BURROWING INTO THE GREY MATTER OF THE NONPROFIT UNIVERSE: CHANGING PATTERNS OF IRS REGISTRATION AND STATE INCORPORATION, 2001-20051 Kirsten Grønbjerg, Indiana University Helen Liu, Indiana University Thomas Pollak, Urban Institute Ginger Elliott-Teague, Oklahoma University ABSTRACT This paper is a supplement to Grønbjerg, Liu and Pollak (forthcoming) and details findings summarized in that paper on the overlap and disjunction between two major listings of nonprofit organizations: the Internal Revenue Service national listing of tax-exempt entities (also known as the IRS Business Master File) for a particular jurisdiction and organizations incorporated as nonprofits in that same jurisdiction. Building on previous analyses, we hypothesize that nonprofits that are more likely to be in contact with the general public, have special coordinating responsibilities, are larger, and are older will be more likely to have secured both types of legal status and that their rates of obtaining dual status will increase more over time than their counterparts. Using Indiana as a case study, we use an updated comprehensive database of Indiana nonprofits from 2005 to test these hypotheses by examining the types of nonprofits that are included in both of the two legal status listings compared to those that have secured only one of them. We also examine how these patterns have changed since 2001, when the database was originally developed. We find that rates of incorporation increased among IRS-registered nonprofits, although more for some types of IRSregistered entities than for others, and find support for some of our expectations, but not all. INTRODUCTION Many studies of the nonprofit sector at both a national and a regional level are based on a subset of IRSregistered nonprofits – those registered under sub-section 501c3 (e.g., “charities”) that file IRS Form 990s – for which data are now easily available in a research-friendly format from the National Center for Charitable Statistics. In late 2008, there were about 554,000 of these so-called “charitable filers” and they accounted for only 37 percent of the more than 1.5 million IRS-registered entities included in the Business Master File (see Table 1). However, their combined revenues (more than $1.6 trillion) and assets (about $3.1 trillion) account for more than three-quarters of respectively revenues and assets controlled by all exempt organizations. That is because this group includes hospitals, universities, major museums, and foundations. Focusing on this subset, therefore, is viewed as an efficient strategy for describing the economic dimensions of the nonprofit sector. Note, however, that IRS-registered non-charities, including large nonprofit insurance companies, employee benefit associations, and credit unions, also account substantial revenues and assets and that major religious institutions, such as the Catholic Church, generally do not report financial information to the IRS. 1 This is a revised version of a paper we presented at the Association of Research on Nonprofit Organizations and Voluntary Action, Atlanta, GA, November 15-17, 2007. We thank other panelists and the audience for valuable feedback and suggestions. We are particularly grateful to all research assistants involved in developing the original 2001 database and subsequent updates, especially Richard Clerkin, Svetlana Diomin, Jamie Helsen, Laurie Paarlberg, and Hun Myoung Park. We also gratefully acknowledge funding for the database work on which this paper is based from the Efroymson Fund at the Indianapolis Foundation (a program of the Central Indiana Community Foundation), the Indiana Research Fund at the Center on Philanthropy at Indiana University, and the School of Public and Environmental Affairs at Indiana University Bloomington. 1 As noted by Grønbjerg, Liu and Pollak (forthcoming), another stream of research suggests that unquestioned faith in using charitable filers, or even all IRS-registered nonprofits, as the primary basis for understanding the nonprofit landscape is misplaced. Smith (1997), for example, argued that the "dark matter" of the nonprofit sector may contain millions of organizations that do not register with the IRS, let alone file Form 990. Similarly, Grønbjerg and Paarlberg (2002) found that only 23 percent of some 54,000 nonprofits included in a comprehensive nonprofit database for Indiana in 2001 were both IRS-registered and state incorporated (the “bright matter” of the nonprofit universe). They also found that single-listed nonprofits, those that are either registered with the IRS or incorporated at the state level (the “grey matter” of the universe) are generally smaller and younger than those appearing on both lists. Using a large sample of nonprofits drawn from the same Indiana database, Grønbjerg and Clerkin (2005) conclude that the list of incorporated nonprofits provide the more unbiased estimates of the nonprofit sector, because it includes the majority of churches and also captures newer organizations more effectively than the IRS registration listing. TABLE 1 Number and Size of IRS-Registered Exempt Entities by Charitable Status: 2008 and 2001 Number of Registered Entities by IRS- Subsection 03- Religious, educational, charitable, scientific, literary… entities All Other Total Number of Entities Filing Form 990* 03- Religious, educational, charitable, scientific, literary… entities All Other Total Total Revenues Reported on Form 990* ($ billions) 03- Religious, educational, charitable, scientific, literary… entities All Other Total Dec. 2008 1,071,357 443,464 July 2001 