A time for giving

February 2010

Cathy Pharoah and Tom McKenzie review the impact of Christmas on giving and ask if we should season up our fundraising initiatives

It is often argued that the sector needs more detailed and frequent surveys of individual giving that can track monthly or quarterly change, to supplement its annual giving surveys. However, while month-on-month retail sales or consumption figures are tracked by industry and government to get early warnings of significant change in the economy, real underlying trends are usually established through figures comparing performance at different periods for exactly the same months, quarters or yearly cycles. It is not difficult to see why.

Seasonal factors likely to have a powerful short-term impact on our national expenditure on different kinds of goods can easily be identified – Christmas, summer sales, weather, availability, major calendar events such as Wimbledon and the football seasons etc. But how far do seasonal factors affect levels of giving? Is Christmas a high point and are there others throughout the year such as, for example, the end of the tax year in March/April or Easter? Do trends in charitable giving follow seasonal patterns in other household expenditure? This article analyses patterns in household giving to charities at monthly and quarterly periods, drawing on extensive data from the national Expenditure and Food Survey (EFS) [1], and identifies whether seasonal factors should be taken into account more when planning income generation strategy.

Quarterly trends in household donating

The first point to notice is that there is indeed an apparently quarterly ‘shape’ to donating, as can be seen in Figure 1 below, which tracks changes in giving by quarters, focusing on average weekly household giving for the quarter. (It should be noted that all the comparisons reported in this article relate to the average weekly giving amount for a specific quarter or month). The average weekly value of donations shows a U-shaped trend across the year, falling between the first and second/third quarters of the calendar year, and then rising in the fourth quarter. These results indicate a clear Christmas effect in the final quarter, while donations are down by 11 per cent in the summer quarters [2]. Comparing the donations trend-line with that for total spending by UK households shows that there are some striking similarities and differences of shape. Average weekly household spending rises steadily from a low point in the first post-Christmas quarter of the year through the summer quarters to a high in the fourth Christmas quarter.

It is possible that the summer quarters see raised general household expenditure because of the summer holidays including outings, travel and hotels, but less charitable giving because people may be outside the UK and beyond the reach of street collections, fundraising events, direct mail or telephone campaigns. The value of regular giving, through direct debit or the payroll etc, is less likely to be affected.

Monthly trends in household donating

But what about the possible effect of the end of the tax year on giving, a time when many higher-rate tax-paying donors and their financial advisers will be looking at particular opportunities to reduce annual tax bills while making a tax-effective donation to charity? This could particularly influence major giving, and some organisations target fundraising drives at this period, so it is worth trying to detect any effect on giving through breaking down quarterly trends into a more detailed monthly analysis of trends in donating.

The results shown in Figure 2 below show considerable variation in the value of average weekly donating across the months of the first quarter. January sees a low point in donating, mirroring the generally low level of household spending in the month after the excesses of Christmas. Subsequently, the value of average giving rises to an annual peak in February, remaining higher throughout March when it starts to decline to the low levels of giving seen in the summer months.

The extraordinary peak seen in the data for February is due to an unusually high level of giving in that month in 2007, which was the year when the UK economy was at its highest point before the credit bubble burst. So it is also possible that this higher level of giving in February, and giving through March and April also remaining higher than in January, are due to the effect of the tax-year end. The summer months of June and July see a dip in giving, which begins to rise again in August.

Do all donors experience the same seasonal effects?

With the huge variations in income and spending that exist in the UK, it cannot be assumed that all types of donor are affected in the same way, or to the same degree, by seasonal trends. So we compared patterns of average donating by different socio-economic groups across the year, and found that there were indeed some striking variations in giving behaviour.

The chart in Figure 2 above compares trend-lines for average giving across the year amongst five different household types, as defined by the ‘acorn’ classification used in much social and market research. In descending order of affluence, the groups are ‘wealthy achievers’, ‘urban prosperity’, ‘comfortably off’, ‘moderate means’ and ‘hard-pressed’[3]. Results show that the marked peak in giving in the early part of the year is mainly due to giving amongst top ‘wealthy achievers’ group, and partly to giving amongst the ‘comfortably off’. This would support the theory that there is a boost in giving in the early part of the year related to tax benefit, and that there are particular income-generation opportunities for charities whose targeting should specifically include financial advisors to the wealthy with up-to-date information on tax-effective giving. (Numbers in the ‘urban prosperity’ group were rather small, and results for this group should not be given too much significance.)

Interestingly, poorer households appear to be particularly charitable towards the end of the year. The only socio-economic group where donations are significantly higher in the fourth quarter than the rest of the year is the ‘hard-pressed’ group. At £1.46 the average weekly donations from this group remain considerably lower than the average £2.38 for other groups in the last quarter, but nonetheless they are up by 71 per cent compared to the rest of the year [4].

