Treasury staff insight

FEU Special Topic: The current economic slowdown

The Treasury
Publication type:
Publication category:

Staff and teams are writing in their individual capacity and the views are not necessarily a "Treasury" view. Please read our disclaimer.

FEU Special Topic: The current economic slowdown


  • Economic activity in New Zealand and a number of other countries is slowing with weaker growth expected to persist.
  • The Treasury is not forecasting a recession in any meaningful sense of the word but does forecast slower growth reflecting the need for monetary policy to respond to excessive inflation.
  • Downside risks have increased and it would not take much for low rates of growth to turn slightly negative but even then, the slowdown is likely to be differentiated from prior recessions:
    • the overall dip in GDP likely to be smaller
    • unemployment is expected to peak lower
    • interest rates are focused on constraining demand rather than shoring up demand following a shock
    • net migration is expected to add to the population
    • the terms of trade remain elevated.
  • New Zealand recessions tend to be accompanied by shocks to the global economy. Global risks have intensified, with policymakers increasingly focused on reducing inflation in a backdrop of continued COVID‑19 and geopolitical volatility.
  • The characteristics of recessions and their impacts across the economy and society vary over time.
  • Policy is not set in stone but may need to adapt to changing conditions.

Slower growth ahead…

The New Zealand economy, along with many others, appears to be entering a period of slower growth. GDP data is likely to be volatile from quarter to quarter - for example, GDP fell modestly in the March quarter, but we expect a rebound in the June quarter. However, over time we do expect the trend to slow. Treasury's most recent forecasts anticipated an average quarterly growth rate of 0.1% over 2023, compared to the long-term average of 0.6%.

… as economy faces multiple headwinds

The economy is facing multiple headwinds as well as a couple of stronger areas. However, the overarching reason to expect a slowdown is that in recent times the economy has “overheated” - demand to consume goods and services has exceeded the economy's capacity to supply, contributing to inflation. Left unchecked, high inflation could result in adverse social and economic consequences.

Bringing inflation down will require a better match between supply and demand in the economy. Economy-wide supply is generally either very slow moving (eg, productivity growth) or is dictated by global factors (eg, global shipping availability). Therefore, in practice bringing inflation under control will mainly mean demand cooling to a more sustainable level.

In an effort to cool demand in the economy, the Reserve Bank has so far increased the OCR from 0.25% to 2%. How much further interest rates need to rise in order to cool demand sufficiently is uncertain, and depends on broader economic conditions, such as:

  • the impact of COVID-19 on economic activity, including the effect of opening the border
  • the ongoing impact of the Russia's invasion of Ukraine
  • the terms of trade (New Zealand export prices relative to import prices)
  • whether global supply chain disruptions start to ease
  • the speed of wage growth
  • asset prices, such as houses and equities.

However, whether the slowdown is principally due to external factors such as the terms or trade, or due to higher interest rates, either way the existence of high inflation gives us strong reason to anticipate a slowdown of some kind.

Businesses and consumers are well aware of the above discussion, and a slowdown appears to be widely expected. This has included a surge in ‘recession' focussed internet searches (Figure 1).

Figure 1: Intensity of ‘recession' internet searches

Figure 1: Intensity of recession internet searches

Source: Google Trends

The Treasury is not forecasting a recession

In our Budget Update we do not forecast a recession - but do forecast a material slowing in activity. This is also true of other forecasters. It is worth comparing our forecast outlook to experiences in previous recessions to highlight key differences as well as identify any similarities.

Before looking at the characteristics of past recessions it is worth noting that there is no official definition of what constitutes a recession. A common definition - that a recession is when there are two or more successive declines in real GDP - aligns with other definitions a reasonable amount of the time, but not always, and is far from perfect (and therefore is sometimes called a ‘technical recession'). A box later in this Special Topic examines the definition of recession in more detail as well as summarising the results of prominent research that dates previous New Zealand recessions.

Prior to COVID-19 New Zealand had experienced 9 periods of recession in the post WWII period

Periods of recession are summarised below from most recent to earliest.[1]

The Global Financial Crisis (GFC - 2007 to 2009) recession was a relatively lengthy downturn. Over much of the early and mid-2000s, inflation had remained above target. The Official Cash Rate (OCR) was raised from 5% at the end of 2003 to 8.25% in 2007. House prices had been rising rapidly but stalled in mid-2007 before declining over 2008. The 2007/2008 drought added to weakness in the domestic economy before the recession was extended by major shocks to the world economy and finance company failures. The terms of trade had risen over much of the early to mid-2000s, reaching three-decade highs at the start of 2008. Sharp falls in the terms of trade as the global economy shrank drove a large exchange rate depreciation and was accompanied with a substantial easing of monetary policy.

