliable measure of capital input for a firm or industry can be
problematic. However, the ABS has introduced new methods and refinements in
recent years that have improved the measurement of capital input at the aggregate and
broad industry levels.

In line with much of the Productivity Commission s work, this paper focus on
multifactor productivity.


Trends in Australia’s productivity

Figure 1 illustrates three phases of Australia s productivity performance over the
second half of the 20th Century:
•   strong productivity growth in the post-war period of reconstruction and expansion,
    through to the mid-1970s;
•   a pronounced deceleration from the late-1970s through to the early 1990s; and
•   a renewed surge in productivity growth from the early 1990s.

Figure 1     Australia s multifactor productivity, 1964-65 to 2000-01

       100
                                                                                                    1999-00

                                                                                     1993-94
        90
                                                                    1988-89
                                                         1984-85
                                              1981-82


        80
                              1973-74


                  1968-69
        70


                                                Actual             Trend

        60
        1964-65     1969-70      1974-75   1979-80        1984-85          1989-90       1994-95   1999-00

Source: ABS 5204.0 and unpublished estimates.


The 1990s surge peaked in 1999-00. There was a downturn in 2000-01, as the
economy (output growth) slowed and hours worked declined slightly, but the
measured capital input continued to rise at its earlier pace. This downturn in
productivity is more likely to be short-lived than indicative of a marked slowing in
underlying trend, for reasons that I will come to.

It is apparent from the chart that the 1990s presented the longest period of continuous
positive growth in MFP on record. Clearly, there was much more at work than a
cyclical uplift out of the early 1990s recession.


                                                                                                              3
The peak years in figure 1 define what is know as productivity cycles . The 1990s
cycle, for example, started in 1993-94 and ended in 1999-00. The ABS calculates the
average rates of peak-to-peak growth and publishes them as indicators of underlying
rates of productivity growth. Importantly, the use of peak-to-peak productivity cycles
ensures that business cycles do not have spurious effects on estimates of productivity
trends.1

Figure 2 shows the underlying rates of productivity growth over productivity cycles,
as computed by the ABS. The rate of growth in labour productivity is indicated by the
height of each column. The rate of multifactor productivity (MFP) growth is indicated
by the black portion at the bottom of each column. The lighter shade in the top portion
of each column represents the rate of capital deepening indicating increases in the
use of capital per unit of labour.

Figure 2     Growth in labour productivity and multifactor productivity over
             productivity cycles, 1964-65 to 1999-00

             5

                                               MFP growth         Capital deepening

             4


                                 2.9
                                                                                                     3.0
             3
                    2.5
                                              2.4
                                                            2.2
                                                                                                      1.2
                                 1.4                                                     2.0
             2      1.3
                                              1.4
                                                            1.4
                                                                                         1.3
                                                                           0.8
             1
                                                                                                      1.8
                                 1.5
                    1.2                                                    0.4
                                              1.0
                                                            0.8                          0.7
                                                                           0.4
             0
                 1964-65 to   1968-69 to   1973-74 to   1981-82 to      1984-85 to    1988-89 to   1993-94 to
                  1968-69      1973-74      1981-82      1984-85         1988-89       1993-94      1999-00


Source: ABS 5204.0 and PC estimates.


Two important points are apparent from the chart:
•   Firstly, it has been variations in MFP growth in other words, efficiency that
    have almost wholly accounted for variations in labour productivity growth. The
    rate of capital deepening has been fairly constant across the cycles (apart from the
    period of strong employment growth in the late 1980s).




1 In contrast, Quiggin (2001a) makes productivity comparisons between business cycles. This
  can obscure important information about productivity growth that occurs independently of
  the business cycle, as happened in the 1990s. Quiggin (2001b) also claims that the 1990s
  surge in productivity is partly due to increases in capacity utilisation. But the strong growth
  in capital in the 1990s was more consistent with an increase in capacity than in its rate of
  utilisation (Parham, 2001).


                                                                                                                4
•     Secondly, the underlying rates of productivity growth both for labour and MFP
       were at record highs in the 1990s cycle. MFP grew at an annual average rate of
      1.8 per cent, just over 1 percentage point (or 2_ times) higher than the previous
      average from the early 1980s.

