Download MEMORANDUM FOR FROM DATE

MEMORANDUM FOR
Minister of Finance
FROM
Alan Bollard
Governor
DATE
21 December 2011
SUBJECT
Briefing for Incoming Minister – Appendix
The business cycle and monetary policy, 199820111
FOR YOUR
Information
1. This note gives a brief outline of New Zealand‟s macro-economic and monetary
policy experience since 1998. It compares the experience to past cycles and to
experience abroad; breaks the cycle down into shorter periods; and narrates the
events that drove the cycle.
We also draw some conclusions about the
appropriateness of monetary policy when viewed from the advantage of long
hindsight (which, of course, was not an advantage that policy makers enjoyed at the
time). We make passing reference to fiscal and other macroeconomic policies which
were also relevant to management of the cycle.
1. The cycle compared with New Zealand history
2. New Zealand‟s economic expansion from 1998 to 2007 was long by the standards of
post-war experience, and resulted in high total growth. The expansion lasted for 38
quarters and quarterly production GDP was just over 37 percent higher at the end of
2007 than in mid-1998. Real per capita GDP rose by 23 per cent from 1998 to
2007.2
3. Other post-war expansions have averaged around 21 quarters and closed with GDP
around 26 percent higher.3 Only two other expansions lasted longer than 30
quarters: 1959-66 (31 quarters, with a 39 percent rise in GDP or roughly a 22 per
cent rise in GDP per capita) and 1968-76 (34 quarters, with a 32 percent rise in GDP
and around a 19 per cent rise in GDP per capita).
4. The contraction that started in 2008 lasted for five quarters, and quarterly GDP
shrank by just over three percent.4 This was slightly longer and shallower than the
average since World War II.
1
2
3
4
Principal author: Willy Chetwin, Adviser, Economics Department.
This takes the contraction as starting in the June quarter of 1998 (the first quarter of positive
quarterly growth after the 1997/98 recession) and finishing in the last quarter of 2007 (before the
economy entered recession in 2008).
Hall, V B and C J McDermott (2009) “The New Zealand business cycle” Econometric Theory, 25,
pp1050-1069
This takes the contraction as starting with the first quarter of negative growth in the recession,
2008Q1, and finishing with the last quarter of negative growth (2009Q1) before growth returned.
Ref #4603576
2
2. The New Zealand experience in international context
5. New Zealand‟s annual growth rate from 1998 to mid-2011 averaged 2.4 percent,
weaker than Australia‟s (3.3 percent) but slightly higher than the UK, USA and the
EMU. Note that there was less variation across these countries in growth per capita.
6. New Zealand and Australia boomed from 2001 to 2007, with total growth very similar
across the two. The other economies faced a period of slow growth from 2001 to
2003, and even in 2004 and 2005 grew slightly more slowly than New Zealand and
Australia. New Zealand stood out on the downside in 2006 with a brief dip in growth
to below 1 percent while momentum abroad slowed by less.
7. In 2008 and 2009, Australia was the only one of the group that avoided technical
recession. New Zealand‟s trough was less deep than those in the USA, UK and
EMU.
8. New Zealand‟s average headline inflation rate was 2.4 percent – in the middle of
the pack. The time profile of inflation was broadly similar to that in the other
countries, and most similar to that in Australia.
9. The rates in New Zealand and Australia stayed above 1 percent in 2009 while those
in the USA, UK and Europe all became negative. In late 2010 inflation in all of these
economies picked up strongly, with New Zealand and the UK the highest.
10. New Zealand‟s short term and long term wholesale interest rates have been the
highest of the group, with Australia‟s a close second. Partly because of smaller falls
in New Zealand‟s and Australia‟s short rates through 2001 and 2008/09, those two
countries‟ rates have the smallest standard deviations of the group.
3. Defining ‘phases’ of the 1998-2011 period
11. This discussion breaks the period into five phases. Those phases aim to reflect a
mix of the main drivers of cyclical conditions and the RBNZ‟s main monetary policy
concerns at the time. While other events have clearly been important, most have
been more transitory, such as droughts, or can be thought of as the endogenous
outcomes of the main themes identified here.
12. Table 1 provides summary statistics on the identified phases, and Appendix 1
contains a chart of GDP growth that identifies the phases. Section 4 narrates the
main economic and monetary policy events in each phase and provides tentative
observations about policy. Section 5 summarises those tentative views on monetary
policy through the period.
