DISCUSSION DRAFT

THE AVERAGE GIFT SIZE FACTOR

A Research Project to
Explore The Relevance of Average Gift Sizes
In Evaluating the Fund-raising Costs
of Charitable Organizations

Sponsored by:

INDEPENDENT SECTOR
1828 L Street, N.W
Washington, D. C. 20036

Staff:

Wilson C. Levis, Director
Anne New, Consultant
Average Gift Size Study Project

Draft Reprinted With Permission of:

Philanthropy Monthly
Box 989
New Milford, CT 06776
July/August 1981

This project is funded, in part, by a grant from
The Equitable Life Assurance Society of the United States.

Comments should be sent to:
Average Gift Size Study Project
INDEPENDENT SECTOR 1828 L Street, NW.
Washington, D. C. 20036 (202) 6594007

By Wilson C. Levis
Director, Average Gift Size Study Project, and
Anne New, Project Consultant

"A growing number of users of financial statements are seeking information that will enable them to evaluate fund-raising costs."

Paragraph 94, Statement of Position 78-10
American Institute of Certified Public Accountants (AICPA)

Wilson C. Levis and Anne New are independent consultants in the non-profit field. Among their current assignments both are project consultants for the National Center for Charitable Statistics at the National Information Bureau and for the Average Gift Size Study Project at INDEPENDENT SECTOR.

Government regulators, voluntary watchdog organizations, the media, and the general public all have an interest in knowing how much of the money contributed to charity is spent for fund-raising...and in determining how much is too much.

Until recently, a single method of evaluation has been the principal technique employed by almost everyone involved in evaluating fund-raising costs: the ratio of fund-raising expense to public support--usually expressed as a percentage.

This method of evaluation required two elements: the total amount of public support and fund-raising expense. It was simple to figure: the first element was divided into the second and the result multiplied by 100 to obtain a percentage figure. If the percentage was higher than some arbitrary ceiling--say, 30%--the fund-raising costs were judged to be excessive.

Now, however, a new method of analysis is being put forward and tested in a project sponsored by INDEPENDENT SECTOR and funded in part under a grant from The Equitable Life Assurance Society of the United States. The project is called the Average Gift Size Study Project because it will explore what may be a more meaningful way to determine acceptable fund-raising costs for a wide range of average gift sizes.

The proposed method would not necessarily replace the simple cost percentage but would supplement and refine it in important ways: (1) it would respond to the donor's question to which the fund-raising cost percentage has no ready answer, "How much of my gift will be used for fund-raising?"; (2) it would provide new insights, particularly useful in evaluating fund-raising costs of the many charities that have large numbers of supporters who give $2.00, $3.00, $4.00 or $5.00 each; (3) it would highlight significant differences between various types of agencies and their constituencies--differences that may have a direct bearing on the "reasonableness" of agencies' fund-raising costs.

The project postulates that a third element, the number of gifts, should be added to the information available to donors, regulators, watch dog agencies, and the public. With this element any interested person can quickly calculate two additional pieces of information: The average gift size (public support divided by the number of gifts) and the average cost per gift (fund-raising expense divided by the number of gifts).

Table 1, Average Gift Size Factor in Fund-raising Cost Evaluation, presents this basic concept.

Why should these two figures be considered in evaluating fund-raising costs?

For one thing, within certain limits, the average size of the gifts received by a given agency appears to define the nature of the agency's constituency. That is, some agencies appear to have constituencies that give them only small amounts. Other agencies have constituencies that can and do give in large amounts. There are great differences in the average sizes of the gifts received by the two types of agency. As will be seen, these differences can have a marked effect on fund-raising cost percentages.

On the other hand, the average cost per gift--again within certain limits--may help to define the effectiveness of the agency's fund-raising operation. Research on this point is a key objective of the project.

By looking at these two figures (i.e., average gift size and average cost per gift) in typical cases, we may discover relationships that will provide a more realistic basis for evaluating the cost effectiveness--and therefore the reasonableness--of an organization's fund-raising operation and fund-raising expense.

The project will, therefore, begin by testing the following theories.

Theory 1. The average cost per gift (last year's fund-raising expense divided by the number of gifts) provides current donors with a reasonably reliable estimate of how much of each gift was used for fund-raising; it also provides prospective donors with a reasonably reliable estimate of how much of each gift will be used for fund-raising.

Theory 2. The technique of counting the number of gifts and dividing the total into the fund-raising expense is valid for computing average cost per gift.

Theory 3. Counting the number of gifts--both overall and by category, if desired--(see Table 2) is simple and inexpensive; can easily be audited and reported in the notes to financial statements; is currently being done in-house as a part of virtually all fund-raising operations.