774,916 484,848 1,514,821 1,259,764 Dec. 2008 553,808 248,349 July 2001 346,375 158,749 802,157 505,124 Dec. 2008 $1,610.7 $384.4 July 2001 $920.7 $259.0 $1,995.1 $1,179.8 % change 38% -9% 20% % change 60% 56% 59% % change 75% 48% 69% Assets Reported on Form 990* ($ billions) Dec. 2008 July 2001 % change 03- Religious, educational, charitable, scientific, literary… entities $3,146.1 $1,906.1 65% All Other $980.6 $685.5 43% Total $4,126.7 $2,591.6 59% * Organizations that filed Form 990 (including 990EZ and 990PF) within 24 months of the BMF release date, as reported in NCCS Core Files and IRS Business Master Files SOURCE: Internal Revenue Service, Exempt Organizations Business Master File ( July 2001, December 2008) The Urban Institute, National Center for Charitable Statistics, http://nccsdataweb.urban.org/ ©2009. Retrieved April 9, 2009 However, as Table 1 also shows, much has changed since the analyses summarized above were carried out. By 2008, the number of IRS-registered entities has increased by 20 percent over 2001 and those registered as charities by 38 percent, but non-charities declined by 9 percent. Those filing financial information increased by close to 60 percent, whether charities or not. However, combined revenues and assets increased even more. Revenues grew by 75 percent (not adjusted for inflation) for charities, although only by 48 percent for non-charities, and total assets by 65 percent for charities and 43 percent for noncharities. These developments raise important questions about whether the nonprofit universe itself truly 2 has expanded, or whether it is just that the IRS-registration system now covers a larger segment of the universe than before. There are related questions about whether the patterns identified by Grønbjerg and Paarlberg (2002) about dual-listed (bright matter) nonprofits still pertain. Based on their findings, we hypothesize that nonprofits that are more likely to be in contact with the general public, have special coordinating responsibilities, are larger, and are older will be more likely to have secured both types of legal status and that their rates of obtaining dual status will increase more over time than their counterparts. COMPOSITION OF THE NONPROFIT UNIVERSE We seek to address these questions by comparing the 2005 updated edition of the Indiana nonprofit sector database and its composition with the original one developed in 2001. Our analysis complements the arguments presented in Grønbjerg, Liu and Pollak (forthcoming) which explore in some detail the reasons why nonprofits may not be IRS-registered or state-incorporated. They also report the results of several small surveys of nonprofits found only on the state corporate registration system (the “dark grey matter” of the nonprofit universe, so-called because these organizations are invisible to researchers that consider only IRSregistered nonprofits). They conclude that most of these do not appear to meet IRS registration requirements, but that geographic biases in the IRS listing and delays in processing applications for IRS exempt status are also important, while clear failure to comply with IRS registration is relatively rare. Here we provide more details on the bright and grey matter composition of the nonprofit universe by analyzing the 2005 comprehensive database of Indiana nonprofits, a segment of which (nonprofits incorporated in Indiana, but not registered with the IRS) served as the sampling frame for the survey findings reported in Grønbjerg, Liu and Pollak (forthcoming). We examine characteristics of nonprofits recorded in the database that are found on only one these two listings to those that are on both listings. We also compare these patterns to 2001, when the database was originally developed, in order to examine changes over the 2001-2005 period. The Comprehensive Indiana Nonprofit Database For both the original database and the 2005 update, the Indiana Nonprofit Sector (INS) research team relied primarily on three statewide sources of nonprofit listings (hereafter, 3DB listing): the IRS list of taxexempt entities, the Secretary of State listing of Indiana incorporated nonprofits, and the yellow page listings of congregations in Indiana. For the 2001 effort, the research team also included nonprofits from several supplementary listings: local listings of nonprofits in eleven communities around the state and nonprofits identified by a telephone survey of Indiana residents. (Survey respondents were asked to provide name and address for all nonprofits with which they had direct personal engagement during the prior twelve months as paid employees, volunteers, or participants in meetings or events.) Nonprofits identified exclusively through these two additional efforts in 2001 are excluded from the analysis presented here in order to have comparable information for both years, however they are part of the “dark matter” of the nonprofit universe, since they do not appear on official nonprofit listings. For more information about the 2001 effort, see Grønbjerg (2002). For the 2005 update of the comprehensive database, this involved manipulating the listings of (a) all 32,900 records from the IRS list of registered tax-exempt organizations under section 501(c) with a primary reporting address in Indiana, (b) all 37,200 organizations incorporated as nonprofits with the Indiana Secretary of State (ISOS) that had not formally filed dissolution papers, and (c) all 10,000 churches, congregations, temples, and mosques listed in the yellow pages of phone directories for the state. The team 3 first updated the