Australia s productivity performance in the 1990s was also strong by international
standards. Figure 3 shows that Australia was one of only three developed countries to
experience a strong productivity acceleration in the 1990s.2 Furthermore, unlike the
so-called Golden Age of the 1960s, there was no worldwide productivity boom in the
1990s.

The productivity acceleration in the US has received much attention, partly because of
the importance of the US economy, but also because it appeared unexpectedly at a
stage in the business cycle when a slowdown in productivity growth would normally
have occurred. However, according to the Productivity Commission s investigations,
Australia s productivity surge started earlier than that of the United States and was of
much greater magnitude (Parham, Roberts and Sun 2001).

Figure 3         Changes in trend multifactor productivity growth in the 1990s in
                 OECD countriesa

                 Finland
               Australia
                 Ireland
                Canada
                Sweden
               Denmark
                 Norway
            United States
            New Zealand
                Belgium
               Germany
                    Italy
             Netherlands
                 France
                  Japan
          United Kingdom
                  Spain

                       -2.0    -1.5     -1.0     -0.5     0.0      0.5        1.0     1.5   2.0

a
    Change in average annual rate of Productivity growth from 1980-89    to 1990-99

Source:          OECD 2001a




2 International productivity comparisons should allow for the fact that productivity can grow
  relatively fast in countries that are in a process of catching up to productivity leader
  countries. Australia outperformed the predicted rate of productivity growth that allows for
  catch-up among high-income OECD countries. This contrasts with the two earlier phases,
  when Australia s productivity growth was below the catch-up benchmark.


                                                                                                  5
The timing, strength and largely localised nature of Australia s productivity surge
suggest that there were some peculiarly Australian explanations at least in its early
stages. A closer examination of the industries responsible for the improved
performance provides some insights about those.


Industry contributors to the productivity surge

It should be said that industry productivity growth estimates need to be treated with a
little more caution than do aggregate measures. Productivity is particularly difficult to
measure accurately in some service industries. With that in mind, the industry
contributors to the 1990s productivity surge came from unexpected quarters.

Figure 4 shows MFP growth rates in industry sectors over the past two productivity
cycles. (Year-to-year estimates from 1974-75 are presented in appendix A.)

In the first cycle (1988-89 to 1993-94) there is evidence of relatively strong
productivity growth in the traditional contributors to aggregate productivity growth.
These are Agriculture and Mining, with their typically strong underlying growth
(notwithstanding some cyclical variation) and Manufacturing, which makes a major
contribution to aggregate productivity growth despite somewhat weaker
performance due to the sector s relatively large size.

These traditional sectors were joined in the 1980s and early 1990s by two other strong
performers: Communication services and Electricity, gas & water. Their improved
performance should not have been a surprise, given the major efficiencies achieved in
these largely government enterprises through the microeconomic reforms in that
period.

Figure 4: Industry MFP growth over the last two productivity cycles

           8.0
                                                     1988-89 to 1993-94                 1993-94 to 1999-00

           6.0


           4.0


           2.0


           0.0
                                                                               le


                                                                                          l
                                                     er
                  re




                                           g




                                                                                                                                                                    r
                            g




                                                                                         ai




                                                                                                                                                         s
                                                                  n




                                                                                                       s




                                                                                                                             ns
                                                                                                                  e




                                                                                                                                           e




                                                                                                                                                                  to
                                                                                                                                                     rv
                                        rin




                                                                                                   st
                                                                           sa
                           in




                                                                 io




                                                                                                             ag
                                                     at




                                                                                        et
                 tu




                                                                                                                                       nc
                                                                                                                            io




                                                                                                                                                               ec
                                                                                                  re
                       in




                                                                                                                                                    se
                                                             ct
                                                    w




                                                                          le


                                                                                     R
                                     tu
                 ul




                                                                                                             or


                                                                                                                       at


                                                                                                                                      ra
                       M




                                                            ru




                                                                                                                                                             ts
                                                                      ho
            ic




                                  ac




                                                                                               d
                                                d




                                                                                                                                                c.
                                                                                                           st


