NatStats 2010 Conference
Bill Evans
Chief Economist
September 2010
Introduction
•
Data provides a key indicator for economic policy;
regulation and financial markets.
•
Consider two examples of how the historical unreliability of
data releases can affect policy.
•
Bank regulation policy associated with “excessive credit
growth”.
•
Monetary Policy associated with the measurement of the
Consumer Price Index.
Countercyclical Capital Buffer
•
Basle guidelines for a countercyclical capital buffer.
•
Required “top up” of capital for the banking system will be driven
by the deviation from trend ( HP filter) of the credit (broadest
measure) to nominal GDP ratio.
•
Banks will only be required to “top up” continuously once the
deviation is in the buffer zone. This top up will rely on stable
reliable measures of credit and nominal GDP.
•
The significance and volatility of Australia’s statistical discrepancy
in measuring GDP adds unwelcome uncertainty to the banks’
capital planning.
•
This study looks at the problems associated with the statistical
discrepancy with other issues associated with revisions to GDP
and credit aggregates additional challenges.
Aust Credit to GDP: deviation from trend
20
% dev from trend
Sources: ABS, RBA, Westpac.
15
10
Buffers activated
Credit to GDP gap (lhs)
5
0
-5
-10
-15
Dec-79 Dec-84 Dec-89 Dec-94 Dec-99 Dec-04 Dec-09
Countercyclical capital buffers for Aust
20
% dev from trend
ppt of RWA
4
Sources: ABS, RBA, Westpac.
15
3
10
2
5
1
0
0
-5
-10
Credit to GDP gap (lhs)
Trigger point for buffer (lhs)
Maximum buffer (rhs)
Level of buffer (rhs)
-15
Dec-79 Dec-84 Dec-89 Dec-94 Dec-99 Dec-04 Dec-09
-1
-2
-3
Australian GDP: three measures
index
index
135
135
GDP(E)
GDP(P)
125
125
GDP(I)
115
Since Sep ’08
(E) +5.2%,
(I) +3.3% & (P) +2.9%
105
115
105
Sources: ABS, Westpac Economics
95
Jun-00
95
Jun-02
Jun-04
Jun-06
Jun-08
Jun-10
Australian GDP: three measures
Sep ’09 National Accounts
% ann
% ann
Sources: ABS, Westpac Economics
Revised to:
(E) 2.1% from 1.1%
(I) 0.5% from 0.7%
(P) 0.0% from -0.1%
6
6
4
4
2
2
1.1%yr
0.7%yr
0
0
GDP(E)
-2
Sep-01
GDP(I)
GDP(P)
-0.1%yr
-2
Sep-03
Sep-05
Sep-07
Sep-09
Australian GDP: three measures
Dec ’09 National Accounts
% ann
% ann
Growth is: Too strong; About right; or Too weak
6
6
4
4.1%yr
4
2.6%yr
2
2
GDP(E)
GDP(I)
GDP(P)
1.4%yr
0
0
Sources: ABS, Westpac Economics
-2
Dec-01
Dec-03
-2
Dec-05
Dec-07
Dec-09
Countercyclical Capital Buffer
•
We estimate that once a banking system finds itself in the buffer
zone economic data will have a profound impact on their capital
planning.
•
Under the example of 2009 when the growth rates of the various
measures of GDP differed by 2 ppts the application of the
countercyclical buffer could have varied the amount of extra
capital that banks needed to acquire by up to $2.5 billion.
•
Since 2008 this discrepancy could have varied the level of capital
required to be held by the banking system by around $4 bn.
The Distortions from Measurement Error
•
In 2009 various measures (GDP (E); GDP (I); and GDP (P))
registered growth in the Australian economy at variously 4.1%;
2.6% and 1.4%
•
Which measure should be used for the countercyclical capital
buffer measurement.
•
We estimate that a 2 ppt shift in the GDP measure could result in
banks having to increase their capital base by a further $2.5 bn.
•
That would have serious implications for equity and capital
markets.
•
This does not take into account actual data revisions to GDP and
credit aggregates.
Statistical discrepancy: mind the gap
2
% GDP
% GDP
Sources: ABS, Westpac Economics
Australia
Japan
1
2
1
US
0
0
-1
-1
-2
Dec-80
-2
Dec-86
Mar-94
Mar-00
Mar-06
Monetary Policy and the CPI
•
The inflation report has been the most important driver of
monetary policy decisions since the Bank formally adopted an
inflation target.
•
The concept “deposit and loan facilities” seeks to measure the
cost of retail banking services by measuring the margins; fees
and taxes for each financial product offered by deposit taking
entities.
