Archive for January, 2011

Does project expense equal programme expense?

Thursday, January 27th, 2011

Many participants from our work shop ”The NGO Director´s Guide to Finance“ remember the functional-natural matrix.

The functional-natural matrix is a beast with rows and columns. Columns can be used to show different functional categories, like different programmes, plus the classical categories “fundraising“ and “administrative overhead“. Programmes are concepts that are initiated by management or the Board, with funds dedicated to certain aims. A programme is like a plan. The rows can be used for different natural classifications, like “rent“, “salaries“, “transport“ and others. I usually prefer the functional classification in the columns, and the natural classification in the rows, because there could be many natural categories, but usually, with many medium-sized NPOs, there are not more than two or three programmes.

As already mentioned, programmes are initiated by management or the Board. Thus, usually, the concept is different from the word “project“. The term “project“ is generally rather used for donor-defined plans. However, this definition is not very strict. There is no overall standard for how to use the terms ”programme“ and “project“. For the moment, it is only important to remember, that the functional-natural matrix deals with a classification of expenses, which was intended by management or Board.

How to prepare this beast called “functional natural matrix”?

Okay, here is the recipe:

First you need an accounting policy that defines at least YOUR approach towards defining fundraising and administrative expenses. For this a general wording could be used, for instance:

„In case of mixed expenses, management will provide an estimate how costs should be allocated to the different categories. The segregation will be disclosed in writing (file: “accountant´s memos“). The segregation of costs to functional categories is also indicated on the original source documents and authorized by the programme manager and the Chief Accountant“…,

 or likewise. Of course, you should take care that your approach is consistent, and in case your national standards already set a standard, which is also comparable with other organisations! The US GAAP provisions, for instance, provide excellent guidance for NPOs on how administrative costs could be separated from administrative costs. The IFRS are silent on this topic: there is no special guidance for NPOs in IFRS. However, IAS 1 determines that entities classifying expenses by function shall disclose additional information on the nature of expenses (including employee benefits expenses, depreciation and amortization expense).

You will also need a good definition of your natural classification.  This should not be so difficult. Depending on your needs, you may feel the need to create sub-brackets for salaries, communication, transport, rent, office supply etc.

Second, you need accounting software that allows to add tag numbers to your expense register. I personally like a maximum of three levels (for instance: function, nature, and, if really necessary, one more category, like: donor, geography, or other).

Then, on the basis of your accounting register, you can issue the functional-natural matrix. Here is a simplified example:



Very nice! You can read the matrix by rows and columns. You can say, for instance, that 54.5 percent of the vehicle costs (USD 15,000) occurred in programme 3.  You could analyze and compare the programmes to each other. You have an overview on the funds invested in fundraising. Yes, this is what you call “decision usefulness“, the Board will say. And they will be glad with the beast.

But then, the donors will take the functional-natural matrix, and they will turn it left and turn it right and then upside down, and they will ask: “Where are our funds?“

Well, this difficult to say: because programme accounting is not project accounting. For instance, it might be your policy to allocate certain depreciation expenses to a programme, while the donor does not accept depreciation as project expense. In other words, the way how the donor counts, and the way how you count might be different. The project expenses could be “somewhere“ in the functional natural matrix: a portion in Programme A, a portion in Programme B, and, if donors are nice, also a portion in fundraising and in administrative costs.

In other words, donor accounting is only derived from your general accounts. The best solution would be, of course, if you could “sell“ your wonderful functional-natural matrix to potential donors with the words: “Do you want to take part in our programme Okapi Renaturalization 2011? Do you want to take a portion of our general administrative expenses?“ In this case, if donors agree on your way of accounting, you can minimize administrative costs. Your accounting will be their accounting!

However, the world is not always made for NGOs with a good accounting system. It might be that the wishes of the donor prevail. In this case, you have to double the procedures. First, you count for yourself, then you count for the donor.

 Is it necessary to integrate donor accounting into your chart of accounts and your electronic accounting system? Phew! This is a hard question. Of course, it would be very convenient if the software already produced the project report. However, in many cases, donor requirements are so different from generally accepted accounting principles that it is virtually impossible to integrate these requirements into a consistent accounting system.