The Asian Financial Crisis & drought (1997 to 1998) saw a fall in export demand from Asian economies triggered by a regional financial crisis. In addition, a sustained drought hit the agricultural sector. Interest rates rose slightly, an atypical monetary policy response, largely attributed to the Reserve Bank's new ‘Monetary Conditions Index' approach to policy.[2]

During the early 90s recession (1990 to 1991) a slowdown in the world economy was exacerbated by high public debt in New Zealand and the threat of credit downgrades. In response there was a sharp fiscal contraction. Monetary policy was relatively tight following periods of high inflation for much of the 1980s. Under the Reserve Bank Act (1989) monetary policy was to focus on achieving inflation in the 0% to 2% range by the end of 1993. Falls in house prices were accompanied by unemployment rising above 10%.

The late 80s recession (1987 to 1988) involved a global share market crash hitting New Zealand following a long period of erratic growth and high inflation; the value of the New Zealand share market halved, and an economy already part way through restructuring was weakened by the failure of major banks and corporations.

The second oil price shock (1982 to 1983) followed the initial oil price shock of 1979 and accompanying global recession. It placed a substantial burden on the already strained current account deficit and forced devaluation of the (then fixed) New Zealand dollar. Attempts to support demand by ‘Think Big' projects were hampered by the relative import intensity of the measures.

The first oil price shock (1976 to 1978) involved a rise in oil prices and drastic reduction in the terms of trade from 1973 to 1976, triggering a domestic recession. This was intensified by substantial migration outflows and attempts to bring inflation under control from 1976.

During the Wool Bust (1966 to 1967) a collapse in wool prices resulted in New Zealand losing an eighth of its export income overnight - a product of the undiversified economy of the time. Maintaining a fixed exchange rate required controls on imports and increases in the (then directly controlled) prices of electricity and other utilities.

The COVID-19 period saw unprecedented declines as activity was restricted for health reasons

After a small fall in real GDP in the March 2020 quarter, the first lockdown saw a 10% quarterly decline in the June 2020 quarter - dwarfing previous quarterly declines. However, as restrictions were removed GDP bounced back in September 2020 to exceed the pre-COVID peak. The lockdown periods were unique in that parts of economic activity were deliberately curtailed to prevent physical interactions and increased spread of COVID-19. No doubt there will be ongoing debate about how to classify this period in the future but for this discussion it has been treated as an additional recession period.

Previous recessions provide a number of insights

Despite the varied nature of past downturns, there are a number of insights and trends:

  • International shocks accompany each downturn (at times worsened by domestic imbalances).
  • Often these international shocks coincide with tighter domestic monetary policy in response to periods of sustained inflation and excess demand.
  • Domestic policy responses can worsen recessions if monetary policy does not respond sufficiently quickly, or fiscal policy does not respond or is poorly targeted.
  • The terms of trade have become less central in driving shocks over time. Downturns of the 1960s and 1970s were driven primarily by large falls in the terms of trade; downturns of the 1980s and 1990s were accompanied by the terms of trade holding steady, while fluctuations in the terms of trade post- GFC were not associated with downturns (these fluctuations occurred around an upward trend).
  • Net migration has typically (but not always) fallen during a downturn. The behaviour of net migration in an economic downturn is likely to reflect the performance of the New Zealand economy relative to other developed economies, particularly Australia.
  • Private investment has typically contracted sharply and is only slightly offset by increases in government investment.
  • Exchange rate depreciation supported stabilisation in the GFC, but adjustments had been more modest in prior downturns.

How does the current economic outlook compare to periods of recession in New Zealand?

Growth to slow rather than sizable GDP falls

A feature of the Budget Update economic forecasts is that economic growth is predicted to slow considerably later this year. In an underlying sense, this slowing has likely already begun but we anticipate that factors such as a resumption of international tourism and a fading of the negative impacts from Omicron will initially support growth.