Ref #4603576
3
Table 1: Averages by phase – main macro-economic indicators
Annual
1
growth, pct
Annual
2
inflation, pct
Post-EAC recovery
1998Q1-2000Q2
Global headwinds
2000Q3-2003Q2
Domestic boom
2003Q3-2006Q3
Oil price spike
2006Q4-2008Q2
GFC and sov. debt
2008Q3-2011Q2
Overall expansion
1998Q2-2007Q4
2.9
Overall contraction
2008Q1-2009Q1
Notes:
1
2
3
4
5
TWI, level
1.4
55.5
Pct of
potential
5
GDP
0.1
3.6
2.7
5.9
53.0
1.1
3.3
2.7
6.6
66.8
0.5
3.1
2.4
6.7
61.6
-0.1
-0.1
3.0
3.7
63.9
-3.1
3.2
2.4
6.4
60.6
5.0
-0.7
3.8
7.2
63.6
-5.6
Average of the annual growth rate, real production GDP
Average of the annual inflation rate, CPI (excluding credit services prior to 1999Q3)
Monthly average 90-day rate, nominal
Monthly average TWI, nominal
Cumulative sum of percentage point output gaps over the relevant period, where the output gap is that from the
September 2011 Monetary Policy Statement.
CPI inflation measures
Nominal TWI
%
%
7
7
Target measure*
6
6
CPI non-tradables
CPI tradables
5
5
4
4
3
3
2
2
1
1
0
0
-1
-1
-2
-2
-3
-3
98 99 00 01 02 03 04 05 06 07 08 09 10 11
*CPI excluding section prices to 1999Q3, then CPI
New Zealand interest rates
%
10
90 day
%
10
8
10 year
4
6
4
2
2
0
0
-2
Index
10 year minus 90 day
98 99 00 01 02 03 04 05 06 07 08 09 10 11
75
75
70
70
65
65
60
60
55
55
50
50
45
98 99 00 01 02 03 04 05 06 07 08 09 10 11
45
-2
%
%
10
10
9
9
8
8
7
7
6
6
5
5
Aus
4
4
3
3
NZ*
2
2
EMU 1
1
UK
0
0
USA
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
*Minimum overnight floating rate prior to March 1999
Ref #4603576
Index
World policy rates
8
6
4
90-day
interest rate,
3
pct
6.2
4
4. Narrative: macro-economic and monetary policy developments
Recovery after the East Asia Crisis: 1998 to mid-2000
13. Through 1998 New Zealand and other affected economies came out of the post-East
Asia Crisis recession. In 1999 New Zealand entered a strong recovery phase
supported by world growth, a falling currency and domestic investment ahead of Y2K
and the America‟s Cup. Droughts and a brief dip in domestic confidence slowed
growth in the first part of 2000. In mid-2000 the dotcom bubble ended and global
demand was the main source of downside risk.
Real trade growth (SNA)
Ann %
20
15
10
5
0
-5
-10
-15
-20
-25
Prices for export commodities and oil
Ann %
Imports
Exports
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
20
15
10
5
0
-5
-10
-15
-20
-25
Index
300
ANZ world commodity prices
Dubai oil prices (RHS)
USD
140
100
250
80
200
60
40
150
100
120
20
98 99 00 01 02 03 04 05 06 07 08 09 10 11
0
14. CPI inflation fell after the East Asia Crisis. It rose from mid-1999 to reach 4 percent
by the end of 2000 as demand pressure returned, the currency fell and excise taxes
and oil prices pushed up near term inflation.
15. The RBNZ was – as were other economists – slow to recognise the size of the East
Asia Crisis world slowdown, and the way the MCI operating rule was being used
caused short rates to spike. When the RBNZ eased monetary conditions from mid1998, 90-day rates fell from 9 percent to just over 4 percent in early 1999. As the
recovery strengthened, tightening resumed – now using the OCR – in late 1999. The
OCR had been introduced at 4.5 percent in March 1999 and was raised to 6.5
percent between late 1999 and May 2000. It stayed at 6.5 percent, with a tightening
bias, through 2000.
Recovery after the East Asia Crisis: 1998 to mid-2000 – the monetary policy story
Immediate aftermath - evaluation
16. The RBNZ was slow to loosen policy in the East Asia Crisis, and so the way the MCI
operating rule worked caused markets to ramp rates up to high levels when the
exchange rate fell. Weak demand and warranted both a lower currency and lower
interest rates. The tightening likely exacerbated the economic effects of the East
Asia Crisis and droughts, and is probably part of the reason why New Zealand‟s
growth in 1998 was much weaker than in the other advanced economies discussed
in this note.
Ref #4603576
5
A rapid recovery caused inflation pressure to rise . . .
