The Diamond Jubilee and the London 2012 Olympic and Paralympic Games made 2012 an unusual and difficult year for policymakers and anybody interested in understanding the behaviour of the UK economy. ONS designated both events as ‘special events’ under the ONS special events policy as they had a potentially significant effect on many key economic statistics. This article takes a retrospective look at each event and considers the impact on a range of published economic indicators. It does not attempt to put a definitive figure on the impact of each factor, as it is not possible in either case to identify with certainty what the counterfactual would have looked like – in other words what might have happened in the absence of the event itself. However it looks at the possible orders of magnitude of the effects, and considers the contribution that they made to the volatility of some key economic series during 2012.
The first such event was the additional bank holiday called to celebrate the Queen’s Diamond Jubilee in June. In addition, the timing of the usual late-May holiday was deferred by a week. This created a four day holiday (including a weekend) for many people in early June.
The economic impact of creating an additional bank holiday varies between different parts of the economy. Those sectors that work on a five-day working week, Monday to Friday, might be expected to produce less in a month with more holidays than usual and hence fewer working days. Other parts of the economy, such as restaurants and hotels, might see higher output as a result of the extra holiday.
Monthly data for both May and June were affected by the altered bank holiday arrangements. The month of May featured one more working day than usual, while June had two extra holidays and therefore lost two working days compared with a ‘normal’ June.
In terms of calendar year quarters, the second quarter (covering April to June) had one fewer working day than normal as a result of the Diamond Jubilee.
Although ONS undertakes seasonal adjustment of its data series to allow for regular seasonal patterns of behaviour, including adjustment for the varying number of working days in different periods in some industries, it does not make comparable adjustments for occasional events such as the Diamond Jubilee. Any impact from this would therefore be expected to show up in ONS’ published economic statistics.
As well as the Diamond Jubilee, the second quarter was characterised by unseasonably wet weather in April and June. This may also have had an effect on the economy, particularly in sectors such as construction where some activity is dependent on the weather, however it is not been possible to distinguish the impact of the weather on economic activity from that of the Diamond Jubilee.
The second special event was the staging of the Olympic and Paralympic Games in London in the July to September quarter. The Olympics ran from 27 July to 12 August, with the Paralympics following between 29 August and 9 September.
The Games may have had an impact on a number of economic data series as a result of altered behaviour by consumers and businesses. For instance it might be expected that there would be a different pattern of foreign visits to the UK, with some people – including competitors and spectators – attracted by the Olympics, while others may have been deterred by fears of overcrowding.
In order to ensure comparability of statistics between different countries, ONS statistics for the UK economy conform to international reporting regulations laid down by the System of National Accounts (SNA) and the European System of Accounts (ESA). As a result, the national accounts are produced to show accruals of income and spending. This means that expenditure is recorded when goods or services are consumed, and not necessarily when they are purchased.
This distinction is important for the Olympics. Many people purchased Olympic tickets well in advance of the Games themselves, with the first substantial batches of tickets being sold in spring 2011. But regardless of when the tickets were purchased, they are counted as spending and output in the economy in the third quarter of 2012, when the Games actually took place. Since such ticket sales amounted to some £580 million, or around 0.2% of GDP, this single factor produced a temporary boost to GDP in that quarter.
ONS published an article in October 2012 detailing this, and other potential effects of the London 2012 Olympics on the UK national accounts.
Figure 1 shows the level of output during 2011 and 2012 as measured by the index of services and the Index of Production. A clear downward spike occurs in both series in June 2012, and it is reasonable to suppose that the Diamond Jubilee was the main reason for this.
Output fell by 1.4% between May and June 2012 in the services sector and by 2.4% in the industrial sector. In both cases, output returned to the May level, or above, in July.
The services sector is the largest single part of the UK economy, accounting for 77% of GDP, while the Index of Production adds a further 16%. Using these weights, the fall in total output implied by the monthly figures for June is almost 1.5%, equivalent to a quarterly change in GDP of around 0.5 percentage points.
If the whole of the fall in output between May and June was due to the Diamond Jubilee, this suggests that GDP would have been 0.5 percentage points higher in the second quarter in the absence of the extra holiday than shown in the published figures.
However it cannot be assumed that the downward spike in services output in June 2012 was wholly caused by the Diamond Jubilee. The level of output in May could have been swollen, compared with previous years, by the absence of the usual late-May bank holiday, suggesting that the Diamond Jubilee’s impact is probably smaller than 0.5 percentage points. In addition, the monthly time series is naturally volatile, and it is possible that the movements in output during this period in part reflect the normal period-to-period fluctuations that are characteristic of economic behaviour.
The impact of the Diamond Jubilee is therefore likely to have been around 0.3 to 0.4% of GDP. This is clearly a very approximate approach, but it is impossible to be more precise. Such an estimate should therefore be regarded as indicating the broad order of magnitude rather than a statistically robust estimate.
During June, the output of the services sector (on a seasonally-adjusted basis) decreased by 1.4% compared with May. The fall was wide-spread across service industries, with the exception of government & other services, which grew by 0.2% in the month. Output in the transport, storage & communication sector fell by 3.3%, and the distribution, hotels & restaurants, and business services & finance sectors also saw contractions, falling by 1.7% and 1.8% respectively.