                                                                                                                       ic


                                                                                                                                  su
                                                          st




                                                                                             an




          -2.0
                                               an
             r




                                                                                                                                                          ke
                                                                                                                                               re
                                                                                                                   un
                                                                      W
          Ag




                                uf




                                                                                                         &
                                                       on




                                                                                                                                 in




                                                                                                                                                      ar
                            an




                                                                                          es
                                           as




                                                                                                                  m




                                                                                                                                           nd
                                                                                                    ns
                                                      C




                                                                                                                             d




                                                                                                                                                     M
                                                                                                              om
                           M




                                                                                        af




                                                                                                                            an
                                        ,g




                                                                                                                                          la
                                                                                                     a
                                                                                     ,c


                                                                                                  Tr
                                     ity




                                                                                                                                      ra
                                                                                                             C


                                                                                                                        e
                                                                                    m




                                                                                                                       nc


                                                                                                                                     tu
                                  ric




                                                                                co




          -4.0
                                                                                                                   na


                                                                                                                                  ul
                                ct




                                                                               Ac




                                                                                                                                 C
                              e




                                                                                                                  Fi
                           El




          -6.0

Source: PC estimates based on unpublished ABS data.




                                                                                                                                                                        6
It is notable that while productivity growth remained relatively strong in these sectors
in the 1990s cycle (except for Manufacturing), they all experienced a deceleration
compared with the previous cycle. None made a contribution to the productivity surge
in the post-1993-94 cycle.

The stand-out performer in the more recent period was Wholesale trade, although
other service industries also increased their rate of productivity growth.

Figure 5 depicts the extent to which different industries contributed to the increase in
the rate of aggregate productivity growth from the first cycle (represented in the left-
most column) to the second (the right-most column). An industry s contribution to the
productivity acceleration reflects both its productivity growth rate and its relative size.
Those industries on the left side of the arch are positive contributors to the
acceleration, whereas those on the right are detractors. The chart shows the new
contributors to aggregate productivity growth to have been the following service
industries:
•     Wholesale trade (overwhelmingly);
•     Construction
•     Finance & insurance;
•     Transport & storage; and
•     a few other minor contributors.

The traditional or 1980s contributors made negative contributions in the 1990s,
particularly Manufacturing.

Figure 5: Industry contributions to the acceleration in aggregate productivity
          growth between the last two productivity cycles

               3



              2.5



               2



              1.5



               1



              0.5



               0
                                                                                                    l
                                le




                                                                                            s
                    4




                                                                                                                                                                                               0
                                                                                                                                                                    er
                                                                                                                                    re
                                                                                                                        g
                                                                                                    ai




                                                                                                                      ns
                                                n




                                                                                                                                                                                               g
                                                                             e




                                                                                                                                                      s
                                                                                          st
                                                              e
                  -9




                                                                                                                                                                                            -0
                                                                                                                     in
                                               io




                                                                                                                                                                                           rin
                                                                                                                                                  rv
                             sa




                                                                         ag




                                                                                                et




                                                                                                                                                                 at
                                                                                                                                   tu
                                                             nc




                                                                                                                   io
                                                                                       re




                                                                                                                  in
                                           ct
                93




                                                                                                                                                                                        99
                                                                                                                                                se
                                                                                                R
                            le




                                                                                                                                                                w
                                                                                                                                 ul




                                                                                                                                                                                        tu
                                                                        or




                                                                                                                 at
                                                         ra




                                                                                                         M
                                          ru




                                                                                      d
                        ho




                                                                                                                             ric
              19




                                                                                                                                                                                     19
                                                                                                                                                                                     ac
                                                                       st




                                                                                                                                                             d
                                                                                                               ic




                                                                                                                                             c.
                                                                                  an
                                                        su
                                      st




                                                                                                                                                           an
                                                                                                             un



                                                                                                                            Ag



                                                                                                                                          re
                        W




                                                                                                                                                                         uf
                                                                    &
                                     on
          to




                                                                                                                                                                                  to
                                                    in




                                                                                 es




                                                                                                                                                                     an
                                                                                                            m