With the economy proving resilient in its ability to rebound from the impacts of COVID-19, inflation has emerged as the primary challenge. Rising prices have reduced the real purchasing power of households and pushed consumer confidence to near record lows. The Reserve Bank has begun tightening monetary policy in response to high inflation and signalled an intention to continue doing so at pace. This has had an immediate impact on house prices, which are forecast to fall throughout 2022 and 2023.

Interest rates are rising…

Since the Budget Update was published the Reserve Bank has signalled that interest rates are likely to increase faster and by more than assumed. A similar shift in policy stance has also occurred in other central banks, including in Australia and the United States.

In addition to the impact on economic activity from rising interest rates, real government consumption is forecast to decline from the second half of 2022 as expenditure related to COVID-19 unwinds. The overall result is a significant slowdown in real activity after this year.

… to deliberately slow activity and reduce inflation

A marked slowing in economic activity is a key channel by which central banks aim to bring inflation down and involves reducing the imbalance between supply and demand in economies. The Budget Update forecasts included quarterly growth in the 0.1% to 0.3% range throughout 2023 and into 2024, in contrast to an average around 0.6%.

Importantly, the current slowdown differs from previous recessions in multiple ways

The Budget Update forecasts portray an economy that differs from previous recession periods in a number of ways:

  • GDP is forecast to slow rather than experience multi-quarter falls.
  • Unemployment is low and while forecast to increase, is not expected to reach levels experienced in previous recessions.
  • Interest rates are rising to manage inflation rather than falling to shore-up demand. The resulting slowing in demand is the objective rather than an unexpected shock.
  • Net migration is anticipated to pick up and contribute to population growth.
  • The terms of trade remain around record levels.

A key contrast with times of recession is that real GDP is still anticipated to grow, albeit more slowly, rather than experiencing sizable falls (Figure 2).

Figure 2: GDP in previous recessions and the current slowdown[3]

Figure 2: GDP in previous recessions and the current slowdown

Sources: Statistics NZ, the Treasury

Unemployment is low and is not projected to reach levels seen in previous recessions

Slowing demand across the economy is forecast to see unemployment rise from its current lows. Increasing unemployment is a characteristic of most recessions. The current slowdown however sees unemployment start from a much lower base, and while we forecast a sustained rise in unemployment it does not reach the heights of past recessions (Figure 3).

Figure 3: Unemployment in previous recessions and the current slowdown

Figure 3: Unemployment in previous recessions and the current slowdown

Sources: Statistics NZ, the Treasury

Interest rates are rising to contain inflation rather than easing to support demand

As noted above, rising interest rates are a deliberate response to reduce demand. During recessions, central banks usually reduce interest rates to help promote demand. Figure 4 shows that this has typically been the case for New Zealand recessions.

Figure 4: Interest rate (90 days) movements in previous recessions and the current slowdown

Figure 4: Interest rate (90 days) movements in previous recessions and the current slowdown

Sources: Reserve Bank of NZ, the Treasury

Net migration to add to population growth

Another insight was that net migration typically falls around recessions. This can be a contributing factor to lower activity levels or can be a response to reduced opportunities in New Zealand. This is illustrated in Figure 5 with the Budget Update forecasts assuming a steady pick up in net migration gains, acknowledging that there is a high degree of uncertainty in relation to the timing and magnitude of the expected increase.

Figure 5: Net migration

Figure 5: Net migration

Sources: Statistics NZ, the Treasury

Terms of trade remain elevated

New Zealand's terms of trade (export prices relative to import prices) are at record highs. Large falls in the terms of trade were a feature of some earlier recessions. In general, the outlook for prices of key New Zealand exports remains favourable (and may even be supported in some instances by events such as the impact of the Ukraine conflict on commodity prices). Oil prices are obviously extremely high and constrained global supply is placing upward pressure on import prices in general. Over the forecast period we expect the terms of trade to remain elevated.

Falling house prices are a feature of the current outlook as well as a number of recessions

While house price falls were associated with several previous recessions there are important contextual differences (Figure 6). Current declines follow a period of rapid appreciation. They are also influenced by rising interest rates and represent one of the channels through which monetary policy influences activity.