The services sector as a whole grew by 1.4% in July following the 1.4% contraction in June, growth in government & other services picked up while the other main sub-sectors largely regained the fall in output from June.
The index of production fell by 2.4% during June and then rebounded, growing by 2.9% in July. Within this, manufacturing output fell by 3.0% in June, the strongest month-on-month contraction since November 2008. It almost regained its May level in July, with growth of 3.0% in the month.
Construction output fell particularly sharply in June, even allowing for its weak trend through 2012. Unseasonably wet weather may have affected this sector of the economy more severely than others as some construction activity is dependent on weather conditions. The April-to-June period was the wettest on record for the time of year. Output declined by 6.7% in June, compared with growth in May and July of 2.7% and 2.8% respectively. However it is impossible to identify the extent to which this can be attributed to the Diamond Jubilee, as distinct from other factors including the weather.
The loss of a working day due to the Diamond Jubilee was not expected to have a large effect on headline labour market statistics, such as employment and unemployment, as only a relatively small number of workers are contracted on a daily basis. Changes in these series at the time of the Diamond Jubilee and the Olympics are well within the normal range of fluctuation. However, the Diamond Jubilee does seem to have had an effect on hours worked in those quarters.
This can be seen in Figure 4, showing the rolling three-month series of average actual weekly hours worked by all workers. In March-to-May 2012, average weekly hours were unusually high at 32 hours per week. This was in part due to the absence of the spring school half-term holidays and the Spring Bank Holiday in May 2012. Hours in both April-to-June and May-to-July decreased to 31.7 and 31.6 hours per week respectively. This was in line with the occurrence of the Diamond Jubilee celebrations, which included one fewer working day in 2012 than in previous years.
Looking at the LFS microdata points towards a sizeable downward effect on average actual hours worked in the first week of June by around 17% compared with the same week of 2011.
The Olympics ran from 27 July to 12 August and the Paralympics from 29 August to 9 September. In contrast to the Jubilee, the impact of the Games is likely to have been more diverse and more dispersed. A whole range of economic activities may have been affected, including overseas travel and tourism and consumer spending. While we might expect the overall impact on economic activity to be positive, in most cases it is impossible to distinguish the precise effects of the Games from the normal variation that typifies most economic time series. And even the direction of the impact may not always be clear to the extent that the Games might have displaced other activities, the impact in some sectors could even be negative.
One area where the impact is unambiguous and directly quantifiable is with the sale of Olympic and Paralympic tickets. Spending on ticket sales should, according to international reporting guidelines, be accrued to the period in which the benefits were consumed – in this case during the Games themselves – rather than the time of actual purchase. The estimated £580 million of ticket sales therefore added 0.2 percentage points to GDP growth in the third quarter. This is included as part of the output of the arts, entertainment & recreation industry, and the effect can be seen in Figure 5.
In Figure 6, Olympic ticket sales are part of “government & other services”. As well as the increase in output in the third quarter of 2012, Figure 6 also indicates a strong performance by distribution, hotels & restaurants, whose output rose 1.8% in the same quarter.
Overseas travel and tourism
The International Passenger Survey (IPS) is the main vehicle used to collect information about the number of overseas visitors to the UK and trips aboard by UK residents. It also captures the amount spent for each of these groups.
For the duration of the Games, the IPS introduced additional sampling - covering more passenger entry points, more frequently – to ensure that the information was available to provide a robust and representative picture of the scale of inbound and outbound tourism.
The IPS estimated that 698,000 overseas residents completed visits to the UK either for the purpose of watching, participating in or working at the London 2012 Games, or to attend a ticketed London 2012 event even though they were visiting the UK primarily for a different purpose. This represents approximately 8% of the typical volume of visits for the third quarter of any year.
In the three months July to September 2012, there were in total 8.8 million visits to the UK by overseas residents, a fall of 3.9% compared with the same period in 2011 when considered in seasonally-adjusted terms, as Figure 7 indicates. This represents a break with the pattern of modest year-on-year growth in the number of overseas visitors in adjacent periods.
However the spending (£6.5 billion in absolute terms) associated with these visits was 10.3% higher than the previous year when considered seasonally adjusted. Visitors who came to the UK for the purpose of the Olympics spent an average of £1,553 on the visit (including any tickets bought in advance or during the visit), which is more than double the average spend among all visitors (£731). Spending associated with overnight visits to London increased by 11% from £2.8 billion to £3.2 billion, and again the purchase of Olympics and Paralympics tickets will have accounted for part of this increase.
The number of visits abroad by UK residents fell slightly (down 0.2% compared with a year ago) to 19.2 million.
Hosting the Games required substantial labour input, both for the construction of facilities and for running the event itself. Additional employment for construction had been spread over a period of months and years leading up to the Games, and this trend is not therefore readily discernible in the published labour market statistics. We might expect to see additional employment in and around the period of the Games as organisers took on staff to set up and run the Games. However, there is little indication of an Olympic effect in the headline figures for full-time or part-time workers.