                                                                                                                                                        as
                                                                   s




                                                                                                                                        nd
                                 C
          9




                                                                                                                                                                                4
                                                    d


                                                               an



                                                                             af




                                                                                                          om




                                                                                                                                                                    M
        -8




                                                                                                                                                                              -9
                                               an




                                                                                                                                                      ,g
                                                                                                                                    la
                                                                          ,c
                                                              Tr
      88




                                                                                                                                                                            93
                                                                                                         C




                                                                                                                                                  ity
                                                                                                                                 ra
                                           e




                                                                         m
                                          nc
    19




                                                                                                                                                                          19
                                                                                                                               tu



                                                                                                                                                ric
                                                                       co
                                      na




                                                                                                                             ul



                                                                                                                                             ct
                                                                    Ac




                                                                                                                            C



                                                                                                                                            e
                                     Fi




                                                                                                                                         El




Source: Commission estimates based on unpublished ABS data.



                                                                                                                                                                                                   7
ICTs and the new service industry contributors

Part of the success of the new service industry contributors can be explained, at least
in an immediate or proximate sense, by their use of information and communication
technologies (ICTs).

There has been a lot of interest in ICTs in the context of the US productivity
acceleration. A key message that came out of early analyses of the US experience
emphasised the productivity gains that arose directly through the manufacture of
ICTs. More and more powerful microprocessors, computers and communications
equipment were produced, with little or no increase in inputs. (See Gordon (2000) as a
primary source.)

However, more and more evidence is emerging that the use or application of ICTs in
the United States has also been important in generating that country s additional
productivity growth (see, for example, Brynjolfsson and Hitt 2000, Stiroh 2001 and
McGuikin and van Ark 2001). The importance of use (as well as production) was also
a central finding in the OECD s recent Growth Project, which investigated the
reasons for productivity and output differences across countries (OECD 2001a).

The Australian case confirms that production of ICTs is not necessary to access so-
called new economy productivity gains. Comparative OECD (2001b) data show:
•   Australia s use of ICT equipment (hardware and software) to be relatively high
    among OECD countries:
    -   Australia had the third highest investment in ICTs as a proportion of business
        investment in 1999.
    -   Australia was also ranked third on Internet and e-commerce transactions (as
        measured by secure webservers).

•   However, Australia s production of ICTs, in contrast to uptake, is relatively small.
    -   Australia ranked only 20th on the proportion of Manufacturing devoted to
        ICTs.
•   On infrastructure indicators (for example, access paths and broadband availability)
    Australia scores a middle ranking (around 10th-13th).

The new service industry contributors to the productivity surge have links to the use
of ICTs. But the links are complex. ICTs are general-purpose or enabling technologies
that provide a platform for other innovations. As with other enabling technologies
(like electricity) the big productivity gains do not come immediately or directly from
the technology s availability, but from it being combined with complementary
innovations, including in new products and production processes. These take time.
They will generally require additional fixed investments and different work
arrangements.


                                                                                           8
Recent Productivity Commission research (Parham, Roberts and Sun 2001) has
highlighted the following points:
•   The use of ICTs and their contribution to aggregate productivity growth
    accelerated in the second half of the 1990s.
•   Finance & insurance is a relatively high user of ICTs (on all indicators). So is
    Wholesale trade, depending on which indicator is used.
•   The productivity gains in Finance & insurance appear to have been directly related
    to ICT use (development of new information-hungry products, for example,
    financial derivatives for risk-management; and automation of and electronic
    access to banking services). But looking across all industries there is no clear
    relationship between the intensity of ICT use and productivity growth.
•   In many cases ICT-related productivity gains have been indirect — by facilitating
    business transformation: in other words, changing what businesses do and how
    they do it.

The importance of this process of business transformation is well illustrated by earlier
Commission analysis of the productivity gains in Wholesale trade (Johnston et al
2000). It emerged that wholesalers were not necessarily becoming much more IT
intensive. Rather, they were using these technologies more productively. Bar-coding,
scanning and picking technologies, together with inventory management systems,
enabled businesses to streamline their operations and move away from storage-based
to fast flow-through systems, reducing the need for additional storage (capital) and
handling (labour).

