US Charity | Culture & Capital

Trends in US Charity & Nonprofits

US GDP per capita doubled between 1980 and 2021 (in 2015 chained dollars), World Bank
US Household Expenditure per Capita tripled between 1984 and 2020, CEIC
US Charitable giving tripled between 1980 and 2020 (in CPI-adjusted dollars), Giving USA

Over the last 40 years, the American body politic grew wealthy, hyper-consumeristic, and ever more generous. Let's explore the data to hone in on trends in US charity and nonprofits.

US Charity and Nonprofits

August 28, 2022

In 2021, the Urban Institute asked 2 research questions:

What recent donation trends have 501(c)(3) nonprofit organizations experienced?
1. How have those trends varied across organization and community characteristics?
What are the differing impacts on and implications for nonprofits of donation trends?

I modified those questions to:

What are the most pronounced trends in charitable giving between 1980 and 2020?
What does IRS data tell us about the state of the nonprofit sector?

Assertion

Decreased church attendance, diffuse charitable giving online, and a proliferate nonprofit sector contribute to religious organizations’ shrinking share of charitable giving and secular nonprofits’ expanding share of charitable giving.

Key Takeaways

Between 1980 and 2020, religious organizations' share of total charitable giving decreased from 53% to 28%.

Between 1980 and 2020, secular special interest organizations' share of total charitable giving increased from 42% to 72%.

Both decreased church attendance and an increased secular nonprofit population contributed to the decentralization of charitable giving away from religious organizations and toward secular nonprofits/private foundations.

As of August 8, 2022, the IRS accounts for 1,795,106 tax exempt organizations, of which 547,625 possess assets in excess of $25,000. Breaking it down further: 139,160 nonprofits possess assets in excess of $1,000,000, and of those, 9,847 possess assets in excess of $50,000,000. That said, a staggering 1,145,659 nonprofits report no assets at all.

Research Story

Data courtesy of Indiana University Lilly Family School of Philanthropy and Giving USA Foundation shows private foundations’ share of total giving increased from 6% in 1980 to 19% in 2020. In contrast, giving from individuals decreased from 84% in 1980 to 71% in 2020. That’s a 13% one-to-one swap between the two categories of charitable giving.

PDF Version

From 1980 to 2020, religious organizations received a significantly reduced share of charitable giving. Likewise, private foundations' share of charitable donations received increased proportionate to charitable donations given. In fact, the whole secular nonprofit sector accrued an increased share of charitable donations by 71%. In contrast, religious organizations' share of charitable donations fell 44%.

Despite an 87% increase in total donations received (CPI-inflation adjusted), religious organizations’ share of charitable giving decreased by 22 points. This is the most dramatic trend in charitable giving in the 40 year period. Religious organizations' share of total charitable giving reached a high of 53% in 1985, but fell to an all time low of 28% in 2020.

Receiving as a Share of Total, 1980-2020 snap.png

This trend correlates with a Gallop report that American church, mosque, and synagogue attendance steadily declined from 68% in 1980 to 47% in 2020 (as a percentage of the U.S. population).

What besides diminished church attendance explains this trend?

Charity is far more decentralized in the 21st century. For instance, in 2021 the online fundraising platform GoFundMe served as the medium of reportedly $15 billion worth of donations. The ALS Association reported the 2014 “ice bucket challenge” raised over $115 million thanks in large part to its virality via social media platforms. In the 21st century, online charity campaigns can be wildly successful if they go viral.

More generally, the online medium of exchange facilitates giving among like-interested individuals regardless of residence. The distance once overcome by writing a check, buying postage, then traveling to a mailbox, is now completed by a simple succession of clicks from a phone or computer. Not only is it easier to donate, but people are more likely to be exposed to large scale foundation-led charitable campaigns via social network platforms. This could be one factor explaining the trend in giving to foundations versus religious organizations.

The behavioral economist Richard Thaler’s nudge theory applies in two ways. First, the social benefit of charitable giving via social media “posts” encourages participation among otherwise reluctant or apathetic persons. Second, the convenience of participation accompanied by instantaneous gratification encourages participation. In all its convenience, the means to give via the internet overcomes mankind's default setting of indifference.

As a matter of DEFAULT settings, I would recommend corporations establish a default payroll setting whereby employees give 2% of their income to a charity of their choice (or an apolitical humanitarian organization renown for good uncorrupted work). The employee can opt out of course. I believe the favorable outcome associated with the default setting in retirement planning would also be found in this policy. More on this recommendation in my upcoming project (expected completion October 1, 2022).