original database with changes in each of the three listings (e.g., added newly registered or newly incorporated nonprofits – and new congregational listings to the yellow pages – and deleted those no longer on each of the new listings). Next, the team identified duplicates (a difficult and tedious process)2 and copied unique information from secondary duplicates into a primary record for that organization in order to preserve all information contained in the administrative listings. Finally, all secondary duplicates were removed from the overall database. This process was also followed in 2001 when the 3DB listing was originally developed. By 2005, there were 51,100 nonprofits in the state included in the combined and cleaned 3DB listing, down by almost 6 percent from 2001. This reflected a decline in the number of IRS registered nonprofits by 7 percent, while state incorporated nonprofits grew by 15 percent (see Section A in Table 2). As noted above, there were actually 32,900 IRS registered nonprofits with Indiana reporting addresses in the July 2005 BMF, compared to 32,500 in 2001. However, the team removed almost 2,800 IRS-registered entities that had either filed corporate dissolution reports with the Indiana Secretary of State or were designated as inactive on the incorporation listing on the assumption that they have ceased to operate and remained on the IRS listing by oversight. These organizations were also excluded from the 2001 database. Based on this analysis, it does not appear that the overall universe of nonprofits expanded, at least in Indiana. TABLE 2 Three-Database Listings of Indiana Nonprofits, by type of listing and changes, 2001 – 2005 Type of listing A. Summary listings All IRS registered All state incorporated Number of Nonprofits 2005 2001 30,147 33,805 Percent change, 2001-05 Percent Distributions 2005 2001 32,519 29,381 -7.3% 15.1% 59.0% 66.2% 60.1% 54.3% B. Detailed listings Both IRS & Corporation Lists 16,986 12,445 Corporation List Only 16,819 16,936 IRS List Only 13,161 20,074 Neither (Yellow Pages only) 4,099 4,653 ALL - 3DB Listings 51,065 54,108 Source: Indiana Nonprofit Database, 2001 and 2005 36.5% -0.7% -34.4% -11.9% -5.6% 33.3% 32.9% 25.8% 8.0% 100.0% 23.0% 31.3% 37.1% 8.6% 100.0% 2 The listings vary in how frequently and completely they are updated. While incorporation listings are updated annually when corporations submit annual reports and fees to maintain their incorporation status, IRS-registered nonprofits have only since 2008 been required to file annually as well. Also, there are no common identifiers between the two listings to easily identify matches between thm. As a result, researchers must either laboriously look up each incorporated nonprofit against the BMF file available through the IRS or the National Center for Charitable Statistics or identify matches by merging the two listings and then flag those that appear to be duplicates. Identifying true duplicates in merged listings saves time over the “look-up” method, but the researcher has to find matches on the basis of whether organizations have the same (or very similar) names and addresses. Automated matching programs can aid the process, but have significant limits since names may vary slightly (e.g., Michigan City Lions Club vs. The Lions Club Michigan City) or abbreviations may be used on one list but not the other (e.g., PTO vs. P.T.O vs. Parents and Teachers Organization). Street and city names may also vary in format (e.g., FT WAYNE vs. Fort Wayne; Eight Street vs. 8th ST). Finally, organizations may use one address for purposes of reporting to the IRS (perhaps the accountant who files Form 990) and another for purposes of maintaining incorporation status. A full description of the steps used to identify duplicates in the comprehensive Indiana nonprofit database is available by contacting Kirsten Grønbjerg at kgronbj@indiana.edu/. 4 In 2001, the IRS list of exempt entities accounted for the largest share of the full database of any single listing – 60 percent (see Panel A of Table 2). That percentage was virtually unchanged for 2005 (59 percent), suggesting that the IRS-registration system has not expanded its coverage of the nonprofit universe. However, the state incorporation listing did expand to cover a larger segment of the nonprofit universe and by 2005 accounted for the largest share of the entire database – 66 percent of the total, up from 54 percent four years earlier. The combination of these two trends explain why nonprofits found on both the IRS and state incorporation listings made up 33 percent of the 3DB listing in 2005, compared to only 23 percent in 2001(see Panel B of Table 2). Indeed, the number dual-listed nonprofits (the bright matter in the nonprofit universe) increased by 36 percent, while the number of organizations found only on the IRS file (light grey matter) declined by 34 percent, and those found only on the state incorporation listing (dark grey matter) decreased by almost 1 percent. Congregations (part of the “dark matter” of the nonprofit universe not captured by official listings) found only on the yellow pages declined by 12 percent (the overall number of congregations actually increased by 11 percent, but more of them had become incorporated and/or IRS registered). To facilitate our analysis of how the distributions have changed between 2001 and 2005 we use the Index of Dissimilarity (IOD).3 It measures the differences in percent distributions for the two years by subtracting the 2001 percentage values from the corresponding 2005 percentages, summing the absolute values (that is, treating negative values as if they were positive) of all differences, and dividing by 2. In the case of Panel B in Table 2, the index is 5.9 percent, suggesting that almost 6 percent of all nonprofits in 2005 would have to change to a different type of list combination in order to match the 2001 distribution (or vice versa). We use this overall index of 6 percent as the baseline for concluding whether the distributions of bright or grey matter nonprofits changed notably between 2001 and 2005. 