A review of studies of IT use in US firms by Brynjolfsson and Hitt (2000) found that
gains varied widely at the firm level. Firms needed time to work out what could be
done and to undertake complementary investments in organisational change to
maximise the gains. This encompassed innovations in product lines, management
practices, work arrangements, supplier and customer relationships and so on.
Flexibility and adaptation were key ingredients. The authors state:
     As computers become cheaper and more powerful, the business value of
    computers is limited less by computational capability and more by the ability of
    managers to invent new processes, procedures and organisational structures that
    leverage this capability.

This appears to have been corroborated by a major study of IT use in US firms
conducted by McKinsey Global Institute (2001). McKinsey found that only in rare
cases did IT directly deliver large productivity improvements. Two examples in the
study are on-line retail securities trading and cellular equipment that enables better
use of the available spectrum, where the product or service itself was well suited to
IT. The study found that, in most cases, IT was only one among many tools that
managers used to redesign core business processes, products or services. Competitive
pressures were identified as the driving force behind such improved performance.


                                                                                      9
The microeconomic reform catalyst

This brings us back to microeconomic reform in explaining the surprising industry
origins of Australia s productivity surge in the 1990s.

The recent Australian experience has provided some important insights about
improving productivity performance:

•   Advances in technology are not the only source of productivity growth;

•   It is the rate of uptake of technology (not the mere existence of advances) that
    matters for improving productivity.

•   Other facilitating changes within firms are required to get the full productivity
    potential out of new technologies.

At each step, microeconomic reform has played a significant role. It is worth
elaborating briefly.


Advances in technology are not the only source of productivity growth

Economics textbooks often portray technological advance as the only or main source
of productivity growth. In the long run, all things being the same, this may be true.
But a significant contribution to Australia s improved productivity performance in the
past decade has come from reductions in production inefficiencies better
organisation of production — rather than new technology.

This is clearly illustrated by the improved performance of government business
enterprises (GBEs), which dominate the provision of Australia s economic
infrastructure. As noted, these include energy, water and communications, as well as
transport services. Microeconomic reforms have driven many of the productivity
improvements in these areas by bringing clearer commercial incentives and
disciplines to GBEs. The businesses have responded by more clearly defining what
they do, adjusting manning levels and improving investment decisions. The
Commission s recent case studies of GBE reform in transport and water services have
revealed strong productivity responses (PC 1999).

The gains in productive efficiency have not been limited to GBEs. Reforms
impacting on manufacturing have also played a significant role. The Vernon
Committee (1965) was an early catalyst for those reforms. It demonstrated that
Australia s productivity potential was not being fully tapped in Manufacturing
because a complex, made-to-measure tariff structure encouraged industry
fragmentation with small scale production oriented toward the confined domestic
market. Rationalisation of tariffs was expected to bring gains from specialisation and
scale, with production geared more toward export markets.


                                                                                   10
Recent case studies by the Productivity Commission have illustrated that subsequent
efficiency gains were linked to reductions in manufacturing protection. For example,
Whitegoods, TCF and PMV were all in a similar, highly-protected position in the late
1970s. But protection was lowered more quickly in Whitegoods through the 1980s
and its productivity improved more markedly than in the other two industries (PC
1999).


The importance of technology uptake

I have already noted that Australia moved quickly by international standards in its
uptake of ICTs in the 1990s. This rapid diffusion of technology is in marked contrast
to the sluggish behaviour of Australian firms in earlier decades.

In most developed countries, the manufacturing sector was an important focus of
innovation in the post-war boom period. Yet available technologies were not picked
up in Australia. The Jackson Committee of 1975 lamented this failure, in its review of
Australia s manufacturing sector:
    Much of the equipment in factories is old, inefficient and overdue for
   replacement; desirable technical innovations have been delayed; and the physical
   conditions for the workforce leave much to be desired.
   Australia s relatively poor performance can be explained by a variety of factors
   including poor labour relations, outdated or inappropriate technology, lack of scale
   economies and inadequate management techniques.

A senior business leader a few years later noted the implications of government-
supported resistance to change (Uhrig 1979, p. 5):
     during a very long period in which the absence of rapid change was assured,
   an evolutionary process had encouraged the development of managers who were
   best fitted for those circumstances. For this reason, the talents of a great many of
   our managers are administrative and bureaucratic rather than entrepreneurial.