In accordance with international guidelines, the estimated 84,000 unpaid volunteers (which mainly consisted of the so-called ‘Games Makers’) are not included in the employment estimates unless they had a paid job in addition to the volunteering. Many of the contracted workers may have already been in employment prior to the third quarter, therefore making no change to their workforce status.
The London Organising Committee for the Olympic Games (LOCOG) report that over 200,000 people were involved, either as volunteers, paid employees or contractors. While the Games were under way, the total paid workforce peaked at over 8,000 persons. LOCOG estimated that after the games 2,500 temporary construction jobs on the site at the peak of work to transform the Olympic village into a form suited to its long-term role.
However, these estimates measure only those directly involved in the Games. It is likely that the hospitality, transport and entertainment industries will also have seen increased hours worked and an increase in the number of temporary staff employed, especially in London.
While not having a large impact on the headline labour market figures the Games seem to have had an effect on hours worked in the second and third quarters of 2012.
Hours appear to pick up from the Diamond Jubilee over the Olympic period reaching a high of 32.1 hours per week in August-to-October 2012, as can be seen previously in Figure 4. Other factors such as the autumn half-term school holiday falling in the November Labour Force Survey (LFS) reporting period instead of the usual October reporting period, in addition to July and August having strong weekly hours worked partly explain this increase. This saw hours worked rise by more than would be expected at that time of year. Average weekly hours appeared to return to the underlying long-term trend in September-to-November 2012.
LFS microdata show that average hours worked were slightly higher than usual (by around 3.5%) during the last week of July and the first two weeks of August. This suggests that people took less time off for holidays during the Games, although it cannot be attributed completely to these events. Possible explanations include holidays being taken around the Diamond Jubilee long-weekend rather than in August; some occupations might have placed restrictions on staff leave during the Games; and others may have choosen to stay in the UK to be closer to – if not attend – the Games.
During the third quarter, anecdotal reports suggested that households were spending less on internet shopping as they spent more time watching coverage of the Games. Online sales have been increasing as a proportion of total retail sales for some years, rising from 3.4% in 2007 to 9.3% in 2012. However, internet sales as a proportion of total retail sales (Figure 8) fell in August 2012 to 8.2% over the month compared with the average of 9.2% throughout 2012. The Games also had an impact on retail sales after the Games with bike sales reaching a record high and the sporting goods & toys sector dominating growth in retail sales into the autumn. Feedback from retailers suggests that sales in sporting goods stores were boosted further by an increase in the sales of football shirts with both the start of the new season and the European championship.
The discussion above suggests that the Diamond Jubilee could have reduced the level of UK GDP in the second quarter of 2012 by 0.3 to 0.4 percentage points. However it is impossible to estimate the exact effect, and this should be viewed as the broad order of magnitude rather than anything more precise. The impact of the Games is more difficult to determine, and even the direction of the impact is uncertain in some areas. However it is clear that there was a 0.2 percentage points boost to GDP in the third quarter of 2013 from the effect of ticket sales.
Since GDP growth in the second quarter was smaller through the effect of the Diamond Jubilee holiday, the change to the third quarter is automatically bigger than it would otherwise have been. Similarly the higher estimate of GDP in the third quarter from inclusion of Olympics and Paralympics Games ticket sales means that the change in the fourth quarter was automatically smaller by comparison.
It is possible that some of the output lost in the second quarter may have been made up in the third as firms sought to recover the lost production time in order to meet delivery deadlines. However it is impossible to know if this occurred or to what extent.
This complicated the task for those interested in understanding what the published quarterly GDP estimates (see Figure 9) had to say about the evolution of the economy during 2012.
For illustrative purposes, Table 1 below indicates the possible impact that the 2012 special events might have had on GDP growth during the year.
|2012 Q1||2012 Q2||2012 Q3||2012 Q4||2013 Q1|
|(1) Latest published estimates of quarter-on-quarter real GDP growth||-0.1||-0.4||0.9||-0.3||0.3|
|(2) Possible impact of Diamond Jubilee on level of GDP||-||-0.3 to -0.4||-||-||-|
|(3) Possible impact of Olympics on level of GDP||-||-||0.2||-||-|
|(4) Possible impact of Diamond Jubilee on growth of GDP||-||-0.3 to -0.4||+0.3 to +0.4||-||-|
|(5) Possible impact of Olympics on growth of GDP||-||-||0.2||-0.2||-|
|(6) Possible impact of all 2012 special events on growth of GDP||-||-0.3 to -0.4||+0.5 to +0.6||-0.2||-|
|(4) + (5)|
|(7) Possible resulting path of GDP growth||-0.1||-0.1 to 0||+0.3 to +0.4||-0.1||0.3|
|(1) – (6)|
It is clear that a significant part of the volatility of GDP growth during the year may be attributable to the impact of the Diamond Jubilee and the London 2012 Olympic and Paralympic Games. However this does not change the overall picture of an economy that was growing slowly at best during 2012.
Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: firstname.lastname@example.org