17. As activity recovered, the decision to start tightening through late 1999 appears to
have been appropriate: growth rose rapidly causing the previously-negative output
gap to become positive and reach around one percent of GDP in early 2000. The
question that follows is how soon the tightening should have ended or unwound as
the medium term outlook worsened.
. . . and pressure continued even as the outlook became uncertain . . .
18. The RBNZ did not foresee in the first half of 2000 how quickly growth and capacity
pressure would subsequently fall. The easing in demand came from several
sources: the end of the investment around Y2K and the America‟s Cup; higher
interest rates and lower domestic confidence; and from mid-2000 the global outlook
began to weaken when the dotcom bubble burst.
19. By mid-2000 the weaker demand outlook was more evident. However, the RBNZ‟s
concern was still focused on whether inflation expectations might rise because of the
price spike from the falling currency and rising petrol prices. Indeed, the CPI peaked
at 4 percent at the end of 2000 and CPI non-tradables inflation at 3.5 percent. The
inflation target at the time was 0 3 percent.
20. In mid-2000 markets indicated a belief that the RBNZ should have tightened by less
and/or should begin sooner to unwind the rate hikes. Market participants were
assuming less forthcoming inflation and demand pressure than was the RBNZ.
Evaluation
21. The Reserve Bank took the view that the direct effects on inflation of the fall in the
exchange rate were something to „look through‟. However, the Bank remained
cautious about the risk that the higher headline inflation could pose to general
medium-term inflationary pressure, partly because of the stimulatory effect to real
economic activity from the very low level of the exchange rate. The balance
between these risks and downside risks to activity meant that holding the OCR was
judged appropriate.
Global headwinds: mid-2000 to mid-2003
22. Global downside risks persisted from mid-2000 until the first half of 2003, though this
did not hurt New Zealand‟s growth as much as was feared ex ante. While growth
dropped to 1 percent in early 2001, this proved to be a temporary fall and activity
accelerated quickly thereafter.
23. The headwinds included the aftermath of the dotcom bubble, perceived deflation
risks in advanced economies, terrorism, SARS, falling prices for our export
commodities from 2001, and rising oil prices from 2002. Also detracting from New
Zealand growth was a rising TWI from mid-2002, reflecting New Zealand‟s relatively
strong growth outlook, and from mid-2003 a fall in net migration after the sharp and
unexpected rise in 2001.
Ref #4603576
6
Annual net migration (sa)
000s
Estimates of potential growth*
000s
Total
80
APC
80
3.6
60
60
3.4
40
40
3.2
20
20
0
0
Permanent and long term
-20
98 99 00 01 02 03 04 05 06 07 08 09 10 11
-20
APC
2001
2002
2003
2004
2011
3.6
3.4
3.2
3
3
2.8
2.8
2.6
2.6
2.4
2.4
98
99
00
01
02
03
04
05
*Estimates from each year come from the September quarter
Monetary Policy Statement
24. The early stages of a long boom in domestic activity had started in late 2001, with
low interest rates here and abroad, the sudden rise in net migration, and
strengthening in the labour market, consumer confidence and export incomes.
25. The RBNZ entered this period with a tightening bias, and shifted to a neutral bias at
the start of 2001 to account for international risks. The OCR fell cautiously in 2001
from its March level of 6.5 percent, with the RBNZ including caveats about the
upside risks from the surprising resilience of commodity prices and export demand.
26. In late 2001, the RBNZ made 100 bps of insurance cuts – 50 bps in September and
50 bps in November – after the terrorist attacks in the US. The „insurance‟ cuts were
unwound in 2002 as domestic pressure grew. The direction switched again in the
first half of 2003 when the OCR fell in response to global risks – demand, commodity
prices, falling net migration, the SARS scare, a double-season drought, and fears of
an electricity shortage – notwithstanding growing domestic demand.
Global headwinds: mid-2000 to mid-2003 – the monetary policy story
27. The challenge through this period was to predict how a series of external events –
whose international importance and persistence were hard to assess ex ante –
would affect New Zealand‟s inflation and activity outlook. .
Evaluation and lessons
28. Policy looks to have been broadly appropriate through 2000 and most of 2001.
While markets anticipated more loosening than RBNZ was projecting during the first
three quarters of the year, in hindsight the RBNZ‟s judgement seems right to not
have loosened by more.
29. From hindsight, rates should perhaps, on average, have been higher through the
last part of 2001 to mid-2003, leading into the domestic boom that was starting. In
particular, the insurance cuts in late 2001 and the rate cuts in the first half of 2003
could perhaps have been smaller, or could have been unwound more quickly.