Figure 6: Real house prices in previous recessions and current slowdown

Figure 6: Real house prices in previous recessions and current slowdown

Sources: CoreLogic, the Treasury

The Treasury is not forecasting a recession, but the risks are increasing

Given the slow rates of growth that are forecast, it is quite feasible that New Zealand could experience two quarters of negative growth at some point over the coming two years. However, the above discussion shows that Treasury is not forecasting a recession in any meaningful sense of the word. And although the economy is slowing, we are not currently seeing more worrisome signs of sharp (unanticipated) declines in demand. For example, ANZ noted that their June Business Outlook was nearing record lows but:

“… the main reason firms are so pessimistic on the outlook for profitability is not lack of demand, but rather supply-side constraints and cost pressures…

Traditional recession-type problems such as cashflow/debtors and low turnover remain well down the list, but interest rates are now warranting more of a mention, unsurprisingly.”

That said, we expect demand to slow, something that very weak consumer confidence suggests is in train and therefore expect easing demand to increasingly impact on business confidence going forward, particularly for firms providing more discretionary aspects of consumer spending. More rapid declines in house prices could see consumer spending slow by more than expected.

A key feature of past recessions has been the importance of global factors and therefore the risk of a more substantial downturn will be closely tied to the fortunes of the global economy. Risks in this regard have increased since the Budget forecasts, with monetary policy increasingly focused on reducing inflation in a backdrop of continued COVID-19 and geopolitical volatility.

In the US, the Federal Reserve Chair has noted that executing a “soft-landing” will be difficult but that a pathway to get there involves increasing interest rates to manage demand.

Slowing activity has real costs…

Regardless, of whether growth slows to low rates or activity falls slightly, the impacts are real - businesses face less demand and offer less employment. The slowing in economic activity in both the Reserve Bank's May Monetary Policy Statement (MPS) and the Budget Update forecast the number of unemployed to increase by around 50 thousand.

… but impacts are felt unevenly across the economy…

Just as recessions have different macro characteristics, they are also felt unevenly across industries and society.

Table 1 shows how GDP across industries fared during past recessions. Generally, around half of industries saw their GDP fall. The COVID-19 period was the exception where lockdown saw activity fall in all industries apart from public administration and safety. The worst affected industries tend to be mining, construction and manufacturing. Service industries tend to be relatively less affected, albeit with industries such as retail trade and accommodation generally experiencing larger declines than for the wider economy.

Table 1: Recessionary impacts by industry

Table 1: Recessionary impacts by industry

Sources: Statistics NZ, the Treasury

… and across society

Detailed data on how downturns affect subgroups of the population and economy is focused on more recent events, but still provides some useful lessons.

Rising unemployment impacts Māori and Pacific people the most…

Unemployment for Māori and Pacific peoples rises faster than the rest of the population. During the GFC the unemployment rates for Māori and Pacific peoples rose by 3.6 and 7.9 percentage points respectively, substantially more than the 2.4 percentage point aggregate increase.

… with regional impacts varying by recession

Downturns do not affect all regions equally. The unemployment rate in Northland rose by 4.9 percentage points in the GFC, and by 3.2 percentage points In Waikato, compared to a 2.4 percentage point increase nationally. However, the pattern to these more severe regional impacts is not consistent. Following the Asian Financial Crisis, Southland, Canterbury, Manawatu-Whanganui and Northland were the most affected regions. In the downturn of the early 1990s, the labour market effects of the shock were most concentrated in Auckland, where the unemployment rate rose 3.3 percentage points, compared to a rise of 1 percentage point or less in most other regions.

It's not clear that recessions increase measures of inequality

Despite these impacts, downturns do not appear to translate into increases in measured inequality. Measuring inequality is complex, and the choice of metrics matters, especially as key measures of material wellbeing (for example, income and consumption) might show different patterns. However, on a measure accounting for household size, composition and housing costs, the ratio between the top earning and bottom earning fifth of households has not noticeably risen during downturns. The exception is the early 1990s, where other wide-ranging reforms were taking place in the economy. Other measures of inequality (for example, income ratios before housing costs and the Gini coefficient) show similar patterns.

Rising interest rates are necessary to reduce inflation…

While the costs associated with slowing activity are real, the alternative of persistently high inflation and falling real incomes (as inflation exceeds wage growth) would likely have more severe economic implications.

… but as always policy may need to adapt

Both the current economic outlook and the impact of policy on that outlook are uncertain. However, policy is never fixed on an irrevocable path. If aspects of the outlook change, then both monetary and fiscal policy may also need to adapt. An OCR now at 2% (and predicted to peak close to 4%) provides some room in which to adjust monetary policy if required, a point made in the May MPS that “A larger and earlier increase in the OCR reduces the risk of inflation becoming persistent, while also providing more policy flexibility ahead in light of the highly uncertain global economic environment”.

