## High Pay for Nonprofit Execs? Analysis of 100,000 Salaries

Everybody has an opinion about whether nonprofit executives are paid too little or too much, but almost nobody has any real data outside their own experiences. At last! Economist Linda Lampkin (left) analyzed 100,000 nonprofit CEO salaries and has a definitive (if a little statistics-wonky) answer:

Sometimes it seems that the whole credibility of the charitable sector hinges on the issue of compensation. So what exactly is too much pay? And who is getting it? A colleague and I decided to analyze Form 990 compensation data to bring some real numbers to the discussion (database info at the end of this article).

The IRS says nonprofits must pay "reasonable compensation" but there are few specific guidelines other than that total compensation must be compared with "what ordinarily would be paid for like services by like enterprises under like circumstances." "Like enterprises" means organizations that provide similar types of services and have similarly sized budgets. Using the ERI Economic Research Institute database of compensation data from Form 990s in 2009 (the most recent full set available), we calculated the mean and median salaries for almost 100,000 CEOs of charities, by revenue size:

Analysis of 100,000 nonprofit CEO salaries

[1] Mean is the mathematical average, calculated by adding up the data points and then dividing by the number of data points.

[2] Median is the midpoint of the data, so the number of data points having values greater than or equal to the median is the same as the number less than or equal to it.

[3] Standard deviation, a measure of the variability of the data, is the average amount by which individual data points in a data set differ from the arithmetic mean of all the data in the set.

Although the amount of the Standard Deviation (SD) from the mean salaries above shows that there are some charities paying much higher than the average, there are also some paying much lower. And look at the levels of the mean and median salaries -- less than \$50,000 for CEOs in the smallest group (the group with the largest number of charities), less than \$70,000 in the next. In fact, the vast majority of CEOs are earning well less than \$100,000 - not much evidence of lots of very high salaries.

But at the highest level of revenue -- nonprofits with annual revenue of \$10 million or more -- the large difference between the mean and the median salaries and the significant standard deviation indicate a wide range of salaries within this group.

So we decided to take a closer look at the organizations by service type with revenue of \$1 million or more. The chart below says, for instance, that among arts organizations with revenue between \$1 million and \$5 million, only 1.3% pay their executives more than 2 standard deviations away from the mean. Restated: Only 1.3% of arts organizations of this size pay significantly more than average.

These percentages are translated to actual number of organizations reporting high compensation for their CEOs in the table below.

Out of nearly 100,000 charities reporting CEO compensation in 2009, fewer than 1,200 paid their CEOs more than 2 SDs above the mean salary. In other words, only around 1% of nonprofit CEOs make a salary significantly higher than the average.

The analysis of the Form 990 CEO compensation data from 2009 showed:

Finding #1: A great many nonprofits filing Form 990 had less than \$100,000 in revenue and a majority of these had NO paid staff. And even among charities with more than \$1 million in revenue (about 14% of all nonprofits), only 60% have a paid CEO. [Ironic question: does \$0 per year seem unreasonably high?]

Finding #2: The highest paid CEOs -- defined as paid more than 2 SDs than the average salary - represent only about 1/2 of 1% of all nonprofit organizations.

Finding #3: Higher than expected compensation is not evenly distributed by type and size of the charity. Hospitals have the greatest number of CEOs paid higher than 2 SDs above the mean, at all revenue sizes. The categories of health (excluding hospitals) and public and societal benefit organizations also had more CEOs falling outside the 2 SD range than all charities.

Setting nonprofit executive pay

Whatever compensation decisions are made, nonprofits need to follow best practices by collecting appropriate data and using them to make defensible decisions that reflect the organization's compensation philosophy. Nonprofit resources need to be spent right - for instance, providing enough compensation to attract, retain, and motivate the staff, while leaving sufficient dollars to achieve the organizational mission. That becomes even more challenging in difficult economic times.

So the next time you hear that discussion about outrageously high nonprofit salaries, add some context to the debate with these numbers -- there are some high paying organizations, but not that many. And use the ERI data as a resource to document that your charity's compensation is on target.