Ref #4603576
7
30. The insurance cut in 2001 (November) was, in hindsight, unnecessary; and both
cuts could perhaps have been unwound more quickly. The cuts sought to pre-empt
the risk of further major weakening in the global economy, consistent with a forwardlooking approach (and with the objectives that were subsequently – from 2002 –
explicitly included in the PTA).
31. In part, a wait-and-see approach to moving rates might have led to smaller cuts, and
so smaller reversals subsequently. At the same time, the 2003 OCR reductions in
particular might reflect an overestimation of potential output growth, which highlights
the difficulty of estimating capacity pressure in real time.
32. At the time, the RBNZ noted relatively strong domestic pressure and projected rising
interest rates. The 100 bps of insurance cuts were fully unwound by July of 2002
when the downside scenarios failed to materialise and strong domestic pressure
continued.
33. At least some of the 75 bps rate reduction in early 2003 turned out to be
unnecessary, with the benefit of hindsight. Again, these cuts sought to pre-empt the
risk of global weakening and domestic contraction. The RBNZ saw this as
consistent with the intent of the new PTA, saying in the June 2003 Monetary Policy
Statement that
The reformulated Policy Targets Agreement requires us to target inflation with a
medium term focus and to avoid unnecessary instability in output, the exchange
rate and interest rates. This OCR decision is intended to help prevent an
unnecessarily sharp downturn, while delivering inflation that remains comfortably
within the target range over the next few years. Influencing our decision is
increased certainty that the economy has started to turn down, and the presence
of downside risks – foreign and local – whose effects may be exaggerated by
fragile confidence. In addition, CPI inflation is expected to receive extra
downward pressure over the year ahead as a direct result of the higher exchange
rate, but this effect will be temporary.5
34. The new PTA agreed in September 2002 changed the specification of the inflation
target. The target range had previously been 0 to 3 percent CPI inflation, whereas
now the PTA said that:
For the purpose of this agreement, the policy target shall be to keep future CPI
inflation outcomes between 1 percent and 3 percent on average over the medium
term.
5
The new PTA agreed in September 2002 set out the target in Clause 2 (b):
For the purpose of this agreement, the policy target shall be to keep future CPI inflation outcomes
between 1 percent and 3 percent on average over the medium term.
Clause 4(b) stated that in pursuing the target, monetary policy should seek to avoid unnecessary instability
in output, interest rates and the exchange rate.
Ref #4603576
8
35. In short, the 2003 cuts look to have taken advantage of a transitory – exchange-rate
induced – dip in headline inflation to insure against broader downside risks.
Domestic pressure and non-tradables inflation were strong at the time, meaning that
inflation was projected to climb from near the middle of the target range from the end
of 2003.
36. With both the 2001 insurance cuts and the rate reductions of 2003, the goal of being
pre-emptive was in line with the 2002 PTA. From hindsight, the objective might
better have been served by watching developments for longer before moving: there
were risks in both directions and in each case inflation was relatively high. A waitand-see approach may have averted rate cuts that needed to be unwound quite
shortly afterwards.
37. The OCR reductions in late 2001 and particularly 2003 also reflected the challenge
of estimating capacity pressure in real time. From 2001 the RBNZ had been raising
its estimate of how quickly the economy could grow without inflation becoming a
problem. The chart above shows that the 2011 ex post estimate of potential growth
rose only slightly from 2001 to 2003, while our contemporary estimates and
forecasts rose over the period. In effect, the contemporary projections treated a
larger part of realised growth as coming from increased capacity rather than excess
demand. That partly explains why the RBNZ saw a lower OCR as consistent with
the inflation target under these cyclical conditions.6
The domestic boom: 2003 to early-2007
38. From 2003, a domestic boom and growing imbalances dominated the macroeconomic and monetary policy stories. Cheap, fixed-rate credit was supported by
global liquidity and low offshore rates. As in many countries, property markets –
here residential, agricultural and to some degree commercial – ran hot for several
years. Borrowing led to weakening household and institutional balance sheets as
asset markets became overextended. A rising currency and cheap imports made for
strong consumption and investment. While our export commodity price index rose to
what was at the time a record (late 2007/early 2008), non-commodity exporters and
import competitors were squeezed.
6
The estimates of rising potential output growth likely developed when domestic inflation had surprised on
the downside following strong growth in 2001 and 2002. In hindsight, the RBNZ was at the time
overestimating the earlier strain on capacity, meaning that it was expecting inflation to be stronger. When
inflation turned out weaker in 2002 and 2003 than expected, the RBNZ attributed part of this potential
growth having risen.