Pew Research found online social media usage rapidly climbed from 5% in 2005 to 48% in 2010, reaching 72% in 2020. While having social media does not equate to use, additional Pew research confirmed a high frequency of use among those who have it. In fact, a 2021 survey found 70% of surveyed Facebook users visit the site daily and 17% visit weekly- the grand majority use the site frequently.

This online ecosystem of social media exchange includes charitable giving. According to The Nonprofit Times, Facebook served as the medium of $5 billion in charitable giving since 2015. As societies build around the digital nexus of exchange, charitable giving spreads through the convenient and instantaneous flow of information.

Note: Online expansion does not explain the downward trend in donations to religious organizations, which predate widespread internet access and use.

Accessibility goes both ways for secular and religious organizations. Duke University research shows the percentage of church congregations with a website grew from 17% in 1998 to 56% in 2012, to 72% in 2019. Religious organizations opened up to the internet, but as Pew Research shows, religious adherents who build such sites and conduct online outreach campaigns are between 46 and 59 years old, the median age of most congregations. That age bracket is not known for utilizing online platforms in the most productive manner. Religious organizations may be out-campaigned by savvy nonprofit outreach specialists.

Giving USA data indicates secular special interest nonprofit organizations share of total giving increased from 42% in 1980 to 72% in 2020. Charitable giving trended away from religious organizations into a diffuse sector of nonprofits and private foundations. Unfortunately, Giving USA data does not cover the proliferation of the nonprofit sector or the trends therein (in the data I have access to at least).

The IRS Data

The IRS collects and samples Form 990 and 990-EZ to track the total number of active tax exempt organizations and their asset/liability balance sheets. Per IRS records, nonprofit filings increased 39% from 289,777 to 401,607 from 1987 to 2017. Interestingly, nonprofit growth persisted through the initial 2 years of the Great Recession, but dropped 17% from 2009 to 2010, settling at pre-housing bubble numbers.

Note: Here, ”tax exempt organizations” is used interchangeably with “nonprofit.”

Note: The current filing threshold stands at gross receipts greater than $200,000.

Tableau link, Breakdown of 501(c)(3-9) Filings & Assets, 1989-2017

IRS data show private foundation filings tripled from 31,171 to 101,194 between 1985 and 2017. Unlike the other 501(c) organizations, private foundation filings rose between 2009 to 2010. There was a drop in 2011, but it was a negligible 448 filings, a change of 0.5%. It grew again in 2012 and did not stop.

As a whole, growth in the nonprofits was stunned by the Great Recession but not derailed.

Are the IRS records comprehensive?

Private Foundations Filings, 1985-2017.png

Are the IRS records comprehensive?

Unfortunately, due to IRS tax filing requirement thresholds, not all nonprofits are required to file Form 990 or 990- EZ and not all nonprofits file or are accounted for in the data tables. As such, the IRS estimates 25% of total organizations are accounted for in the data, a large gap in information (see IRS reports on Charities and tax-exempt organizations, 1975-2012).

Note: All private foundations are required to file Form 990-PF regardless of gross receipts or total assets on balance. In fact, all tax exempt organizations are required to file at least Form 990-N. Failure to file with the IRS in 3 consecutive years results in the automatic revocation of tax exempt status.

To confirm the accuracy of the above numbers, I cross-referenced other IRS reports, which accounted for 909,224 nonprofits in 2005 and 1,081,891 in 2012. The IRS filing data from 2005 and 2012 accounts for roughly 20% to 35% of the actual nonprofit population. This confirms that the data is reasonably accurate, both as a snapshot of high-performing nonprofits and general nonprofit trends.

The IRS hasn’t issued an in-depth report on the nonprofit sector since 2016. However, the IRS maintains a master file of registered nonprofits current as of August 8, 2022. With this data, we can see if the nonprofit sector continued to grow past post-Great Recession levels. We won’t see year-to-year changes (which we can judge by Form 990, 990-EZ, and 990-PF filings only, here), but this data allows us to separate IRS estimates from the figures as they currently stand.

For the continental US, the dataset is split up into three parts (see table below). Too large for Excel or Google Sheets, I loaded each csv into a Google Cloud bucket and used Google Cloud’s Bigquery SQL Workspace. Then I transformed, cleaned, and analyzed the data.

To focus on national trends, I merged all the tables together using UNION ALL, producing a table of 1,795,106 rows.

(

SELECT *

FROM `donations-data.nonprofits_irs.west` as west

UNION ALL

SELECT *

FROM `donations-data.nonprofits_irs.northeast` as northeast

UNION ALL

SELECT *

FROM `donations-data.nonprofits_irs.midwest_east` as east

)

Per the IRS data extract, nonprofits are categorized broadly into 5 organization types: Corporation, Trust, Cooperative, Partnership, and Association.