Appendix: The ERI database and how you can use it

The ERI database of Form 990 information is available in the Nonprofit Comparables Assessor (CA), a software program that calculates average competitive compensation levels, based on organization revenue size, type of organization, and geographic location. You choose the type of organizations providing similar services (adoption services, day care, museum, etc.), the revenue level (\$1 million, \$50 million, \$2 billion, etc.), and a specific geographic region (all US or a state). The majority of nonprofits can use a free version of CA to provide the data needed to document reasonable pay. The more comprehensive paid subscription versions are designed for larger organizations, watchdog groups, and regulators, including the IRS and state charity officials.

Linda Lampkin is former Director of the National Center for Charitable Statistics at the Urban Institute where she was instrumental in improving Form 990 and in developing databases that make 990 data useful to researchers and practitioners. She co-authored the New Nonprofit Almanac and numerous research papers on the nonprofit sector. After a 30-year career as an economist working for various nonprofits and labor unions, she now heads up the Washington, D.C., office of Economic Research Institute, where she "gets paid to write blogs about the nonprofit sector and play with compensation data." She can't believe her good fortune in finding this job.

Reach her at linda.lampkin@erieri.com or 202-306-7333. The full study on which this study is based can be found at https://www.erieri.com/PDF/CharityExecutivePay.pdf.

• Anonymous

It would be very interesting to know how much the mean and medium compensation of EDs has gone up since 1998, when the IRS first acknowledged that it was appropriate for a nonprofit to look at for-profit 'comparables' in setting salaries. Executive salaries generally (i.e., in for-profit businesses) have been creeping higher as Boards look at benchmarks and then decide that their CEO is 'above average' (all of them come from Lake Wobegon, apparently). I would also quibble with the characterization of an ED's salary as "high" if it is more than 2 SDs above the mean. The SD numbers are big -- signs of a wide range of data points -- so allowing a 2-SD 'reasonable band' seems too generous.

Jun 12, 2012
• Hi, Anonymous - this is Linda and I appreciate your interest in the research. Research on nonprofit comp has been hampered by the lack of usable data (we receive the Form 990 info as pdfs and actually have to type the comp data to create a database - lots of time and expense required) - but I think that research on changes in comp over time would be interesting, too. I also think that for-profit executive salaries have been doing more than creeping higher - ERI does a quarterly review of executive salaries and analyzes trend and that seems to be an entirely different world, with annual average comp of over \$17m!. See http://www.erieri.com/index.cfm?FuseAction=NewsRoom.Dsp_Release&PressRel... The choice of 2-SDs as the measure of high comp was made to try to provide an easily understood indication of the number of nonprofits that might be considered as providing comp that was higher than typical - a group that should be researched more. There is no hard and fast rule for what is "unreasonable" comp and there may be explanations that show what appears high nay in fact be reasonable. The width of what is "reasonable" certainly could be adjusted depending on the exact purpose of the analysis. This report was just meant to indicate that for the entire group of Form 990 filers, there were not that many high-paying organizations.

Jun 13, 2012
• Anonymous

It has been a while since I took Statistical Analysis as an undergrad, but I am confused by the standard deviation figures. The explanation (foot noted) is accurate, but the actual numbers don't make sense. Is the reported standard deviation number the range for one standard deviation below and one above the median? Sorry if this is an ignorant question, but I can see the deviation curve in my mind, and these numbers are not making much sense to me.

Jun 12, 2012
• Anonymous

I think it would be interesting (and perhaps telling) to also analyze executive salaries as a percentage of the overall budget.

Jun 12, 2012
• Anonymous

I agree that it would be interesting to look at salaries as a percentage of the overall budget. In fact, I've been looking, unsuccessfully and for purposes of comparison, for information on the percent of overall budgets of nonprofits devoted to the whole compensation package. Also would like to find information on the percent of nonprofit compensation packages made up of benefits.