Ref #4603576
9
Annual house price inflation
%
30
25
20
15
10
5
0
-5
-10
-15
-20
NZ
Aus
US S&P Case Shiller
US FHFA
UK
Total household credit
%
30
25
20
15
10
5
0
-5
-10
-15
-20
98 99 00 01 02 03 04 05 06 07 08 09 10 11
Private investment ex-dwellings,
% expenditure GDP (sa)
%
17
16
15
14
13
12
11
10
9
APC
% GDPE
18
1150
16
1100
14
1050
1000
12 Annual growth
950
10
900
8
850
6
800
750
4
% expenditure GDP (RHS)
700
2
650
0
98 99 00 01 02 03 04 05 06 07 08 09 10 11
Private consumption,
% expenditure GDP (sa)
%
Real
Nominal
98 99 00 01 02 03 04 05 06 07 08 09 10 11
17
16
15
14
13
12
11
10
9
%
%
62
62
Real
61
61
60
60
59
59
58
Nominal
58
98 99 00 01 02 03 04 05 06 07 08 09 10 11
39. Monetary policy faced a challenge in responding to domestic pressure and
imbalances: currency strength was exacerbating sectoral imbalances, and was
related to the growing policy rate differential. The growing policy rate differential also
led to an increased supply of longer-term hedged funding from abroad to New
Zealand. Supplementary instruments were examined to help the OCR, but none was
seen as being both ready to use and having desirable effects.
40. The RBNZ raised the OCR from 5 percent through the second half of 2003 to 8.25
percent by the end of the period, in two broad phases. Rates reached 7.25 percent
at the end of 2005. The RBNZ paused in 2006 as it watched tentative signs of
economic rebalancing and slowing that did not quite materialise. In 2007, the OCR
recommenced its climb, rising to 8.25 percent.
The domestic boom: 2003 to early-2007 – the monetary policy story
41. Overall, the rise in the policy rate to the end of 2007 was much greater than the
RBNZ or markets had anticipated at the start of the tightening cycle. The OCR
peaked at 8.25 percent.
42. Nonetheless, the main question about monetary policy in this period is – in line with
the inflation target and Clause 4b of the PTA – whether it could have better kept
inflation in check and limited the imbalances that developed. We conclude that
Ref #4603576
10
interest rates might have been higher in an ideal world; but there were some
practical barriers to the effectiveness of a higher OCR, particularly from late 2004
once domestic momentum was strong.
43. In the face of strong domestic inflation during the boom, two main factors appear to
have prevented the OCR from rising faster and / or further. The first was a series of
indications, which turned out to be false, that activity and pressure on capacity would
soon ease. The second was growing imbalances in the economy, and a view that
further increases in the OCR might worsen the sectoral imbalances while having
only a small effect on aggregate pressure.
The nature of the imbalances
44. The imbalances that grew were part of the story of surprisingly resilient domestic
activity, which in turn kept domestic inflation around 4 percent for much of the
period. Inflation was also boosted by a long period of increasing oil prices.
45. Household and farm balance sheets became highly leveraged and property prices
overvalued. New Zealanders‟ propensity to borrow from overseas meant that
import-competing and non-commodity exporting industries were squeezed by a high
currency while domestically-focused industries struggled to keep up with demand.
The nature of uncertainty about the outlook
46. Uncertainty about the outlook was a recurring theme in the period – when and how
quickly activity and inflationary pressure would ease. Both the RBNZ and markets
saw a number of indications that this might happen.7 On the domestic side, the
downside risks related to the increasing amount of monetary tightening „in the
pipeline‟, stretched household and farm balance sheets, overvalued property prices,
a tight labour market and rising wages. On the external side, volatility in the terms of
trade, the currency and the outlook for foreign demand made forecasting hard.
The timing of policy moves: taking account of uncertainty and imbalances
47. The upshot of the uncertainty and of growing imbalances was a tentative approach
to moving the OCR when there was a risk that momentum might be shifting.
48. In 2006, when the economy appeared to be slowing, the RBNZ appropriately held
rates steady rather than easing. In doing so, the RBNZ noted the very high level of
built-up capacity pressure, domestic inflation and, therefore, the strong upside risks
if policy were eased too early and activity subsequently picked up. As it turned out,
momentum did return in 2007 and the RBNZ, again appropriately, further tightened
policy.