SELECT

COUNT(ORGANIZATION) as org_type

FROM `donations-data.national.current`

WHERE

ORGANIZATION = x

Then I wanted to see where each organization type was most concentrated.

SELECT

STATE,

COUNT(STATE) AS num_state

FROM `donations-data.national.current`

WHERE

ORGANIZATION = x

GROUP BY STATE

ORDER BY num_state desc

LIMIT 5

The question follows, is nonprofit registration concentrated in one city within each State?

SELECT

CITY,

STATE,

COUNT(CITY) AS num_city

FROM `donations-data.national.current`

GROUP BY STATE, CITY

ORDER BY num_city desc limit 500

Yes, the top ten cities of registered nonprofits are:

*State rank refers to the total number of nonprofits in each state (or city-state) in descending order.

Unsurprisingly, the cities in the top 10 are in States with a high amount of nonprofits. All of the cities’ States are in the top 10 overall, except Des Moines, IA, which ranks 10th as a city and 22nd as a State. Washington DC is a highly concentrated area for nonprofits as organizations strategically place their headquarters in close proximity to the U.S. Congress.

Tableau link, Top 500 Cities with Nonprofits, Current

How many of the nonprofits possess a favorable deductibility status of deductible? That is, organizations where donations are tax deductible.

SELECT

COUNT(DEDUCTIBILITY) AS tax_benefit

FROM `donations-data.national.current`

WHERE DEDUCTIBILITY = 1

Charitable giving is tax deductible for 1,515,287 nonprofits.

SELECT

COUNT(DEDUCTIBILITY) AS tax_benefit

FROM `donations-data.national.current`

WHERE DEDUCTIBILITY = 2

Charitable giving is not tax deductible for 244,962 nonprofits.

The grand majority maintain a favorable deductibility status.

Wonderfully, this data presents a unique opportunity to look past the IRS filing thresholds. For instance, we can isolate the query for nonprofits with income less than $25,000.

SELECT

STATE,

COUNT(INCOME_CD) as less_than_25k

FROM `donations-data.national.current`

WHERE

INCOME_CD < 3

GROUP BY STATE

ORDER BY less_than_25k desc

Turns out, the vast majority of nonprofits possess total assets of less than $25,000. At the very top of the list, 126,291 of California’s 185,153 nonprofits and 97,853 of Texas’ 134,037 nonprofits possess less than $25,000 in total assets. Further, 113,158 of those 126,291 California nonprofits reported no registered assets. The figure is no less staggering for Texas’ nonprofits as 91,528 reported no registered assets.

I broke down nonprofit asset size by state:

SELECT

STATE,

COUNT(INCOME_CD) AS asset_size

FROM `donations-data.national.current`

GROUP BY INCOME_CD, STATE

ORDER BY STATE

SELECT

STATE,

INCOME_CD,

COUNT(STATE) AS num_state

FROM `donations-data.national.current`

WHERE INCOME_CD = X

GROUP BY STATE, INCOME_CD

ORDER BY num_state desc

Despite inflated figures with $0 asset nonprofits, this inflation is consistent among all States. The hotbeds of nonprofit organizations are California, New York, Texas, Florida, Pennsylvania, and Illinois.

Tableau link, U.S. Nonprofits by State & Total Assets, Current

To compare nonprofit and church populations, I ran the query:

SELECT

INCOME_CD,

COUNT(FOUNDATION) as num_church

FROM `donations-data.national.current`

WHERE

FOUNDATION = 10

GROUP BY FOUNDATION, INCOME_CD

Note: Foundation Code 10 isolates the church population.

The grand majority of registered churches, 250,968 of the 269,904, reported $0 assets. Interestingly, 136 churches report over $50 million in assets. The figure we will keep is 18,936- the number of churches with assets greater than $1.

Unfortunately, we cannot accurately compare the current data with older data. In a similar data set available via the "SOI Tax Stats - Tax-Exempt Organizations and Nonexempt Charitable Trusts - IRS Data Book Table 14," there is no delineation between churches from other similar organizations. Instead, the category tracks "Religious, charitable, and similar organizations" as a conglomerate group. So, best we can do is compare that figure over time.

The data goes as far back as 1991. The category grew from 516,554 in 1991, to 1,280,201 in 2010, to 1,431,266 in 2021. Again, we are limited in what we can assert regarding this data point. If we cannot isolate the data to the foundation and income_id codes (or activity codes), then we can't discern a trend between churches and charities or "similar organizations." Likewise, we cannot discern how many of those organizations have assets- i.e. are active establishments. So, we will have to explore this point in another study.

