Stop looking at admin expense ratios!

In this post, I'm going to argue against using admin expense ratios as a mechanism to assess charities. This is a point which has been made before, however in this article I will also investigate a topic with relatively little data:
- could the admin expense ratio be useful in some particularly extreme cases? I would have thought so, however I did some further investigation, and I actually think that even in this unusual case it seems unhelpful.
The key point here is that the reasons why admin expense ratios are bad are well understood, yet the ratio is still disappointingly widely referred to, both by donors and (worst of all) charities themselves, who harm the sector by perpetuating the usage of this inappropriate measure.

So here's a list of the ills of this metric:

It's a poor measure of impact
You want to choose which charity to give to on the basis of its impact - i.e. based on how much good it does. The amount it spends on administration is obviously a different thing. Please check out this page from Charity Science which illustrates the idea nicely.

It encourages "creative accounting"
There is actually no really firm definitions of what is and isn't administration. If you buy a desk or computer for staff in your charity's counselling programme, would you call that administration cost, or would you say that it's an essential part of the counselling programme and should be accounted for there? When you paid some consultants to help you to expand your alcohol and drug addiction service to a new town, was that admin cost, or was that part of the alcohol and drug programme? What if the consultants were helping with growth of the whole organisation instead? (note that I'm familiar with UK charity sector accounting - there may be jurisdictions where admin costs *are* precisely defined) Also you could argue that a CEO's time is necessary for the effective management of each programme, and allocate his/her salary to each programme. This point is also made in this article.

Charities can decide to be savvy and make their admin costs look as low as possible, or they can try to maintain their integrity (if they even have time to think about the topic). Trying to maintain your integrity is unlikely to reap many rewards, especially since there are many ways of twisting the results to your favour. Ultimately, this lack of consistency is one of the reasons why the measure is meaningless.

It discourages investment in more efficient administration expenditure
Employing an extra admin worker can often be a huge improvement in efficiency - without that admin worker the admin work doesn't somehow go away! So instead of getting an admin employee to do it, you pay someone with other specialist skills to do it instead. If you take someone with a specialist background in nursing/supporting children/policy/whatever your charity does and get them to do admin, you've found a great way to run your charity inefficiently. Not only is a poor use of money, but it's a good way to demotivate your staff and worsen staff turnover.

It discourages longer term sustainability (in some cases)
Certain organisations do well under the admin expense ratio measure. This is the organisation set up by someone with a desire to do good, and who is independently wealthy enough that they can put the time into running the organisation without needing to be paid. This is actually quite a good thing in the short term, it does essentially mean that the donor's money goes (somewhat) further because the charity leverages volunteer time. So I can give the ratio a little bit of credit, but actually there are better ratios to reflect this.

Ultimately this is not good for the longer term sustainability of the charity though; over time, if it wants to grow/survive, it will need to professionalise, because it can't rely on that sort of goodwill forever. When that founder wants to retire or stop he or she may need to be replaced by a staff member. This may well be the best thing, but the admin expense ratio would suffer.

Surprisingly, even *extremely* high admin expense ratios may not tell us much
It is claimed that a high admin expense ratio is an indicator of inefficiency. If you're looking at sector-wide averages, this is intuitively false, since greater investment in efficiency should make better returns and Caroline Fiennes sets out reasons and some data to support this in her book (you can see this in chapter 2 here).

However I would have guessed that where the admin expense ratio is extremely high, it may tell us something useful. For example, Givewell founder Elie Hassenfeld commented that Charity Navigator is useful for identifying fraudulent charities, but otherwise useless, and I imagined this comment may carry across to the admin expense ratio. So I looked into it, and found little evidence to support this; that said, I may not have done sufficiently rigorous research, so if anyone knows better, please let me know. If you want to know what I did, have a look at "Further comments about whether there are any merits to the admin expense ratio" below.