Evaluating the policy response
49. In an ideal world, monetary policy might have done more to take the heat out of
inflation and lean against the growing economic imbalances. Non-tradables inflation
7
Indeed, the RBNZ’s 2007-2010 Statement of Intent observed that forecasters and markets had consistently
been surprised by economic outcomes. The Reserve Bank was continuing to work on whether there was
scope for upgrading its projection and analytical tools.
Ref #4603576
11
was around or above 4 percent for much of the period, and this stemmed from
genuine pressure on resources.
50. With the level of inflationary pressure in the economy, it was appropriate to be very
cautious – as the RBNZ was in 2006 – when thinking about whether a falling growth
outlook should mean falling rates. On the other side, one could argue that strong
domestic inflation meant the RBNZ might have raised rates earlier from the second
half of 2003 to 2005.
51. Could monetary policy have done more to respond to the run-up in domestic inflation
and capacity strain (and so hopefully limit any subsequent bust, consistent with
Clause 4b of the PTA)?
52. A higher OCR would have been desirable through 2003 and leading into 2004, and
may well have had more effect on aggregate demand and inflation than after the
boom was underway.
53. Once the domestic boom was underway, there was scope for a higher OCR to have
helped at the margin. However, higher rates in this period would also have pushed
up the currency and worsened some of the growing sectoral imbalances.
54. Other instruments were considered through 2005 to 2007, but none was readily
available to help monetary policy do its job while taking pressure off the currency.
Exchange rate intervention in 2007 had a temporary effect on the strong currency,
but this appears not to have been large. The RBNZ has since looked further at how
it uses macro-financial analysis in building a view of the economy, as well as at
macro-prudential tools that might assist in the future.
55. Among the things making the OCR‟s job harder, fiscal stimulus added to pressure to
demand and the exchange rate from 2005. The RBNZ expressed on a number of
occasions concern about the timing of the fiscal stimulus, coming as it did at a time
when the output gap was significantly positive.
56. House prices were rising quickly, making the real interest rate on housing loans look
low. Moreover, a large proportion of household and farm borrowers were on fixed
rates which were held down by low offshore rates and only partially affected by the
OCR.
Oil price spike: 2007 to mid-2008
57. From mid-2007, with domestic inflation already strong, oil prices rose sharply on the
back of a steady climb since 2003. They peaked in mid-2008 and dropped back over
the second half of that year. Food prices moved similarly.
58. In the background were early, but growing, signs of financial disruption in the US and
Europe. Those forces together helped trigger in 2008 the end of a 10-year
expansion during which GDP rose by around 37 percent. Domestically, property
Ref #4603576
12
prices fell from mid-2007 and the cumulative rise in policy rates started to bite.
Supporting activity were fiscal policy and export commodity prices.
59. The RBNZ held the OCR steady at 8.25 percent from mid-2007: there was a strong
build-up of inflationary pressure, short-term effects of food and oil prices, and an
increasing risk of weakening activity.
Oil price spike: 2007 to mid-2008 – the monetary policy story
60. Coming late in a long period of strong activity, a cost-side shock like the oil price
spike of 2007 and 2008 threatened to cause one or both of higher inflation and
weaker activity. The policy response had to balance these two effects. As it turned
out, the spike, reinforced by the weakening world outlook, helped tip New Zealand
into the recession of 2008-09. Monetary policy watched and waited for certainty
about the direction of inflationary pressure, and by mid-2008 was expressing an
easing bias.
Context
61. The cost shock came on top of rising effective interest rates, stretched household
and farm balance sheets, the beginning of falling property prices and growing
concern about the global outlook. It also came at a time when domestic inflation
was very high and there was a risk that further cost pressure would pass into
persistent inflation.
62. From mid-2007, there were early signs of pressure. The gap between the OCR and
90-day bill rates widened materially. The RBNZ also began broadening liquidity
facilities from August 2007. In mid-2008, sharply-higher funding costs for banks and
a fall in export demand became important. The spike also came at the same time as
a sharp rise in food prices, which affected both New Zealand‟s import and export
prices.
Evaluating the policy response
63. Though still early to judge, monetary policy during this period looks to have been
broadly right. With built-up inflation pressure very strong, the RBNZ needed to worry
about whether a transitory price spike – as this appeared to be – would pass into
expectations and persistent inflation.
64. In a situation with lower domestic inflation and momentum, monetary policy would
have had far more room to accommodate the transitory price spike and focus more
on the medium-term risks to activity. However, in the situation to hand in 2007 and
2008, there might have been a high cost had the RBNZ eased early and
subsequently found that prices reversed before affecting activity.