The absence of 501(c) subcategorization via activity codes (more on this below) in the current dataset limits the ability to verify the consistency between the current and old data. I cannot discern where the discrepancies lie between the accounts. I can neither get more granular with the analysis nor assert, as the comparison shows, that a drop in the number of tax exempt organizations occurred between 2021 and 2022.

Historical table 16 (*Form 990, 990-EZ, 990-N, 990-PF data) is not helpful either, because the nature of that data is exclusionary of non-filing organizations. Additionally, estimates are baked into that data table, which leaves room for error. So, when the data show the number of tax exempt organizations fell from 1,828,187 to 1,795,106 between 2021 and 2022, I cannot definitively affirm it. If that decrease were verifiable, however, the decrease would be an outlier akin to the drop in Form 990* filings from 2009-2010 and Form 990-PF filings from 2011-2012. If that were the case, we'd want to see monthly figures to see if the IRS accounts for organizations on a cyclical or seasonal basis and has not gotten to tabulating all the data yet. The IRS website cautions this possibility, but again I cannot confirm it.

Note: If anyone knows where I can get the full tax exempt organization data, something like what I have for the current numbers, I would greatly appreciate it.

What does the current data tell us?

The last IRS report on “Charities and Other Tax-Exempt Organizations'' was published covering tax year 2012. The report accounted for 1,081,891 tax exempt organizations under section 501(c)(3). As illustrated above, current IRS data indicates there are now 1,795,106 blanketly tax exempt organizations.

The latest IRS data table “SOI Bulletin Historical Table 16,” covers tax years 1985-2018. That data table accounted for 408,767 tax exempt organizations. Despite changes in filing thresholds (see IRS Notes below), current IRS data accounts for 547,625 tax exempt organizations with total assets greater than $25,000. Bridging the gap in data from 2019 to 2021, current data indicate a continued proliferation of secular special interest nonprofit organizations.

On the whole, we discern 1,044,253 organizations that were created after 1995 (and are still active). W know this because of the IRS discontinued activity codes. Organizations created after 1995 have a “0” in the activity code cell. Organizations may come and go, but over a million were created and remain (at least on a balance) active over the last 25 plus years.

Decreased church attendance rates, increased charitable giving exchanged online, and secular nonprofit growth (despite high turnover?) contribute to religious organizations’ shrinking share of charitable giving and secular nonprofits’ expanding share of charitable giving.

IRS historical data table 16 shows the population of certain types of organizations dropped over the 30-year period:

1. Labor and agriculture organizations went from 72,009 to 48,093

2. Social welfare organizations, 142, 473 to 103,958

3. Fraternal beneficiary societies, 98,840 to 42,660

4. Voluntary employees' beneficiary associations, 14,708 to 5,890

I will explore the broader context and any additional data to analyze these developments later studies.

I'm left wondering...

I can do a far more comprehensive macroeconomic study on charity if I had more data.
1. If the data goes back, say, to the 1910s and the beginning of tax deductibility, I’d be able to see how policy impacted charitable giving.
2. How far back do the records go?
  1. How else might we quantify charitable giving?
With more data on the nonprofit population I could identify trends in the private social welfare/net apparatus.
1. With this data, I could compare the state of private welfare before and after the expansion of Federal welfare/grant-distributing programs.
  1. How much of the nonprofit sector is funded almost exclusively by Federal, State, county, municipal government?
    1. Are specific types of nonprofits working with particular levels of government or are they vertically integrated into the grant funding/receiving processes?
How does the expansion of the nonprofit sector correlate with the expansion of the Federal administrative state (especially following the so-called neoliberal policies of Reagan/Bush/Clinton)?
1. How much do nonprofits and independent agencies depend on one another?
  1. Who is accountable if programs or initiatives fail?
    1. How is failure or success quantified/qualified?
To what extent do AGI deductibility limits affect individual giving, especially among the top 50% of AGI earners?
1. What about the latest AGI deductibility limit of 2018?
How much taxes do tax exempt organizations pay via payroll or sales taxes?
Are there DC or other State capital lobbyists on nonprofit boards?
What kinds of secular special interest nonprofits are growing in number over the last 10 to 20 years? Is there a growth in assets commensurate with the increased population (per capita/organization)?
Do nonprofits monopolize a type of grant? or location?
1. Can a nonprofit subsume or undermine competition in a deleterious fashion and create market conditions, in the private welfare sector, akin to those found when a corporate monopolizes an industry, product, or service?
If I had data akin to the current dataset for all available years, I could determine organizational life cycles and turn over rates via EINs (employer identification numbers).
As a matter of demographic comparison to historical precedents: has the demographic of charitable givers changed since the advent of online giving? Has it grown or shrunk a certain demographic's giving tendency?
1. Fewer people are going to places of worship where selfless doctrines are preached as a virtue. Social media platforms do not promote selfless acts as a virtue, but rather as a display. Yet, they can be effective places of promulgation and exchange.
  1. I predict the data will show the youth, the least church going and most actively online demographic, will not donate at the rate of their predecessors.