Thus, the project report is usually a handmade thing. Of course, it could be helpful, if your expense register already contains a tag number that informs you on the project (if you go back a few paragraphs: This could be level three, in the systematic suggested above). In this case, if you filter the expense register for this project, the bulk of the work will be already done.

Then, in a second step, probably, the purchase of cars has to be added, or depreciation has to be taken out, another exchange rate has to be used, or the cash basis instead of the accrual basis of accounting has to be used. You can see from this that I prefer a semi-automated process.

By experience, it is very helpful to register first all expenses as expenses, i.e. as decreases of unrestricted net assets, and then later, in a second step, on the basis of the project reports, a reclassification is introduced. At first glance, this technique may seem strange to many accountants. However, the technique makes sense, as you will see in a few moments. The idea is that project expenses will appear as a reclassification of unrestricted net assets on the basis of the project reports to the donor.



1.      At the end of the accounting period, all expense registers are closed onto unrestricted net assets:

Debit decreases of unrestricted net assets (statement of activities

Credit Salary expenses, rent expenses, office supply expenses etc.

2.      A project report is prepared for the donor. With this report, the organisation claims that the following expenses would reduce the amount of restricted funds (as lines in the statement of activities)

Debit  temporarily restricted net assets of Project A123  

Credit unrestricted net assets – reclassification.

The underlying document for this second booking (and disclosure in the statement of activities) is the project report. The result of this booking would be that temporarily restricted net assets are reduced (i.e. a project expense is disclosed in the statement of activities), while the unrestricted assets are once again increased)a reclassification is disclosed in the statement of activities).

If you have correctly allocated all project expenses with your tag number, you can even determine a gain or loss per project! – and, of course, this gain or loss would be expressed as an increase or decrease in unrestricted net assets. Well, you could say that this difference is not really a gain or loss, but either a contribution to the administrative overhead, or a deficit that was sponsored by unrestricted funds.  

In the example above (in the matrix), it could be that one donor finances Programme B, and this donor allows that a car park is bought from his funds. Thus, in project accounting the cars are registered as a project expense with a value of, let us say, USD 500,000; while in project accounting only the straight-line portion of USD 100,000 (straight–line depreciation over five years) is mentioned.

Thus, you can deduce the project accounting from your general accounts:


Programme 2 total expenses: USD 135,000

Plus capitalized expenses (car purchase): USD 500,000

Minus depreciation costs: (USD 100,000)

Eligible project cost: = USD 535,000

Difference between programme expense and eligible project expense: USD 400,000

USD 400,000 is the “gain“ from the project: The organisation has “earned” an increase of USD 400,000 in unrestricted net assets. You also can see in this example, that the reclassification (USD 535,000) is sometimes higher than the true expenses.

This is, in short, the difference between the programme perspective and the project perspective.

Please use the “comment” function in this blog, in order to submit your comments.


Frank Fabel, CPA


Tuesday, January 4th, 2011

Accounting for administrative overhead:

Charging a percentage or allocating specific costs?

Many donor organisations and many donee organisations with multi-donor agreements suffer from this problem: the recipient organisation receives funds from different donors; all donors allow to pay a portion of the salary of the accountant – but then quarrels start of who pays how much. Let us listen in on the following year-end talk between the director and chief accountant in the NGO “Save the Okapi”:


Director: “How is it, cooking the books, maitre?”

Chief Accountant: “Of course, we are not cooking the books. We are only using all possible advantages for the sake of the organisation and the mission!”

Director: “Which means?”

Chief Accountant: “Let us take my salary: Okapi without Frontiers in Paris is prepared to pay
20 percent of my salary. The Okapi Salvation League from Seattle also pays USD 500 of my salary and requires that this should be equivalent to 20 percent of my work time. The World Okapi Fund agreed to pay two fifths. And the Okapi Therapie Zentrum from Germany pays another fifty percent. Altogether, this makes up 130 percent. I am prepared to work overtime, thus, I think it would be ethically correct that everybody pays their portion as agreed. I cannot see that we are obliged to disclose this excessive inflow of funds to our donors. Thus, we will earn 30 percent of my salary as unrestricted funds. The Board could be proud of me. By the way, what about a salary increase?”

Director: “We can talk about this after the audit. I fear that auditors may mention this in the audit report. I do not want to risk the relation to the donors. So far it has always been an honest and open relationship. Is there no other way how to allocate your salary to the donors?”