•
It is done by sampling various accounts – sample size 7000
individual accounts which contain 3 million transactions annually.
Issues such as nonlinearities of fees; waiving of fees; various
types of accounts which differ across institutions; substitution
between fees and margins; accuracy of account information all
make this procedure complex and potentially unstable.
Monetary Policy and the CPI
•
As an institution which needs to forecast the components of the CPI
we find this component totally impossible to get any lead whatsoever.
•
Not surprisingly we are unaware of any other country which includes
this concept in its CPI.
•
Such is the Reserve Bank’s scepticism about the component that the
Bank often refers to inflation measures ex deposit and loan facilities.
•
Policy decisions can “turn” on 0.1% of a quarterly CPI. In the lead up
to the GFC the RBA always tightened when the core CPI printed 0.8%
or more and only tightened on a 0.7% 50% of the time.
•
In the last 3 years deposit and loan facilities contributed at least 0.1
ppt to the CPI on 75% of the reads.
•
The measure is very important for the CPI. Of the 92 items in the CPI
this measure with a 4% weighting is exceeded only by house
purchase (8%); rents (5.9%); motor vehicles (4.1%).
RBA responds to inflationary pressures
1.6
%
Trimmed mean quarterly change
% 1.6
GFC
1.2
forecasts
1.2
0.8
0.8
0.4
0.4
0.0
Dec-03 Dec-04 Dec-05 Dec-06 Dec-07 Dec-08 Dec-09 Dec-10
0.0
Deposit & loan quarterly change
10
%
5
0
-5
-10
Sources: Westpac Economics, ABS
-15
Mar-99
Mar-01
Mar-03
Mar-05
Mar-07
Mar-09
Deposit & loan contribution to headline CPI
0.75
ppt cont’n
ppt cont’n
quarterly
0.75
0.50
0.50
0.25
0.25
0.00
0.00
-0.25
-0.25
-0.50
-0.50
Sources: Westpac Economics, ABS
-0.75
Mar-99
-0.75
Mar-01
Mar-03
Mar-05
Mar-07
Mar-09
Deposit & loan cont’n to ann headline CPI
1.0
ppt cont’n
ppt cont’n
annual
1.0
0.5
0.5
0.0
0.0
-0.5
-0.5
Sources: Westpac Economics, ABS
-1.0
Jun-02
-1.0
Jun-04
Jun-06
Jun-08
Jun-10
Some Conclusions
•
Data measurement can have serious implications for policy ;
regulation and financial markets.
•
Revisions; statistical discrepancies; measurement errors; have
and will be critical to fine tune.
•
Maintaining “elegant” approaches which cannot be measured in a
stable fashion may be counterproductive
•
Using inherently volatile data to dictate regulatory rules seems
dangerous
•
Measuring elegant concepts which are inherently random to drive
the key policy variable is risky and unhelpful to markets
•
Policy and regulation should rely on stable data series which are
not prone to instability and volatility
•
At the very least the “deposit and loans” component of the CPI
should be dropped.
Disclaimer
Westpac Institutional Bank is a division of Westpac Banking Corporation ABN 33 007 457 141.
Information current as at date above. This information has been prepared without taking account of your
objectives, financial situation or needs. Because of this you should, before acting on this information,
consider its appropriateness, having regard to your objectives, financial situation or needs. Westpac’s
financial services guide can be obtained by calling 132 032, visiting www.westpac.com.au or visiting any
Westpac Branch. The information may contain material provided directly by third parties, and while such
material is published with permission, Westpac accepts no responsibility for the accuracy or
completeness of any such material. Except where contrary to law, Westpac intends by this notice to
exclude liability for the information. The information is subject to change without notice and Westpac is
under no obligation to update the information or correct any inaccuracy which may become apparent at a
later date. Westpac Banking Corporation is registered in England as a branch (branch number
BR000106) and is authorised and regulated by The Financial Services Authority. Westpac Europe
Limited is a company registered in England (number 05660023) and is authorised and regulated by The
Financial Services Authority. If you wish to be removed from our e-mail, fax or mailing list please send an
e-mail to economics@westpac.com.au or fax us on +61 2 8254 6934 or write to Westpac Economics at
Level 2, 275 Kent Street, Sydney NSW 2000. Please state your full name, telephone/fax number and
company details on all correspondence. © 2010 Westpac Banking Corporation. Past performance is not a
reliable indicator of future performance. The forecasts given in this document are predictive in character.
Whilst every effort has been taken to ensure that the assumptions on which the forecasts are based are
reasonable, the forecasts may be affected by incorrect assumptions or by known or unknown risks and
uncertainties. The ultimate outcomes may differ substantially from these forecasts.