65. With household balance sheets already stretched and interest rates rising, the
higher oil and food prices persisted for long enough to help slow consumption and
housing demand. Indeed, even after dropping, oil prices remained high by historical
Ref #4603576
13
standards. By that time the GFC was directly affecting New Zealand and the weight
of risks was to the downside, notwithstanding high headline inflation.
66. By mid-2008, it had become clear that activity in New Zealand was falling and
activity abroad was probably doing likewise. In the June 2008 Monetary Policy
Statement the RBNZ expressed an easing bias.
GFC and follow-on: mid-2008 to present
67. The seriousness of the US and European financial disruption became clear in the
first half of 2008. This would affect both world demand and New Zealand‟s funding
costs.
68. Domestically, the housing market was starting to slow and a drought had hurt
agricultural supply. Wage pressure and inflation expectations remained strong and
fiscal stimulus continued. Oil and food prices fell sharply from mid-2008, though
rebounded from early-to-mid-2009. Signs of domestic and global recovery began in
2009, but were disrupted by renewed stress abroad when sovereign debt concerns
caused renewed weakness. Earthquakes in late 2010 and early 2011 knocked
domestic confidence.
HLFS employment
%
5
Real NZD exchange rates
000s
Number (RHS)
4
3
APC
2
0.9
AUD/NZD
70
65
0.8
60
1900
0
-1
QPC (sa)
-2
98 99 00 01 02 03 04 05 06 07 08 09 10 11
TWI
1800
1700
80
75
2100
2000
1
-3
2200
0.7
98 99 00 01 02 03 04 05 06 07 08 09 10 11
55
50
45
69. The RBNZ lowered the OCR in July 2008 from 8.25 to 8 percent, and the cuts
accelerated as the outlook worsened and oil and commodity prices fell. The OCR hit
a low of 2.5 percent in April 2009 and remained there until the middle of 2010. There
was a brief sojourn up to 3 percent in mid-2010 as the outlook improved, before
sovereign debt concerns and the Christchurch earthquakes led rates to drop back to
2.5 percent in March 2011.
GFC and follow-on: mid-2008 to present – the monetary policy story
70. The RBNZ rapidly lowered the OCR from 8.25 percent in June 2008 to 2.5 percent in
April 2009. A large and fast drop like this was the appropriate response to a sharply
worsening growth outlook and falling pressure on persistent inflation. Rates remain
at historically low levels more than two years later.
Ref #4603576
14
International policy movements
71. Similarly sharp cuts in policy rates occurred abroad. The US felt the real economic
effects of the GFC and oil prices earlier than others and so eased earlier. Others
initially moved, like the RBNZ, more gradually. The Bank of England, European
Central Bank, Reserve Bank of Australia (RBA) and RBNZ all accelerated their rate
cuts after the Lehman Brothers bankruptcy and market disruption of September
2008. With the exception of the RBA, all of those other central banks have since
held rates at or very close to the trough reached in 2009.
Evaluation
72. Easier monetary policy was allowed by a rapid reduction in resource pressures, and
to some degree needed because other sources of support for activity were scarce.
With banks increasingly cautious about lending and facing higher funding costs,
expanded liquidity facilities were also important for helping easier monetary policy to
pass through to the economy.
73. While the increase in the OCR in mid-2010 lasted only briefly, it looked to have been
an appropriate response to the increase in capacity pressures that occurred from
late 2009 through early 2010. The factors that subsequently brought rates back
down – the Canterbury earthquakes and the burgeoning market reaction to
sovereign debt problems in advanced economies – were not foreseen when the
OCR was first raised.
74. Bringing the OCR back down in early 2011 was similarly an appropriate response to
unforeseen events, namely the major Christchurch quake in February 2011 and the
downgrade to the world outlook.
5. Summary of ex post policy observations
75. Setting monetary policy at the time and with limited information is always complex,
and the effectiveness of policy can be limited by many complicating factors. But
against that background, and in an attempt to draw out some lessons with the benefit
of hindsight, we offer the following tentative observations.
76. The way the MCI was used exacerbated the negative economic effects, compared
with the way the OCR has subsequently been used, when the RBNZ was too slow to
loosen after the EAC.
77. Policy in 1999 looks to have been broadly consistent with the PTA objectives. One
could argue that it was too tight through the first half of 2000, although subsequent
events such as 9/11 and SARS would in any case have dominated the effect of a
marginal policy change.
Ref #4603576
15
78. Policy from 2001 to 2003 was looser on average than it might have been had we
been able to foresee the surprising continuation of the housing boom post-2003.