Chief Accountant: “Maybe we can issue one work contract for each donor. So, as a result, I will have four work contracts. Ain’t I a genius?”

Director: “Do you want to confuse me? You will still end up with 130 percent! No, this is no way out. How would it be if we take the total compensation of all donors as 100 percent, and everybody paid a portion. Can you calculate what this would amount to?”

Chief Accountant: “Yes, of course. If 130 percent were 100 percent, then 10 percent would be 13 percent and 0.77 percent would be approximately 1 percent. Thus, a donor who wishes to pay
20 percent, would in reality pay 15.4 percent of my salary. Understood?”

Director: “I’m afraid, none of the donors would understand me!”


The Director is right: A lot depends on communication with donors. And donors can get very angry if they do not understand which portion of the salary of the chief accountant they pay.

But how to allocate the portions correctly? The option with different work contracts is no real way out. In a sense, this makes things even worse. There are more papers, and it makes things more confusing. Maybe, simple-minded auditors or donors could be satisfied with this technique. But is this satisfactory for management? Certainly not, as this approach links the organisation very close to the donor. In fact, directors who act in this way are only donor-driven.

Also the technique to allocate the actual portion of the accountant’s salary to the donors is no practical option. This would require monthly recalculations of the portion, and the probability that the donor will understand this is close to zero.

But why not allocate an administrative overhead? It can be so simple: the programme costs could be linked to the respective donor in the usual way, with an unequivocal earmarking of the original documentation. Administrative costs would be charged as a fixed sum or percentage of a parameter that has to be agreed. For instance, the donor could pay 10 percent of the expensed project costs in order to cover the administrative overhead. How these 10 percent are used would be the responsibility of the recipient organisation.

Or: The director could agree with the donor that a certain portion of the administrative expenses are borne by the donor. For instance: Okapis of the World contributes 30 percent of the turnover of the NGO; thus Okapis of the World should also contribute 30 percent of the administrative overhead.

Or: The NGO determines that in the last years, administrative expenses have amounted to
16.5 percent of the total turnover. Thus, the Director negotiates with all donors that they should at least pay a portion of 16.5 percent of the administrative overhead, without presenting invoices for this.

All these solutions have something in common:

The director is required to act! These regulations will not come into force as an action in the accounting department, but as a negotiation between donor and donee.

It goes without saying that this approach only works if there is a policy of how administrative expenses will be counted. This is not so easy: For instance, it could be that expenses are mixed programme and fundraising expenses, or mixed programme and administrative expenses. Here is a classical example: Save the Okapi transported some Okapi babies to the next zoological garden in order to save their lifes. In front of the cage in the zoo, there is also a coworker who explains the necessity to save the okapi and AT THE SAME TIME tries to recruit new members for the organisation. In other words, programme costs (saving okapi babies, awareness raising for okapi issues) and fundraising expenses are mixed. In these cases, it would be the task to estimate to which extent the expenses would be programme and to which extend fundraising expenses. It would be the task for auditors to understand and critical review the calculation.

Or: the chief accountant is at the same time a veterinarian, and 50 percent of his or her worktime is spent on emergency operations for okapis in the bush, and the other 50 percent he or she works in the office. And they keep a table on the time spent on emergency operations and the time spent on the bookkeeping operations. Later, auditors complained that there were too many blood traces in the books, and doctors complained that office scissors were used for operations…

Okay, this was an unrealistic example. But it could be that directors partly carry out administrative work and partly work for the programmes. In accordance with the internal estimate, costs would be allocated.

The problem here always is whether the donor accepted this specific policy. However, the fear that a donor would not agree on this rational approach is no argument against introducing a policy.

In short: everything is better than the stupid attempt to allocate administrative costs to grants in the same way as programme costs.

Programme costs could be allocated, as long as these expenses are clearly separated. For instance, it is better, if one person works in one programme, and not in three or four.

We would like to encourage directors of recipient organisations to agree on a certain portion for administrative expenses, which does not have to be supported by original documents. The mechanism could be freely negotiated. However, we think that a portion of the project turnover, based on the experience in the past, would be the most just approach. We also would like to encourage donor organisations not to glue to simplified systems. A sophisticated system of determining the percentage of administrative overhead is more efficient and more transparent in the long run.


Frank Fabel, CPA