The stronger competitive forces in more recent times have strengthened the incentives
to adopt technologies that can be used to upgrade productivity and competitiveness
and to develop new products and markets (PC 1999, Johnston et.al. 2000). That same
pressure has seen a turnover in senior managerial ranks, with greater rewards for
productive endeavours relative to rent seeking. In the process, and also reflecting the
more general shift to services, Australia has transformed itself from a technological
laggard into an advanced user of new technology.

An important Business Council of Australia study in the 1990s (Carnegie and Butlin
1993) noted:
   The internationalisation of the Australian economy and the new performance
   standards it requires are the predominate drivers of enterprise innovation. They
   have led to broad improvements in standards, moves to increase value to


                                                                                    11
   customers, the search for new products, the ability to turn problems of scale into
   competitive advantages and the successes of international niche marketing (pp.
   330-31).


The need for flexibility and adaptation to get the most out of new technology.

The clear message emerging from studies of firms use of ICT is that, generally
speaking, the large productivity gains do not come from bolting-on ICT to existing
modes of production. The gains come from using ICT as part of a process of business
transformation. Flexibility and adaptation are key.

Again, the case of Wholesale trade the stand out performer of the 1990s
illustrates the point. The benefits of transformation to fast flow-through systems in
that sector depended importantly on the increased flexibility in labour markets.
Industrial relations reforms promoted enterprise flexibility and autonomy, including
through the introduction of split shifts and reduced scope for demarcation disputes.

In contrast , the persistent labour market rigidities in much of Europe appear to have
reduced the potential for those countries to reap ICT-related productivity gains.


The gains from reform and productivity growth

The study of Australian wholesale trade also illustrates why the benefits of reforms
can show up in perhaps unexpected places. The distribution of motor vehicles is one
major area of wholesaling. With the increase in international competition (following
the reduction in trade barriers), domestic producers looked for savings all along the
 value-chain not just in the production of motor vehicles, but also in their
distribution and marketing. In other words, increased competitive pressure in
downstream markets can have a strong influence upstream. (This was also the story
of GBE reform.)

What has also emerged is that the benefits of productivity gains have not been wholly
absorbed by businesses, but have generally been passed on to users and consumers.
In particular, notwithstanding the major turnaround in the productivity of wholesale
trade, gross profit margins declined (Parham, Barnes, Roberts and Kennett 2000). It
seems that a more competitive environment not only drives more productivity gains, it
also means that the consumer receives more of the benefit from productivity gains
through lower prices.

The gains to the community from productivity growth have been substantial at the
national level. Growth in real average incomes of Australians accelerated from 1.4 per
cent a year in the 1980s to 2.5 per cent a year in the 1990s. MFP growth accounted
for over 90 per cent of that acceleration (Parham, Barnes, Roberts and Kennett 2000).



                                                                                   12
Without it, Australian households would on average have been $7000 poorer annually
by the end of the decade.


The productivity outlook

Notwithstanding the possibility of external economic shocks, there are grounds for
optimism about the general productivity outlook for Australia. The underlying rate of
productivity growth may not continue at its 1990s rate, when there was considerable
scope for us to catch-up to other countries after decades of lagging. But it is unlikely
to fall back to the rate of the late 1970s and 1980s.

For one thing, the heightened incentives and disciplines for improved performance are
not temporary. The reduction of barriers to competition and removal of impediments
to innovation can be expected to have lasting effects on the dynamism of our
economy. And, to the extent that the economy has become more flexible and
adaptable, its capacity to deal with any future external shocks and to continue to
benefit from technological advances will have improved.

In this respect, although a lot has been said about the potential productivity gains from
computer networks (especially e-commerce through the Internet and specialised
networks), their realisation remains largely ahead of us. The use of e-commerce only
became widespread in the United States at the end of the 1990s, with Australia a little
way behind. But, for reasons just outlined, Australia is now well placed to reap the
productivity gains that may flow from e-commence. We can also take heart from the
fact that US analysts are optimistic about the scope for ongoing productivity gains in
that economy flowing from ICT use (Greenspan 2002).

