How has the UK business cycle changed over the past 100 years?

Despite lessons from macroeconomics and economic history, recessions still strike with similar frequency and force to their predecessors. Business cycles remain unpredictable in their timing and severity – with spells of relative economic tranquillity or turbulence as much down to luck as policy.

The Economic History Society was established 100 years ago amid the major recession of 1926. This article asks how the British business cycle has changed in the century since then.

To answer this question, we need to establish the business cycle facts. The classic starting point in business cycle analysis is to identify the peaks and troughs in economic activity. The UK Business Cycle Dating Committee has developed a quarterly chronology covering the period from 1920 to 2010 (Broadberry et al, 2023), which is based on national accounts and expert judgement. As a member of this committee, I have tried to extend it forward to the present using the same approach (see Table 1).

Table 1: A chronology of the UK business cycle, 1920 to 2025

Peak	Trough	Contractionpeak to trough		Expansiontrough to peak		Cyclepeak to peak
Peak	Trough	Quarters	Amplitude (%)	Quarters	Amplitude (%)	Quarters	Amplitude (%)
1926:I	1926:III	2	-8.9	–	–	–	–
1930:I	1932:III	10	-6.9	14	17.3	16	6.9
1943:II	1947:II	16	-12.9	43	62.8	53	51.5
1973:II	1975:III	9	-5.4	104	142.4	120	111.2
1979:II	1981:I	7	-5.2	15	17.3	24	11.0
1990:II	1992:II	8	-2.4	37	35.8	44	28.7
2008:I	2009:II	5	-6.4	63	57.4	71	53.6
2019:IV	2020:II	2	-22.1	42	22.2	47	14.3
Mean		7.4	-8.8	45.4	50.7	53.6	39.6

Source: An update of Broadberry et al, 2023.

Over the past century, economic contractions – measured from peak to trough – have lasted slightly more than seven quarters on average, during which time real GDP (an inflation-adjusted measure of goods and services output) fell by 8.8%. Expansions – measured from trough to peak – have lasted more than 11 years on average, during which time economic activity rose by 50.7%. Across the business cycle, lasting more than 13 years on average, the gain in GDP in the UK has been 39.6%.

In terms of change over the last one hundred years, most of the metrics do not show clear trends. But we can see that contractions lengthened up to the Second World War and shortened afterwards. For example, contractions increased from two quarters in the recession of 1926 to 16 quarters in the recession of 1943-47 and returned to two quarters during the pandemic-induced recession of 2020.

In the UK’s modern economic history, there have been eight recessions (see Figure 1, which shows the dynamics of GDP through these).

In terms of the short-run losses, the first (1926) and last (2020) recessions of the century stand out. GDP fell by 9% in 1926, while the Covid-19 recession of 2020 saw output drop by more than 20%, marking it out as the episode with the largest output costs.

In terms of longer-run losses, the recession of 1943-47 – at the tail end of the Second World War and demobilisation – is striking. Output remained 13% down four years after the peak, whereas it was close to or had surpassed the pre-recession level in the other episodes.

Figure 1: Recession dynamics, 1920 to 2025

Source: An update of Broadberry et al, 2023.

Have the shocks changed over the past century? Experts have classified each of the recessions up to 2010 by the underlying shock (Broadberry et al, 2023). For the recession of 2020, it does not require a committee of experts to identify that the cause was Covid-19.

Other than a cluster of commodity price shocks and economic policy blunders between the 1970s and 1990s, the five remaining recessions have had five different causes: the labour supply shock of the General Strike in 1926; the international shock of the Great Depression in 1930-32; the war in 1943-47; and the global financial crisis of 2007-09 (see Table 2). Therefore, there is no clear or changing pattern.

Table 2: The origin of recessions, 1920 to 2025

Peak	Trough	Shock
1926:I	1926:III	Labour supply
1930:I	1932:III	International shock
1943:II	1947:II	War
1973:II	1975:III	Commodity price shock and economic policy
1979:II	1981:I	Commodity price shock and economic policy
1990:II	1992:II	Animal spirits and economic policy
2008:I	2009:II	Global financial crisis
2019:IV	2020:II	Covid-19

Source: An update of Broadberry et al, 2023.

Moving beyond the focus on recessions, we can also look at macroeconomic volatility or the scale of fluctuations in key economic measures – see Figure 2, which plots the decadal standard deviation of quarterly GDP growth at constant prices and seasonally adjusted.

The century began in turbulence with greater volatility than in the 1930s. The 1970s stand out as a breakpoint, dividing the first 50 years into a period of relatively high volatility and then half a century of more moderate volatility.

Of course, the startling (and rather worrying) feature of the figure is the sharp increase in volatility that we have experienced so far in the 2020s. This is driven by the very large drop in economic activity during the Covid-19 lockdowns and the very rapid recovery. If we set the affected quarters (2020:I to 2021:II) to the average of the unaffected quarters, essentially forcing their contribution to zero in the volatility calculations, we see that the majority of this spike is due to Covid-19 rather than a more general feature (highlighted in orange in Figure 2).

If we look past the pandemic as a one-off, the takeaway is that there has been a gradual reduction in macroeconomic volatility over the last century. If we consider Covid-19 as part of a return of major shocks, then the story is very different.

Figure 2: Macroeconomic volatility, 1920s to 2020s

Notes and sources: The standard deviation of quarterly real GDP growth from the Office for National Statistics (ONS), 2026. The orange bar shows the contribution from Covid-19 to the volatility of the 2020s.

Looking at an estimate of the business cycle in Figure 3, there are intervals in which it looks different, such as the stop-go cycles of the 1950s and 1960s and the so-called Great Moderation between the recessions of 1991 and 2007-09. Yet, there seems to have been some reversion in the last two decades to longer-run historical norms.

Figure 3: The UK business cycle, 1920 to 2025

Notes and sources: Based on the Hamilton (2018) filter and (log) real GDP from the ONS (2026); shaded areas denote recessions.

So, has the business cycle changed? The long view suggests that it has done so less than we might like to think. Despite developments in macroeconomics and the lessons of economic history, recessions still strike with similar frequency and force to their predecessors.

The deeper insight is that business cycles are shaped by shocks that arrive with unpredictable timing and severity. As a result, spells of relative tranquillity, such as the early 2000s, and turbulence, such as the present, are as much down to luck as policy.

In November 2003, the then chancellor of the exchequer, Gordon Brown, reflected on this history in a speech:

‘Remember the old days, what was called the British problem: stop-go, boom-bust, unstable cycles … Britain the country usually first in, worst hit and last out of any world downturn.’

A century of data shows that the UK may have put an end to the old problem of stop-go, but it has not escaped boom and bust.

Where can I find out more?

Dating business cycles in the United Kingdom, 1700-2010: Article in the Economic History Review.
Dating business cycles in the United Kingdom, 1700-2010: Article from the Economic Statistics Centre of Excellence (ESCoE).
Communicating the UK economic cycle: Explanation of movement in gross domestic product (GDP) and wider considerations around technical recessions in the UK from the ONS.