Current fiscal policy settings, reflected in a projected return to surplus, supports monetary policy tightening to reduce inflationary pressures.

Similar to previous downturns, the forecast fiscal position is likely to erode if activity falls faster than currently anticipated as revenue declines and benefit expenditure increases. Fiscal policy may also need to consider distributional impacts, which as outlined above vary from downturn to downturn.

Promisingly, while the risks of slower economic activity have increased since Budget, tax revenue remains above forecast.

What is meant by Recession?

There is no single definition of what constitutes a recession, and a single definition is unlikely to be appropriate across time or countries. The National Bureau of Economic Research (NBER) in the United States has a long history of dating US business cycles. The NBER traditionally defines a recession as “a significant decline in economic activity that is spread across the economy and that lasts for more than a few months.” The dating of recessions by the NBER's committee involves assessing a wide variety of economic information to evaluate the depth, diffusion, and duration of a downturn.

Two quarters of negative growth?

A common definition for recession is that it reflects two or more quarters of negative real GDP growth. This definition is sometimes attributed to a 1974 New York Post article by Julius Shiskin (Commissioner of the United States Bureau of Labour Statistics)[4]. The article notes some approximate rules that cover the depth, diffusion, and duration dimensions above.

Duration - “declines in real G.N.P. for 2 consecutive quarters; a decline in industrial production over a six month period.”[5]

Depth - “A 1.5 per cent decline in real G.N.P.; a 1.5% decline in non-agricultural employment; a two-point rise in unemployment to a level of at least 6 per cent.”

Diffusion - “A decline in non agricultural employment in more than 75 per cent of industries over six month spans”.

Shiskin noted that many people use the common definition of two quarters of decline which, while simplistic, had worked quite well in the past for the United States. This common definition is only a very partial one and does not include many important aspects that are considered relevant. Because of this, this common definition is often described as a “technical recession”.

The “technical definition” has a number of limitations reflecting that it only covers one of the approximate rules (which Shiskin considered to only be a rough translation of the NBER's approach). For example, consecutive GDP declines of 0.1% would count as a technical recession but a 2% decline followed by a 0.5% rise would not - even though GDP would be lower in both quarters in the later example.

In lamenting the tendency for headline grabbing recession calls, economist Brian Easton outlines a range of challenges with calling recessions based on the latest GDP data[6]. These include that GDP estimates are subject to error (with this margin of error likely greater than the March quarter 0.2% decline) and that data is often revised over time. He also reiterates that there is no official definition of recession in New Zealand. There is no universal definition that would be appropriate for all countries.

Past NZ recessions varied in length and severity

Hall and McDermott (2016) identify 9 periods of recession in New Zealand post World War II. They use the Bry-Broshan dating algorithm and test the robustness of their results to alternative specifications. As shown in Table 2, the recessions vary in length from 2 to 7 quarters and involved contractions in GDP of approximately 1% to 9%.

While there is no such thing as a typical recession, on average they have involved contractions of 4.0% post-World War II, with those in the past 35 years (the period for which official quarterly GDP is available) averaging 2.1%.

The period over late 2010 illustrates that the classification of recession periods can be influenced by the methods used to identify them. Real GDP declined modestly over the second half of 2010, a period affected by the September 2010 Canterbury earthquake. The use of a different dating algorithm by Hall and McDermott identified this as another recession period but that it is a relatively marginal call.

Table 2: New Zealand recession characteristics

Table 2: New Zealand recession characteristics

Source: Hall and McDermott (2016), the Treasury

  1. [1] See Hall and McDermott(2016):
  2. [2] Pre-GFC descriptions are based on Reddell and Sleeman(2008):
  3. [3] The current slowdown period captures the period beyond 2022Q3 when growth is forecast to be at its weakest. In this and later graphs, time=0 is the period immediately prior to the first decline in GDP.
  4. [4] ‘The Changing Business Cycle’, Julius Shiskin, New York Times (1 December 1974).
  5. [5] At the time of Shiskin (1974) gross national product was the United States’ key economic metric. They have subsequently switched to using gross domestic product.
  6. [6]
Join the discussion on LinkedIn