Conclusions
Ultimately, I wouldn't be sad about the excessive emphasis on admin expense ratios if there was ever any real evidence that admin expenses are (or ever have been) excessive. I have seen no such evidence. By contrast, my real life experience of the charity sector has shown lots of cases where charity sector staff are loath to suggest ideas like funding their training or development because they don't want to take the charity's precious resources away from the cause they care about so much, despite the clear benefits of such investments. So donors, please stop looking at this.

And charities, if you are looking for the right way to address this issue, you can't refuse to share the info - you have to be transparent. War child, for example, strikes (I think) the right balance between transparency and challenging the irrational question-asking. However, don't do what I've seen many charities do, and trumpet your low overhead ratio. You're just making things worse.



---------------------------------------------------------------------------------------------------------
Appendix: Further comments about whether there are any merits to the admin expense ratio

One more note, I referred above to what Caroline Fiennes said in her book. In it, she claimed that on average charities with higher admin expense ratios are more effective. It may be interesting to note that nfp synergy challenge this here, arguing that her data do not have a large enough sample size to be robust. Ultimately though, they too agree (who wouldn't?!) that the admin expense ratio is a poor measure.

To test my hypothesis that the admin expense ratio would be useful at the extremes, I took a glance through this list of the charities with highest admin expense ratios on Charity Navigator, thinking that where admin expense ratios are really high, there would genuinely be signs of inefficiency. This was not obviously the case, with one case being obviously an example where there were one-off admin costs from a move of office, and other cases mostly not being analysed enough to know whether they are inefficient or not. For example, the public policy research organisation Peter G Peterson Institute for International Economics was criticised for the amount spent running the organisation rather than furthering its goals, however a research organisation strikes me as exactly the sort of organisation where lots of costs could be classified as admin, but are really essential to running the organisation - after all it's not the sort of organisation that gets out in the field and deals directly with beneficiaries, so it's not immediately intuitive to me which of the costs will be admin and which not. For example is a database manager's salary admin? sounds like it, but then data is probably important to research work; and this comes to the crux of the problem, all the work should be important. Similarly, when I looked at other charities on this list I mostly struggled to see that genuine inefficiency was occurring.

Note that the website I was looking at was old, so perhaps I would have been better off if I had looked at more recent data direct from Charity Navigator, however as far as I can tell, there is (unfortunately) no way of doing a search on CN for the charities ranked according to their admin expense ratios, starting with the highest.


---------------------------------------------------------------------------------------------------------

---------------------------------------------------------------------------------------------------------

Further reading:

http://www.uncharitable.net/
http://overheadmyth.com/
https://philanthropy.com/article/Overhead-Costs-Pose-Dilemma/154811
http://www.huffingtonpost.com/paul-brest/administrative-costs-and_b_162505.html?
http://freakonomics.com/2011/06/09/why-ranking-charities-by-administrative-expenses-is-a-bad-idea/
https://evaluatingkatine.wordpress.com/2009/09/21/the-worst-question-to-ask-about-charity/
http://www.charityscience.com/blog/how-charities-are-actually-wasting-your-money-video-and-infographic
http://www.ssireview.org/articles/entry/the_nonprofit_starvation_cycle/
http://ssir.org/articles/entry/overhead_costs_the_obsession_must_stop
http://www.theguardian.com/voluntary-sector-network/2013/may/02/good-charities-admin-costs-research
http://www.amazon.co.uk/Aint-What-You-Give-That/dp/0957163304/ref=sr_1_1?ie=UTF8&qid=1390910624&sr=8-1&keywords=it+ain%27t+what+you+give - click on "look inside" and go to chapter 2
https://www.warchild.org.uk/about/our-finances/how-much-do-we-spend-on-running-costs
http://www.huffingtonpost.ca/marina-glogovac/indicator-of-a-charitys-effectiveness_b_8147516.html
https://www.bostonglobe.com/magazine/2015/12/09/you-been-asking-charities-wrong-question/5kJmeri8i7SXxCPgnMtluL/story.html