Theory 4. The fund-raising cost percentage (last year's fund-raising expense divided by last year's public support times 100) does not provide prospective donors with a reasonably reliable estimate of how much of each donor's gift may be used for fund-raising.

Theory 5. Organizations with the lowest costs per gift always use the lowest percentage of each donor's gift for fund-raising regardless of the overall fund-raising cost percentage.

Cost accounting, which includes the calculation of unit costs, is standard operating procedure in any well-run business. The project will seek to show that the same cost-accounting principles can and should be used in the evaluation of charities' fund-raising costs.

To test its theories the project will:

(1) collect existing gift-size data from prior studies and new data from organizations willing to participate;

(2) conduct in-depth interviews with leaders in accounting, law, and charitable oversight, as well as donors and the executives of participating organizations;

(3) encourage all interested persons (e.g., readers of this article) to comment on the concepts set forth here;

(4) encourage all interested organizations to provide average gift size data;

(5) publish a final paper based on the comments received, interviews conducted, and data collected; the paper will be distributed as widely as possible, but particularly to those who might find it useful in their evaluation of charitable fund-raising and in developing principles and methods for financial disclosure by charitable agencies.

In testing the theories, let us first examine a typical kind of solicitation to which theories 1, 2, and 3 might be applied to a special mail appeal to prospects for $10 to $200 gifts (see Item 4, Table 2).

Campaign Data

A.	Number of prospects	2,500
B.	Direct fund-raising cost	$15,000.00
C.	Cost per prospect (cost per solicitation: B / A)	$ 6.00

To demonstrate the use of the proposed type of unit cost accounting no contribution envelope will be opened until after the completion of the appeal.

D.	Number of contribution envelopes received (number of gifts)	2,000
E.	Percentage of participation (percentage of response: D / A x 100)	80%
F.	Cost per gift (cost per contribution envelope received: B / D)	$7.50*

*Note that the cost of appealing to each prospect is $6.00 but only 80% of the prospects responded positively by sending in a gift envelope. The cost of nonrespondents (the other 20%) must be considered part of the total cost of obtaining the gifts that were received. Thus, the average cost per gift is not $6.00 but $7.50 as noted above.

Comment: In this appeal the average cost per gift of $7.50 was found by the process of unit cost accounting before the gift envelopes were opened. To open all the envelopes, tabulate the dollar results, and compute the average fund-raising cost percentage would not appear to provide any additional cost-accounting information that could improve the accuracy of the $7.50 unit cost per gift.

Each donor in the special mail appeal outlined above could, if desired, receive a report on how much of his or her gift was used for fund-raising. The acknowledgment letter, prepared immediately after each gift envelope was opened (and before dollar results are totaled), could say something like this:

Thank you for your gift of $ ___. The average cost of obtaining each gift in this appeal was $7.50. Therefore, the percentage of your gift which was used for fund-raising was ___%, which is $7.50 divided by the amount of your gift times 100.

Note: It is not the purpose of the project to advocate sending such a statement to donors. The above example simply shows that the information is easily available if unit cost accounting is used. The donor's question: "How much of my gift was spent for fund-raising?" can be given a reasonable answer.

At this point, let's pause and review the relevance of the example so far to theories 1, 2, and 3.

Theory 1 refers to a reasonably reliable estimate of the amount per gift that was and probably will be spent on fund-raising. The agency in our example can, if it chooses, quickly inform donors how much of each gift was used for fund-raising, as we have seen. In next year's appeal, a statement of the average cost per gift in this year's appeal would give prospective donors a figure they could relate directly to the amount of the gift they intend to make. (The agency may wish to make allowance for late gifts and rising costs but the effect of such adjustments on the average cost per gift should be very small in a 12-month period.) In this preliminary example, the average cost per gift does appear to be a reasonably reliable estimate of the amount of a gift that will be spent on fund-raising.

Theories 2 and 3 relate to the technique of computing the average cost per gift and the simplicity and low cost of the process. When used as in the example above, the technique should be reliable. Since the basic solicitation costs apply equally to all pieces in the mailing, the cost of obtaining any specific gift will be the same as that of all the others. As shown, the method is simple and inexpensive. It uses materials (the contribution envelopes and campaign expense records) that are on hand and easily audited. There seems, so far, to be evidence that theories 2 and 3 will prove to be correct under the above circumstances. The project will proceed to test them in more complex situations.

Let's now look at an example of theory 4--that the fund-raising cost percentage so commonly used in evaluating the reasonableness of fund-raising costs does not provide prospective donors with a reasonably reliable estimate of how much of each donor's gift may be used for fund-raising.