While Australia s aggregate productivity performance has outstripped that of most
OECD countries, some industry sectors — notably Manufacturing and Retail trade —
have not shown as much strength as in other countries. As demonstrated earlier,
productivity growth in these industries stagnated in the 1990s, whereas it accelerated
in the United States and other economies (BLS 2002, McKinsey Global Institute
2001).3 On the face of it, there should be scope to reinvigorate productivity growth in
these sectors. One possibility is that, in some areas, firms have been insulated from
the need or constrained in their ability to undertake the reorganisations that can
generate large productivity gains.

What does seem clear, is that government policy will continue to play an important
role in Australia s future productivity performance. Detailed industry research


3 The acceleration in productivity growth in US durables manufacturing is partly related to
  ICTs. The absence of ICT production in Australia would be one reason for Australia falling
  behind US productivity levels. However, Australia s Manufacturing productivity has also
  fallen behind that of many other OECD countries (van Ark and Timmer 2001).


                                                                                         13
confirms the strong links between microeconomic reform and the 1990s productivity
acceleration. Its two-sided influence — generating external competitive pressure and
the internal flexibility for firms to respond — has been particularly important to the
 new economy story in the services sector.

Labour reforms have been the key to this and they will remain of central importance.
But the closer that the Australian economy gets to its technological and productive
frontier, the more important will innovation be to our continuing progress. Innovation
embraces a lot more than technological progress. And while technology can always
be imported, its innovative use is largely a domestic issue. Ultimately how well we
do it depends on the intellectual qualities and attitudes of the managers and
workforces in Australian enterprises. That in turn largely depends on the
effectiveness of our education and training systems. Ensuring that those systems
work well — and there is growing evidence to the contrary — will be one of the key
challenges in sustaining Australia s productivity performance in the future.




                                                                                  14
Appendix A: Multifactor productivity by industry sector, 1974-75 to
2000-01



                      Agriculture                                                        Mining

200                                                               200


 160                                                              160


 120                                                              120

 80                                                                80

 40                                                                40

  0                                                                 0
  1974-75 1979-80 1984-85 1989-90 1994-95 1999-00                  1974-75 1979-80 1984-85 1989-90 1994-95 1999-00




                  Manufacturing                                              Electricity, gas & water

200                                                               200


 160                                                              160


 120                                                              120


 80                                                                80


 40                                                                40


  0                                                                0
  1974-75   1979-80    1984-85    1989-90    1994-95    1999-00    1974-75   1979-80   1984-85   1989-90   1994-95   1999-00




                      Construction                                                Wholesale trade

200                                                               200


160                                                               160


120                                                               120


 80                                                                80


 40                                                                40


  0                                                                0
 1974-75    1979-80   1984-85    1989-90    1994-95    1999-00     1974-75   1979-80   1984-85   1989-90   1994-95   1999-00




                                                                                                                          15
                     Retail trade                                                        Accommodation, cafes & restaurants

200                                                                             200


160                                                                             160


120                                                                             120


 80                                                                                80


 40                                                                                40


  0                                                                                      0
 1974-75   1979-80   1984-85   1989-90      1994-95             1999-00             1974-75             1979-80   1984-85   1989-90   1994-95   1999-00




              Transport & storage                                                                            Communications
200                                                                             200


160                                                                             160


120                                                                             120


 80                                                                                80


 40                                                                                40


 0                                                                                       0
 1974-75   1979-80   1984-85   1989-90     1994-95         1999-00                  1974-75             1979-80   1984-85   1989-90   1994-95   1999-00



             Finance & insurance                                                             Cultural & recreational services
200                                                                             200


160                                                                             160


120                                                                             120


 80                                                                                80


 40                                                                                40


 0                              200
                                160
                                120
                                 80
                                 400
                                                                                         0
                                 1974-75   1979-80    1984-85        1989-90   1994-95        1999-00
 1974-75   1979-80   1984-85   1989-90     1994-95         1999-00                  1974-75             1979-80   1984-85   1989-90   1994-95   1999-00




