A blog post by Lucius Caviola & Nadira Faulmüller, cross-posted from the Practical Ethics blog at Lucius' request.

Imagine a car company advertising as follows: “90c of any dollar you pay for your car goes directly to building cars. Only 10% of our expenses go into planning, designing, and advertising them.” Such a campaign strategy would seem patently bizarre; when buying a product few of us are interested in how much went into administration, all we care about is what we get for our money. Overhead ratio (the proportion of money going into administration) is irrelevant; only cost-effectiveness matters.

This common sense approach to purchasing goods or services does not seem to translate into the non-profit sector, however. Consider the following advertisement by the organisation CARE:  “More than 90 percent of our expended resources – among the highest of all philanthropic organisations – support our poverty-fighting projects around the world. Less than 10 percent of expended resources go toward administrative and fundraising costs.”

CARE campaign strategy is quite common for a non-profit. Many charities attempt to win donors by emphasising their low overhead ratio. Perhaps they are right to so do, as more than a third of US citizens believe that charities should spend much less on administration. Further, lab studies have shown that people tend to donate more to charities with low overhead ratios. Consequently, in order to fulfil donors’ expectations charities are forced to keep their overheads small as illustrated in Dan Pallotta’s TED talk.

Few donors are able to identify the irrelevance of overhead ratio to effective giving. The interventions of some charities are far more cost-effective than others, so that even if only a small percentage of donated money reaches the destination, the overall impact will be much more beneficial than a larger percentage of donations for a less effective intervention.

Indeed, charity evaluators have not found any correlation between overhead ratio and cost-effectiveness: Organisations with a low overhead ratio can still fail to be effective, and organisations with a high overhead ratio can be very effective – maybe precisely due to good administration, including competent staff, infrastructure, fundraising, and evaluation.

Given donor’s proven aversion towards overhead, charities could consider offering ‘overhead-free’ donations (by, for example, soliciting unrestricted funding from a few large donors to cover administration costs). A recent study published in Science has found that average donations can triple when donors are told that overhead costs are already covered. One explanation for this result is that people perceive their personal impact to be greater if all of their money reaches the destination.

These findings raise an important question: Why are people so preoccupied with overhead ratio in the first place? A study we have recently conducted suggests that they may be falling prey to an error called evaluability bias; people focus on overhead ratio because it is easier to evaluate than cost-effectiveness.

We found that when people are presented with information for a single charity (namely its overhead ratio and number of lives saved per dollar), on average charities with lower overhead ratios received more donations, irrespective of cost-effectiveness. However, when people are presented with information for two charities (allowing them to compare the cost-effectiveness of the two charities), they gave more to the charity with the greater cost-effectiveness, irrespective of overhead ratio. These findings suggest that people do care about saving lives, not administration costs, but that unless the information needed to evaluate cost-effectiveness is readily available, they will fall back on the easier-to-evaluate overhead ratio.

So although people are biased about what matters in the non-profit sector, the good news is that appropriate information about cost-effectiveness can go a long way to correcting this bias. We must try to better inform potential donors about the irrelevance of overhead ratio when selecting a charitable cause, and work to improve their decisional architecture to make it easier for them to realise the importance of cost-effectiveness. Shifting resources to the most cost-effective charities is crucial; a lot of individuals’ lives depend on it.

10

0
0

Reactions

0
0

More posts like this

Comments4
Sorted by Click to highlight new comments since: Today at 7:53 PM

I think the "overhead ratio" heuristic is moderately rational because

  1. There are some real scam charities that skim a lot off the top. (Not quite a charity, but a recent example is the claim that Sarah Palin's PAC only uses 3% of its funds for its intended purpose.)

  2. If you have limited time and are only donating small amounts, it's potentially not worth the effort to look up detailed information, so evaluating based on what's available may be better than nothing. (At least you reduce your chance of donating to a scam that way.)

  3. The work of many charities is hard to quantify. Development aid is probably among the easiest, but how do you quantify MIRI's cost-effectiveness and compare it with other charities in its league? If two charities seem to both do good work per employee, but one charity spends more of its funds on its core employees, that charity is the better deal.

I think 1. and 2. are partially true but 3. seems to be the result of fuzzy and faulty thinking to me.

The work of many charities is hard to quantify. Development aid is probably among the easiest, but how do you quantify MIRI's cost-effectiveness and compare it with other charities in its league? If two charities seem to both do good work per employee, but one charity spends more of its funds on its core employees, that charity is the better deal.

Yes, if you're concerned about appearances, but not if you're concerned about expected impact. High variance does not imply that the answer is indeterminate or that the amount of good work per employee is equal.

If you have limited time and are only donating small amounts, it's potentially not worth the effort to look up detailed information, so evaluating based on what's available may be better than nothing. (At least you reduce your chance of donating to a scam that way.)

It is worth than nothing. Donor emphasis on overhead ratios causes charities to spend less on overhead than they consider optimal. So if anything one should assume charities with higher overheads to be less willing to yield to this pressure, and hence more cost-effective (at least as long overhead expenses do not exceed certain upper bounds beyond which one could reasonably suspect corruption or incompetence).

I don't think this argument works, unfortunately. (I know that in a separate post you said you were only wondering about/floating it.) The reason is that the amount that charities would spend on overhead, were it not for donor emphasis on this, isn't in fact what "they consider optimal" for altruistic purposes, because they're not only concerned with altruistic purposes. To put it another way, in your separate post you said that "charities are better judges of how their own resources should be allocated than is the public". And it's plausibly true they're better judges of how to spend resources for altruistic purposes. But again, these aren't their only purposes.

I've worked for many non-profits and in my experience most are primarily concerned with the self-interest of their employees, which the donating public isn't. So in that respect the donating public makes better decisions.

Serving the self-interest of employees is where a lot of overhead spending goes, including through larger salaries than are justified by the good that those salaries do, among other things. Salaries are typically a large part of overhead. This connects to what Brian's said. I agree with him that higher salaries are some evidence of the employees being less dedicated and more self-interested, all else being equal. Other factors are important too of course, including how much they focus on fun or prestigious work.