We find the example in the special mail appeal we have already reviewed. Only now we must look at the total dollar results. When all the envelopes have been opened and their contents tallied, the figures might be as follows:

G.	Total dollar results	$100,000.00
H.	Average gift size (G / D)	$50.00
I.	Average percentage of fund-raising cost to total dollar results (B / G x 100)	15%

How reliable is the percentage figure as an estimate of how much of a donor's gift has been or will be used for fund-raising?

It is reliable for gifts at or near the average gift size. The use of the percentage, however, can greatly distort the true fund-raising costs of any gifts that are not close to the average in size.

In the special mail appeal example, the average fund-raising cost percentage was 15%. But, the fund-raising cost of a $1000 gift could not be only $1.50 (15% of $10.00) when it cost $6.00 per prospect to conduct the appeal. Further, the fund-raising cost of a $200.00 gift could not be as high as $30.00 (15% of $200.00) when the cost per prospect was only $6.00.

In other words, the fund-raising cost percentage of gifts in the lower dollar range will be higher than 15% and for gifts in the upper dollar range will be lower and the use of the average fund-raising cost percentage tends to conceal this fact.

Why is this significant?

Because, under certain circumstances, the use of the overall fund-raising cost percentage alone would cause a donor to make a decision exactly opposite to his or her intent.

Here is an example. (This example assumes that all other factors are equal and the donor's decision would then be based solely on an evaluation of fund-raising costs.)

A prospective donor might look at an agency's financial report and see the overall fund-raising cost percentage was 30%. The donor had intended to make a gift of $1,000 but would decide against it because $300 of the $1,000 would be used for fund-raising. In fact, the average cost per gift may have been only $5.00. A large number of gifts in the $15 - $20 range could cause the average fund-raising cost percentage to be 30% even though the actual cost percentage of a $1,000 gift was less than 1%.

Conversely, the same donor could incorrectly decide to make a $100 gift to an organization because an overall fund-raising cost percentage of 10% led the donor to believe only 10% or $10 of his $100 would be used for fund-raising. In fact, the organization may obtain many very large gifts, have an average gift size of $500 and have an average cost per gift of $50 or 50% of his $100 gift.

Thus, there seem to be strong indications that the fund-raising cost percentage may not provide this prospective donor with a reasonably reliable estimate of how much of his or her gift may be used for fund-raising.

Further, based on the above analysis, it would seem that disclosure of average cost-per-gift could be appropriate at the time of solicitation (if point of solicitation disclosure were called for): "Last year our average per-gift fund-raising cost (for a similar appeal) was $7.50."

Another Example

We believe an educated donor is our best donor. Because fund-raising overhead has become a major point of discussion in the press and elsewhere, we want you to know exactly how much of your gift will be used for fund-raising overhead.

In 1980, we used an average of $2.25 per gift for fund-raising overhead.

The following gift table is provided for your information. Please review it when considering your gift.

CHARITY 45 GIFT TABLE
			Fund-Raising Overhead
Check one	Gift Categories	Amount of Your Gift	Estimated Cost per Gift	Percent of Your Gift
-	Average Gift	$ 5.00	$2.25	45%
-	General gift	10.00	$2.25	23
-	Special gift	25.00	$2.25	9
-	Advanced gift	50.00	$2.25	5
-	Major gift	100.00	$2.25	2
-	Leadership gift	250.00	$2.25	1
		or more $________	$2.25	less than 1%

The fifth and last theory is perhaps the most difficult of all to grasp yet it states a simple mathematical fact: that organizations with the lowest cost per gift always use the lowest percentage of each donor's gift for fund-raising regardless of the overall fund-raising cost percentage.

Table 3, Fund-Raising Performance Comparison, shows what the mathematics are using the examples of table l. In the table, Charity 45 has mostly $5.00 gifts. It has a cost per gift of $2.25 and an average fund-raising cost percentage of 45%. Charity 15 receives mostly $33.33 gifts. It has an average cost per gift of $5.00 and an average fund-raising cost percentage of 15%. Note that by running your finger across the table at any gift level in Column A, the figure on that line in Column C is always lower than the figure in Column E. Comment: It is clear that since the cost per gift remains constant for each size of gift, Charity 45 (which has the lowest cost per gift) always uses the lowest percentage of each donor's gift for fund-raising. The use of the overall percentage alone without the cost per gift conceals this important fact and distorts the donor's perception.

Table 5, Fund-Raising Cost Percentages for Various Numbers of Gifts, shows how fund-raising costs would drop if a donor gave fewer but larger gifts or paid installments on a pledge without additional solicitations.

Further, the problem of educating donors could be turned into an opportunity to increase fund-raising productivity. What if a donor could be upgraded from 10 gifts at $10 (total $100) to eight gifts at $20 (total $160) with the average cost per gift remaining at $5.00? While the gross contribution would go up from $100 to $160, an increase of 60%, the net available for program would go up from $50 to $120, an increase of 140%. For small donors with a deep interest in the charitable organization, an educational program to encourage consolidated gifts or pledges might be well worthwhile.

