Trends in Infrastructure Investment

1
Trends in Infrastructure Investment

• Eighth ACCC Conference
• Sanctuary Cove, 27 July 2007
• Jeff Balchin
• Director

2
Purpose

• Understand historical trends in infrastructure
  investment
• Assess adequacy of infrastructure investment from
  aggregate data
• Attribute cause for any inadequacy of
  infrastructure investment

3
The null hypothesis
4
Method / measurement

• Methodological issues
• Extent of aggregation / accuracy
• Measure quantities / volumes
• Public vs. private participation
• Measures

5
Data series

• Investment
• Engineering construction activity (ABS)
• More disaggregated (9 classes of infrastructure)
• September 1986 onwards
• Excludes buildings / land
• Commenced vs. Undertaken vs. Remaining
• Gross fixed capital formation (ABS, OECD)
• Less disaggregation (3 classes of infrastructure)
• 1960 onwards
• Includes buildings, excludes land
• Capital stock (ABS)
• Same breakdown as GFCF
• Gross asset value reproduction cost
• Assumed deterioration of stock of assets
• Capital services (ABS, OECD)
• Same breakdown as GFCF for Australia
• More aggregated for OECD data

6
Breakdown of data ECA vs. GFCF
Engineering construction activity
Gross fixed capital formation
7
1. Trends in investment
8
Investment Engineering Construction Activity (I)
Infrastructure construction undertaken(chain
volume measures, m, reference year 2004-05)
ABS, Engineering Construction Activity, Cat 8762
Table 6 Deflators purchased from National
Accounts Branch, Underlying indexes for
Engineering Construction, March Quarter 2007.

• Annual expenditure more than doubled since 1990
• Above average growth pipelines, railways,
  roads, harbours
• Average growth telcos, electricity
• Below average growth bridges, water and sewerage

9
Investment Engineering Construction Activity
(II)
Infrastructure construction by owner/operator (cha
in volume measures, m, reference year 2004-05)
Total Infrastructure construction (chain volume
measures, m, reference year 2004-05)
ABS Engineering Construction Activity, Cat 8762,
Tables 8, 10, 11 Deflators purchased from ABS
National Accounts Branch, Underlying indexes for
Engineering Construction, March Quarter 2007.
ABS Engineering Construction Activity, Cat 8762,
Table 6.

• Estimates of remaining infrastructure
  construction activity systematically exceeds
  actual remaining expenditure
• Large increase in infrastructure controlled by
  the private sector
• 13 at start of the period, 42 at the end of the
  period

10
Investment Gross Fixed Capital Formation
Gross fixed capital formation in infrastructure
industries(chain volume measures, m, reference
year 2004-05)
ABS National Accounts, Cat 5204, Table 71.

• More interesting period appears to be the 1980s
• Caused by state government financial problems?
• Accumulated gold plating?

11
Infrastructure capital stock / services
Capital services index (1964-65 100)
Net capital stock(chain volume measures, m,
reference year 2004-05)
ABS National Accounts, Cat 5204, Table 71.
ABS National Accounts, Cat 5204, Table 89.

• Changes in level of services dependent on asset
  lives

• Similar pattern, but dampened

12
2. Adequacy of investment
13
Measuring adequacy

• Upward trend in infrastructure investment is
  unexpected
• Economy has grown
• Need more infrastructure
• How much does the stock of infrastructure need to
  grow as the economy expands?
• In line with GDP growth?

14
Investment Engineering Construction Activity
Infrastructure construction completed(per cent
of GDP)
ABS, Engineering Construction Activity, Cat 8762
Table 6 Deflators purchased from National
Accounts Branch, Underlying indexes for
Engineering Construction, March Quarter 2007.

• Positive story growth in all sectors except
  water, sewerage and bridges at or above GDP growth

15
Investment Gross Fixed Capital Formation
Gross fixed capital formation in infrastructure
industries(per cent of GDP)
ABS National Accounts, Cat 5204, Table 71.

• Picture not so rosy when compared to the heydays
  of the 1960s

16
Infrastructure capital stock / services
Growth of capital services relative to GDP
(1964-65 100)
Net capital stock(per cent of GDP)
ABS National Accounts, Cat 5204, Table 71.
ABS National Accounts, Cat 5204, Table 89.
17
Investment international comparisons
GFCF Electricity, Gas Water supply(Per cent
GDP)
GFCF Post and Telecommunications(Per cent GDP)
GFCF Transport and Storage(Per cent GDP)
The OECD STAN Database for Industrial Analysis
18
Capital services
Growth of capital services relative to real
GDP (non-residential construction assets, 1985
100)
OECD, Productivity Database, 6 October 2006 OECD
Factbook 2006 Economic, Environmental and Social
Statistics
19
Growth of capital services and GDP Half empty or
half full?
Growth rate of GDP to capital services
Growth rate of capital services to GDP

• Two observationally equivalent hypotheses
• Australia is underspending relative to the better
  performers we should be concerned
• Australia has managed to achieve substantial GDP
  growth while only needing a modest increase in
  infrastructure we should be impressed

20
Capital services and GDP growth
Growth of capital services to GDP
Real GDP growth

• Relative GDP growth appears to explain most of
  the differences in growth of capital services
  relative to GDP
• Indeed, the losers (in terms of falling capital
  services to GDP) have been winners (in terms of
  GDP per capita)

21
Why may infrastructure needs decline as a
proportion of GDP?

• Economies of scale / scope / density
• Changing composition of the Australian economy
• Efficiency in investment decision making
• Forecasting tools
• Financial appraisal techniques

22
Changing demand for infrastructure
Industry Shares of GDP
Population Density
Department of Environment and Heritage, State of
the EnvironmentSecond Technical Paper Series.
Australian Bureau of Statistics, Year Book
Australia 2005.

• Reducing importance of manufacturing, increasing
  importance of services

• Population now much more dense

23
How to assess adequacy then?

• Focus on quantity / quality of outputs rather
  than quantity of inputs
• Shortcomings in data a priority
• Cannot answer all questions i.e. whether
  investment is optimal
• A micro-analysis is the only reliable method
• Social costs and benefits

24
Final remarks

• Private sector / capital is now extremely
  important
• Hence, getting the policy settings right is a
  priority
• Making robust inferences on issues relevant to
  infrastructure investment is difficult
• Trends may be easy, establishing adequacy and
  causes much harder
• Focus on outputs rather than inputs would be an
  improvement
• Need for improved data collection
• No substitute for proper analysis of economic
  costs and benefits