IRS Data Notes (and limitations)

In 2010, the nonprofit filing threshold for Form 990 and 990-EZ increased from $25,000 to $50,000, limiting the data on updated organizational balances below the threshold.

“On September 9, 2008, the IRS issued temporary Income Tax Regulations, which eliminate the advance ruling process for a section 501(c)(3) organization. Under the new regulations, a new 501(c)(3) organization will be classified as a publicly supported charity, and not a private foundation, if it can show that it reasonably can be expected to be publicly supported when it applies for tax-exempt status.”

“Every organization that qualifies for tax-exempt status under Section 501(c)(3) is classified as a private foundation unless it meets one of the exceptions listed in Section 509(a).”

“Most tax-exempt organizations other than churches and certain church-related organizations are required to file an annual information return or notice with the IRS.”

Every tax exempt organization must file Form 990, 990-EZ, or 990-N at least once every 3 years to avoid automatic revocation of tax exempt status. On paper, every organization is required to file annually, regardless of receipts.

** Notes from the Council on Foundations: “Of the 1,081,891 active organizations recognized by the Internal Revenue Service (IRS) under IRC section 501(c)(3), about 26 percent filed Form 990 or Form 990-EZ returns for Tax Year 2012”

Data References:

Indiana University Lilly Family School of Philanthropy and Giving USA Foundation

(https://givingusa.org) requested by Jason Hooy August 5, 2021, received from Mr. Rob Tate August 6, 2021.

IRS, "Exempt Organizations Business Master File Extract (EO BMF)," current as of August 8, 2022, accessed August 26, 2022. https://www.irs.gov/charities-non-profits/exempt-organizations-business-master-file-extract-eo-bmf

Other References

Chaves, Mark, Joseph Roso, Anna Holleman, and Mary Hawkins. 2021. Congregations in 21st Century America. Durham, NC: Duke University, Department of Sociology. Accessed August 27, 2022: https://sites.duke.edu/ncsweb/

Jones, Jeffrey M., “U.S. Church Membership Falls Below Majority for First Time,” Gallup, March 29, 2021. Accessed August 26, 2022: https://news.gallup.com/poll/341963/church-membership-falls-below-majority-first-time.aspx#:~:text=U.S.%20church%20membership%20was%2073,2010%20and%2047%25%20in%202020.

Lipka, Michael, “Which U.S. religious groups are oldest and youngest?” Pew Research Center, July 11,

2016. Accessed August 27, 2022: https://www.pewresearch.org/fact-tank/2016/07/11/which-u-s-religious-groups-are-oldest-and-youngest/#:~:text=The%20median%20age%20of%20adults%20who%20identify%20with%20the%20Presbyterian,mainline%20United%20Church%20of%20Christ

Singh, Manoj, “The 2007-2008 Financial Crisis in Review,” Investopedia, May 17, 2022. Accessed August 26, 2022: https://www.investopedia.com/articles/economics/09/financial-crisis-review.asp

Staff, “Facebook-Generated Fundraising Hits $5 Billion Since Inception,” The NonProfit Times, March 29, 2021. Accessed August 26, 2022: https://www.thenonprofittimes.com/fundraising/facebook-generated-fundraising-hits-5-billion-since-inception/

Staff, “Social Media Fact Sheet,” Pew Research Center, April 7, 2021. Accessed August 6, 2022: https://www.pewresearch.org/internet/fact-sheet/social-media/

Staff, “ALS Ice Bucket Challenge Commitments,” ALS Association. Accessed August 26, 2022: https://www.als.org/ice-bucket-challenge-spending

Staff, “Giving ReportL 2021,” GoFundMe. Accessed August 26: 2022:https://www.gofundme.com/c/gofundme-giving-report-2021

Final Note

If you have any questions, concerns, or desire to discuss, please reach out to me.

I will be producing many more data-centric reports on the nonprofit sector, so subscribe!

Thank you,

Jason