Considerations for Regulating Fund-Raising Costs

We began by noting that a single method of evaluating charitable fund-raising costs--the fund-raising cost percentage--is widely used.

This does not mean that it is generally viewed as the ideal method or even a method that has resulted in effective regulation of charitable fund-raising. On the contrary, the evidence indicates that the use of the fund-raising cost percentage has caused resentment among charities, controversy between charities and the charitable oversight system, and frustration on the part of both charities and regulators.

Obviously, if a charity is to do its job, it must realize enough funds from its contributors to support its program adequately. To say this, however, is not to say what percentage of each contributed dollar may be spent on fund-raising and what percentage must be spent on program.

Opinions as to what is a reasonable fund-raising cost percentage can vary widely. Some regulators may choose 15%, others 30%. Many would reject 45% as too high. Yet disclosure of the number and size of gifts of a 45% organization might reveal that it has a lower cost per gift--and, therefore, uses less of each gift for fund-raising--than its neighbor with a 15% fund-raising cost.

Further, unless cost per gift information is disclosed, the oversight system cannot impose fund-raising standards and regulations on the high fund-raising cost percentage, high cost per gift organizations without at the same time impacting negatively on the well managed, low cost per gift organizations with above average percentage figures.

The following question (which barrows the accounting profession's language to emphasize the purpose of financial disclosure) indicates another aspect (regulation and oversight) of the current study.

"[In order that] users [(especially donors)] of financial statements [can have] information that will enable them to [more accurately and completely] evaluate [and more effectively regulate] fund-raising costs,*" should regulators and others in the charitable oversight system and accountants require charities to report the number of gifts they receive each year so average gift size and cost per gift factors can be calculated?

Our inquiry will seek data that may help to provide the answer.

We welcome the comments of readers of this paper. Please address your responses to:

Average Gift Size Study Project
INDEPENDENT SECTOR
1828 L Street, N.W.
Washington, D.C. 20036
(202) 659-4007

Organizations interested in providing data should contact us as soon as possible.

*Paragraph 94, Statement of Position 78-10, AICPA.

---

THE SLIDING SCALE VS. THE SINGLE PERCENTAGE FOR EVALUATION OF FUND-RAISING COSTS

Charts A and B are presented both as tables (charts A-1 and B-1) and as graphs (charts A-2 and B-2) for better understanding.

CHART A--Single Percentage (e.g., 30%): Can a single percentage be appropriate for both $1 million gifts and $1 gifts?

CHART A-1
Size of gift	Acceptable Fund-Raising Performance
	Fund-Raising Cost Percentage (up to)	Cost Per Gift (up to)
$1 million	30%	$300,000.00
100,000	30	30,000.00
10,000	30	3,000.00
1,000	30	300.00
100	30	30.00
10	30	3.00
1	30	0.30

Chart A-2
$1M. G i f t S i z e $1	Acceptable Fund-raising Performance	Unacceptable Fund-raising Performance
	0%.............30%	........................................100%
Fund-raising Cost Percentage

CHART B--Sliding Scale Percentages (0% to 100%): Does this make more sense?

CHART B-1
Size of Gift	Acceptable Fund-raising Performance
	Fund-raising Cost Percentage (up to)	Cost Per Gift (up to)
$1 Million	1 %	$10,000.00
100,000	.	.
10,000	.	.
1000	.	.
100	.	.
10	.	.
1	100	1.00

Few people would disagree with the idea that a single fund-raising performance percentage cannot be appropriate for both $1 million gifts and $1.00 gifts (see charts A-1 and A-2 above). If the lowest cost per gift is near $1.31, good fund-raising performance for $1.00 gifts would theoretically have to be 100% fund-raising costs. Further, it is likely that an acceptable percentage for $1 million gifts is near 0%. Therefore, it would appear that a sliding scale fund-raising performance table or curve ranging from near 0% fund-raising cost for $1 million gifts to 100% for $1.00 gifts might make sense (charts B-1 and B-2 above). According to chart B, good fund-raising performance can exceed 50% fund-raising costs all the way to 100%. However, it might be reasonable to assume that at some point an organization's fund-raising cost percentages would be unacceptable--regardless of fund-raising performance (see chart C below).

This raises yet another basic question: should the voluntary sector, including the charity oversight mechanisms and the charitable organizations they oversee, develop on a consensus basis a generally accepted "industry'' standard for some maximum fund-raising cost percentage which charities should not exceed year after year?

Unacceptable percentage, if fund-raising cost percentage is over x% regardless of acceptable fund-raising performance.

Return to Fundraising Productivity Series Home Page