Jun 11, 2014
• Anonymous

This is a very interesting area for research that would be very useful for non-profit boards as they look to understand how their overall executive team compensation as a % of net patient revenue compares to their peers. I've come across very little useful research in this area. I'd love to know if you find anything.

Jul 23, 2014
• Anonymous

The \$100-500K is a bit disturbing for Salaries(needs to be broken down a bit further for true figures). Too many gaps from a \$100k to a \$500k organization to determine a Salary Range hope you agree

Jun 12, 2012
• Anonymous

Even if we were all paid exactly the same within our budget range, I think some people would say we are paid "too much" overall. It seems to me that an equally valid measure (though harder to quantify) is a comparison between for-profit and not-for-profit organizations and their CEOs. The final section ("Setting Nonprofit Executive Pay") seems right on target in this regard - pay (within reason) what it takes to get good people and achieve your mission. Value is not quantified only in dollars, it's a balance of quality, demand, and price.

Jun 12, 2012
• Salaries don't take money away from the mission. Paying adequate salaries improves the ability to fulfill the mission. Staff should not be shy about asking for higher pay and Boards should not get away cheaply by trying to guilt-trip committed people into working for less. David M. Patt, CAE

Jun 13, 2012
• Great data. I think the point about the vast majority of nonprofits having revenues under \$100K is particularly important, especially given the trend to want to compare nonprofits with for-profits, because the vast majority of small businesses in America look very similar to the vast majority of nonprofits- they also have no paid staff (i.e, earning a salary). We would do better to compare our org's to the local home day care or family-run dry cleaner instead of Facebook or Kodak. Across the board, whether you are a for-profit or non-profit small business with little to no staff, you do not have much of a public voice or political influence, but it's the majority.

Jun 13, 2012
• Anonymous

Dear LInda, Thanks for the useful info and hard work. One question, does your revenue include non-non cash revenue? We have 2.5 million in cash revenue and another 9 million in donated revenue. Just wondering if that puts us in the 10m+ or the 1m to 5m range. Thanks.

Jun 14, 2012
• Anonymous

admin wrote an interesting post today. Here's a quick excerpt, but since this is the Not-For-Profit Accounting blog devoted nonprofit baccounting/b issues I'll assume my readers know all about 990s. But there may be folks new to the sector out there, maybe you know a new board member, so please pass

Aug 17, 2012
• Anonymous

I agree the 990EZ is not that tough to do. If you don't get it filed timely you will have pintlaees and may even find your tax exempt status revoked if you are substantially late. Just get it done.

Aug 17, 2012
• Whatever compensation decisions are made, nonprofits need to follow best practices by collecting appropriate data and using them to make defensible decisions that reflect the organization's compensation philosophy.Staff should not be shy about asking for higher pay and Boards should not get away cheaply by trying to guilt-trip committed people into working for less. David M. Patt, CAE

Dec 21, 2012
• Anonymous

I work for a nonprofit organization that pays its ED \$100,000 a year when we only generate around \$450,000 in funding/donations/revenue. Tell me how that's not taking advantage of the children we serve.

May 14, 2013
• High Pay for Nonprofit Execs? Analysis of 100,000 Salaries

has this been updated since the original posting?

Feb 13, 2017
• This article was writteni n 2012, so the statistics listed are what was availble then. You may access updated information at the Economic Research Institute's website at www.erieri.com.

Feb 13, 2017
• What about benefits, there is a salary but what if there are no medical, vacation, paid sick time, no health or life insurance etc. Are your studies including or excluding these items?

Jun 15, 2017

### Filtered HTML

• Web page addresses and e-mail addresses turn into links automatically.
• Allowed HTML tags: <a> <em> <strong> <small> <sup> <sub> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd> <h1> <h2> <h3> <h4> <img> <br> <br/> <p> <div> <span> <b> <i> <pre> <img> <u><strike>
• Lines and paragraphs break automatically.

### Plain text

• No HTML tags allowed.
• Web page addresses and e-mail addresses turn into links automatically.
• Lines and paragraphs break automatically.