Transcript the underground economy and australia`s gdp

THE UNDERGROUND ECONOMY
AND AUSTRALIA'S GDP
Tony Johnson
ABS
Introduction
"Measuring the Non-Observed Economy: A handbook"
(2002) OECD, IMF,ILO, CIS
Key aim - to promote best practice to ensure estimates of
GDP are as exhaustive as possible within the production
boundary defined in SNA93
Some academic economists claim Australia's GDP
understated by 15% because of missed underground
production
ABS has recently reviewed its treatment of underground
transactions in the national accounts
The "non-observed" economy
Economic activity typically missing from data sources used
to compile the national accounts. Includes transactions that
are either:
Underground
Illegal
Informal
Household production for own use
Australian practice - Illegal activity is excluded from GDP.
The remaining components are included, but are often
based on scant and fragmented information. The third and
fourth components are relatively small for Australia
Underground activity
Definition -
"Those activities that are productive and legal but are
deliberately concealed from the public authorities to avoid
payment of taxes or complying with regulations" (Handbook
2002, p 13)
"Skimming" of cash receipts and overstating (or
understating) of expenses by businesses
"Moonlighting" by householders
Underground economy in Australia
Estimates are made using aggregate tax audit information
available for a single period only
Percent adjustments by industry have been held constant
over time
Supply and use balancing may add some further implicit
adjustment
GROSS DOMESTIC PRODUCT ACCOUNT, INCOME
ESTIMATES
2001
Explicit adjustment for
underground activity
$m
Adjustment to initial
income estimate
%
CoE
GOS
Non fin. corps.
Private
Public
Total
Financial corporations
General government
Dwellings
TOTAL GOS
GMI
Total factor income
Taxes less subsidies
on production and imports
GDP
0
0
3205
0
3205
0
0
0
3.3
0
2.8
0
0
0
3205
5277
8482
1.6
9.7
1.5
0
8482
1.3
-
Upper limit to missing GDP
Review data sources and make judgements about potential
for missed underground transactions
Assign hypothetical "upper limits" to missing transactions at
most detailed level possible
Individual judgments subject to potentially large error, but in
aggregate provides an indication of what could be plausible
for missing GDP
Assumptions used
Underground activities are conducted only in cash
General government and government businesses are not
involved
Large and medium businesses are unlikely to be involved
Imputed components of GDP are unaffected (by definition)
Small businesses and individuals are the major contributors
ƒ In Australia, at least 65% of GDP (as measured) is
unlikely to be missing underground transactions
UPPER BOUND ESTIMATES OF
UNDERGROUND ACTIVITY
2000-01
Upper bound estimates of
potentially missing
transactions
Upper bound estimates of
potentially missing
transactions
$m
CoE
GOS
Non fin. corps.
Private
Public
Total
Financial
corporations
General government
Dwelling owned by
persons
TOTAL GOS
GMI
Total factor income
Taxes less subsidies
on prodn. and imports
GDP
%
6382
2.0
5356
0
5356
0
5.5
0
4.6
0
0
0
0
0
5356
20256
31994
0
2.6
37.4
5.5
0
31994
4.8
Upper bound estimate (cont.)
Taking into consideration:
 adjustments already made
 implicit adjustments in supply-use balancing
 the generous upper bounds chosen
Australia's GDP highly unlikely to be understated by
more than 2% on account of missing underground
transactions, and most likely less
Additional plausibility checks
A high degree of accuracy can be attached to population
data from periodic censuses
Make assumptions about upper bounds for involvement and
average value of underground income according to labour
force status and occupation
Divide derived upper bounds by number of households. This
can be compared with after tax household income from
household surveys
ƒ Results were consistent with that previously described
Could GDP be understated by 15%?
estimates derived from currency demand models as high as
15% of GDP
Implications of 15% can be tested for plausibility
ƒ We concluded highly implausible when assessed at the
detailed level
Large businesses would have to be widely involved in
understating both income and wages and salaries paid
and this is highly unlikely in the Australian context
Discussion points
Are countries making explicit adjustments for underground
transactions?
If so, what order of magnitude?
What data sources and methods are used
Is it an important quality issue for national accountants?
Do countries have a view on the legitimacy of monetary
model estimates of the cash economy?