Along with the expectation that the housing market would ease, the RBNZ was
overestimating potential growth. The RBNZ therefore saw relatively little upside risk
to inflation and focused more on minimising the risk of weak output growth (i.e. part
of Clause 4b). Contributing to this focus on the risk of weakening activity was the
change to the Policy Targets Agreement in 2002.
79. It was appropriate, given the inflation target, to be tightening policy through 2004 and
2005. Looking back, tighter policy might have been preferable, in particular by
entering the period with rates at a higher level (see previous point on rates through
2003).
80. Holding rates steady through 2006 was consistent with the PTA‟s objectives, given
the apparent weakening of activity playing off against the strong accumulated
capacity stress. However, with hindsight, it may have prolonged inflationary
pressures.
81. Policy in 2007 and subsequently looks broadly right. The RBNZ was appropriately
cautious about moving to loosen until it was confident that the strong inflation
pressure really was easing.
82. From 2004 to the present we saw a number of factors restricting the effectiveness of
monetary policy:
a. During the housing boom, real interest rates appear to have been very low
when based on expectations of house prices which were driving investor
credit demand.
b. Banks were able to hedge fixed-term interest rate risk abroad at rates which
were cheap relative to New Zealand‟s policy rate. This resulted from very low
international policy rates and a strong appetite by offshore investors for NZ
dollar risk. Domestic policy was therefore slower to pass through to effective
interest rates and the economy.
c. The exchange rate was high and rising, reflecting strong New Zealand
economic conditions and domestic policy relative to the easy international
financial conditions. There was concern that further rises in the OCR risked
adding to upward pressure on the currency, which could have added to the
growing sectoral imbalance.
d. Fiscal policy was adding to upward pressure on interest rates and the
exchange rate, by reducing national savings and by adding to excess
demand.
83. For these reasons, the RBNZ has continued to underline the importance of other
policies giving support to monetary policy, particularly at the extremes of the cycle.
The Bank‟s current examination of potential macro-prudential policy tools is relevant
Ref #4603576
16
in this regard. The experience since 2004 also clearly demonstrates the ongoing
influence of international financial conditions on New Zealand, even under a fully
floating exchange rate regime.
Ref #4603576
17
Appendix 1
Annual GDP growth
%
Terrorist attacks
in USA
US policy rate
drops to 1%
•
•
US policy rate rises
to 5.25%
US policy rate drops
to 0.25%
•
NASDAQ falls
•
•
6
Lehman Brothers
bankruptcy
•
NZD commodity
prices peak
4
NZD oil and
commodity prices
pre-GFC peak
•
•
2
0
-2
-4
Ref #4603576
House prices
accelerate
•
•
4
•
Net migration
peaks
Budget signals fiscal
stimulus
Housing market
turns
•
•
0
•
EAC recovery
(domestic and world)
6
2
Drought
Drought
breaks
%
Global headwinds: dotcom bust, growth
and deflation risks, terrorism, SARs
Oil spike
GFC, sovereign debt
Domestic boom: property, consumption,
investment
98 99 00 01 02 03 04 05 06 07 08 09 10
-2
-4
18
Appendix 2: macro experience in selected advanced economies
Real GDP (sa), index = 1 in 1998Q1
CPI level, index = 1 in 1998Q1
*
1.5
Aus
1.5
Index
1.5
1.4
NZ
1.4
1.4
USA
1.4
1.3
1.3
EMU
1.3
1.2
1.2
1.1
1.1
Index
Index
USA
UK
EMU
1.3
1.2
1.1
1
1
Aus
UK
Index
1.5
1.2
1.1
NZ
1
1
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
*For New Zealand, excluding credit services prior to 1999Q3
Annual CPI inflation
Annual GDP growth rates
%
%
6
6
4
4
2
2
0
0
-2
NZ
Aus
UK
USA
EMU
-4
-6
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
Ref #4603576
-2
-4
-6
%
6
5
4
3
2
1
0
-1
-2
%
NZ
Aus
UK
USA
EMU
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
*For New Zealand, excluding credit services prior to 1999Q3
6
5
4
3
2
1
0
-1
-2
19
Nominal TWI, index = 1 in Jan 1998
Nominal 90 day interest rates
Index
Index
EMU
1.3
1.2
1.2
Aus
1.1
1.3
NZ
1.1
1
1
USA
0.9
0.8
0.9
0.8
UK
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
%
7
7
6
6
Aus
5
EMU
4
USA
UK
NZ
3
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
Ref #4603576
%
10
9
8
7
6
Aus 5
4
3
NZ
2
EMU
1
UK
USA 0
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
Nominal 10 year interest rates
%
%
10
9
8
7
6
5
4
3
2
1
0
5
4
3