The figures indicate that wealthy achievers are the group most likely to step up their giving in August, and that the top income groups are generally less likely to reduce average giving in the summer months than at other times of the year. This may well be because of a greater propensity for the better-off to use planned, regular and tax-effective methods of giving than for other groups.

How strong is the Christmas effect?

The graphs show a marked rise in the average value of donations in November and December. Many charities in the UK target special appeals for donations at the Christmas; for some the campaigning begins in August/September and it is not uncommon for potential donors to return from their summer holidays to find charities’ Christmas catalogues in their post. So do more people give at Christmas than at any other time, and give more?

Participation in giving

A clear increase in participation in giving was found, but this was mainly concentrated among certain age groups:

While 50.1 per cent of UK households make a donation to charity in any given month [5] of the calendar year, this rises to 52.6 per cent in December, a 5 per cent increase.
Both older and younger householdsare significantly more likely to give to charity in December than at other times of the year.
Amongst households where the oldest person is aged between 25 and 34, the participation rate increases from 40 per cent in any given month to 48 per cent in December.
Where the oldest person is aged 65 or over, the participation rises from 54 per cent overall to 60 per cent in December.

These findings suggest that Christmas giving campaigns targeted at the middle-aged, traditionally a high donating-group, may be largely ineffectual as this group is likely to be focused principally on catering for, and giving to, their families at this time. Younger people ‘about town’ with increased time devoted to shopping and socialising are likely to be a good target for spontaneous and impromptu giving, while older people with time to plan their Christmas expenditure may be a much better target for considered Christmas appeals.

What amounts are given at Christmas?

The story around the value of giving at Christmas is more complicated: it was found that:

The average weekly amount donated appeared to rise in December compared to the rest of the year from £2.03 (January to November) to £2.42.
This represents a 19 per cent increase [6].

But this increase in the average weekly amount given in December is mainly because more people are giving and not because donations are larger. When considering the whole of the fourth quarter (October to December) relative to the rest of the year (January to September), average donations go up in the fourth quarter by about 8 per cent. There is little evidence to suggest that people make bigger gifts at Christmas. When the amounts donated are considered as a proportion of household income, there is no statistically meaningful difference between the Christmas period and the rest of the year, as household incomes also rise in December due, in part, to factors such as annual bonuses, increased seasonal employment and, possibly using income from savings. It is encouraging to find that where income is higher, the value of donations also goes up, but this tends to be true of charitable giving generally, and not just at Christmas.

Less encouraging for charities is the finding that any seasonal rise in donations is dwarfed by increases in spending on the presents and cash gifts we make to others. Cash gifts more than double and spending on presents more than triples in December compared with the rest of the year [7]. Average UK household spending tends to rise by 13 per cent, or £64 per household per week in December compared with the other months of the year:

Extra weekly donations to charity contribute less than 1 per cent to the total increase, at 0.6 per cent.
Seven per cent is given away as (extra) presents and cash to other households.

More seasoning in income generation initiatives?

The evidence shows how far charitable giving, like other types of household expenditure, is sensitive to seasonal effects. Taking these into consideration when planning income generation initiatives could increase the effectiveness of targeting and returns on investment. The evidence suggests that Christmas may not be the best time for trying to persuade the wealthiest donors to give, or to give more, and that early spring, when thoughts turn to tax, may be a more effective point for this group.

Less well-off people, however, a group that includes younger donors, appear extremely amenable to persuasion about giving at Christmas and could be targeted more with imaginative spontaneous campaigns. January is a predictably low point for donating as it is for other kinds of household spending, and probably best avoided. But the pattern of falling donating in the summer months merits further thought. Clearly the problem is not, as in January, that households are spending less: generally spending rises in summer. The problem is therefore likely to be less solicitation. Charities could potentially capture a slice of that average £4,000 a year we spend on holidays – unless, of course, they are all on holiday themselves!

[1] Covering the years 2001 to 2007 gave a total sample of 46099 households.
[2] This result is significant at the 5 per cent level (p=0.028)
[3] See www.caci.co.uk/acorn/acornmap.asp for details on the classification.
[4] This result is significant at the 1 per cent level (p=0.001). The figures are based on spending ~ recorded in a diary over a two-week period and have been converted to November 2009 pounds. Averages have been weighted to account for structural differences between the sample and the population.
[5] Households in the EFS keep a diary of their spending over a period of two weeks. We have imputed the probability of households giving to charity in any given four-week period (month) from the observed participation rates in the two-week survey period.
[6] This result is only marginally statistically significant (p=0.071) due to high variation in actual amounts donated.
[7] These differences are statistically significant at the 0.1 per cent level.