                                                      Industry                     Market sector




                                                                                                                                                     16
References

ABS (Australian Bureau of Statistics), Australian System of National Accounts, 2000-
  01, Cat.˚no.˚5204.0, ABS, Canberra.
BLS (Bureau of Labour Statistics) 2002, Productivity and Costs. Preliminary Fourth-
  Quarter      and     Annual       Averages       for    2 0 0 1 , February 6,
  (http://www.bls.gov/news.release/prod.nr.htm)
Brynjolfsson, E. and Hitt, L. 2000, Beyond Computation: Information Technology,
   Organizational Transformation and Business Performance , Journal of Economic
   Perspectives, vol. 14, no. 4, Fall, pp 23-48.
Carnegie, R. and Butlin, M. 1993, Managing the Innovating Enterprise: Australian
   Companies Competing with the World s Best, Business Council of Australia,
   Melbourne.
Gordon, R. J. 2000, Does the New Economy Measure up to the Great Inventions of
   the Past? , Journal of Economic Perspectives, vol. 14, no. 4, Fall, pp. 49-74.
Greenspan, A. 2002, Remarks by Chairman Alan Greenspan, At the Bay Area Council
   Conference, San Francisco, California, January 11.
Jackson Committee (Committee to Advise on Policies for Manufacturing Industry)
   1975, Policies for Development of Manufacturing Industry, vol. 1, Report to the
   Prime Minister, (Mr R.G. Jackson, Chairman), AGPS, Canberra.
Johnston, A., Porter, D., Cobbold, T. and Dollamore, R. 2000, Productivity in
   Australia s Wholesale and Retail trade, Productivity Commission Staff Research
   Paper, AusInfo, Canberra.
McGuikin, R.H and van Ark, B. 2001, Perspectives on a Global Economy. Making
  the Most of the Information Age: Productivity and Structural Reform in the New
  Economy. Research Report R-1301-RR, The Conference Board, New York.
McKinsey Global Institute 2001, US Productivity Growth 1995-2000. Understanding
  the contribution of Information Technology relative to other factors, McKinsey
  Global Institute, Washington, DC, October.
OECD (Organisation for Economic Co-operation and Development) 2001a, The New
  Economy: Beyond the Hype: Final Report on the OECD Growth Project, OECD,
  Paris.
OECD 2001b, OECD Science, Technology and Industry Scoreboard. Towards a
  knowledge-based economy. 2001 Edition. OECD, Paris.
Parham, D. 2001, Evidence of growth speaks for itself , Australian Financial
   Review, 13 December, page 50.


                                                                                 17
Parham, D., Barnes, P., Roberts, P. and Kennett, S. 2000, Distribution of the
   Economic Gains of the 1990s, Productivity Commission Staff Research Paper,
   AusInfo, Canberra.
Parham, D., Roberts, P. and Sun, H. 2001, Information Technology and Australia s
   Productivity Surge, Productivity Commission Staff Research Paper, AusInfo,
   Canberra.
PC (Productivity Commission) 1999, Microeconomic Reforms and Australian
  Productivity: Exploring the Links, Commission Research Paper, AusInfo,
  Canberra.
Quiggin, J. 2001a, The Australian Productivity Miracle: A Sceptical View , Agenda,
   vol. 8, no. 4, pp. 333-348.
Quiggin, J. 2001b, Productivity isn t working , Australian Financial Review, 6
   December, page 54.
Stiroh, K. 2001, Information Technology and the US Productivity Revival: What Do
    the Industry Data Say? , Federal Reserve Board of New York January 24 (mimeo).
Uhrig, J. 1979, Australian Manufacturers: Adjusting to the Realities of the Future,
   The Second Stan Kelly Memorial Lecture, 29 August, Research School of Pacific
   Studies, Australian National University, Canberra.
van Ark, B. and Timmer, M. 2001, The ICOP Manufacturing Database: International
   Comparisons of Productivity Levels, International Productivity Monitor, no. 3,
   Fall, pp. 49-74.
Vernon Committee (Committee of Economic Enquiry) 1965, Report of the Committee
   of Economic Enquiry, (Dr J. Vernon, Chairman), Commonwealth of Australia,
   Canberra.




                                                                                18

				
DOCUMENT INFO