4 Detailed Patterns and Changes over Time The differential growth rates in nonprofit listings for the state of Indiana raise important questions about how patterns of dual and single source listings have changed since 2001. We turn therefore to a more systematic comparison of the dual-listed (bright matter) and single-listed (grey matter) segments of the 2001 and 2005 databases, although we are limited by the information available in the various source listings. IRS-Registered Nonprofits by State Incorporation Status The IRS listing has the greatest wealth of information, so we focus first on the 59 percent of Indiana nonprofits (in 2005) that are registered with the IRS to examine differences between those that are both IRS-registered and state incorporated and those that are only registered with the IRS, before comparing the dual listed nonprofits with those that are only state incorporated. We use the difference between the percentages. In 2005, 56 percent of all IRS-registered nonprofits with Indiana reporting addresses were also incorporated in Indiana, up from 38 percent in 2001, or an increase of 18 percentage points. See Shryock, Siegel, and Associates (1976, p. 131) for a description of this index. The index ranges from zero (the percent distribution of nonprofits by type of listing is exactly the same in 2005 as it was in 2001) to 100 percent (any type of list with nonprofits in 2001 had none in 2005 and vice versa). 3 4 Because we analyze something close to the entire universe of Indiana nonprofits, normal statistical tests of significance are not appropriate here since they require assumptions about samples drawn from the universe. theory. 5 If there were no changes in which types of IRS-registered nonprofits were also state incorporated (duallisted, bright matter) between 2001 and 2005, we would expect rates of incorporation to increase by 18 percentage points across the board, regardless of the type or size of nonprofits involved. In the analysis below, we flag (in bold) differences of 23 percentage points or more (arbitrarily adding 5 percentage points to the overall difference of 18 percentage points) as our baseline for determining that rates of incorporation increased more than expected for particular types of nonprofits. Similarly, we flag (in underlined italics) differences of 13 percentage points or less (18 minus 5 percentage points) as indicating that incorporation rates increased less than expected. IRS Sub-Section. As expected, there are notable differences in state incorporation rates among IRS subsections. Ignoring sub-sectors with very few registered entities, column 2 in Table 3 shows that for 2005, the percent incorporated exceeded two-thirds for nonprofits registered under subsections 12 (benevolent lifeinsurance or mutual cooperatives, 97 percent), 02 (title-holding corporations, 81 percent), and 03 (charities, 69 percent). These are nonprofits that hold substantial assets (subsections 12 and 02) or deliver programs and services for the general public (subsection 03). As such they are likely to be subject to greater scrutiny than others and may have found it useful to secure state incorporation in addition to IRS-registration. The rate of incorporation was over 50 percent for subsections 06 (business leagues and chambers, 57 percent), 04 (social welfare or civic nonprofits, 56 percent), 14 (credit unions or mutual corporations/associations, 56 percent), 19 (war veterans associations, 52 percent), and 13 (burial associations or cemeteries, 51 percent), followed by sub-section 07 (pleasure, recreation and social clubs) at 44 percent. Rates of incorporation were much lower for sub-section 08 (fraternal beneficiary associations, 17 percent), 10 (domestic fraternal societies, 13 percent), 05 (labor, horticultural, agricultural groups, 11 percent), 15 (mutual insurance company or association, 8 percent), and 09 (employee beneficiary association, 2 percent). We had expected financial organizations (e.g., credit unions, mutual insurance or employee beneficiary associations) to have high rates of state incorporation, but that is not the case. Instead, they appear to operate under other legal forms. 5 A comparison of columns 1 and 4 in Table 3 shows that charities (subsection code 03) was the only type of IRS-registered exempt entity to increase – up by about 1,100 to 17,600, or almost 6 percent (but still less than the increase of 22 percent nationally in organizations registered under this IRS subsection over the same period). Almost without exception, all other subsections declined. As expected, except for the relatively few organizations registered under subsections 9, 15, and 17 and those where rates of incorporation were already very high (subsection 12) in 2001, the percent of nonprofits in each subsection code that were state incorporated increased from 2001 to 2005 (compare columns 2 and 5).But only for subsection 02 was the increase in excess of 23 percentage points. For other subsectors (all those where only a small minority were incorporated in 2001), the increase was considerably lower and less than 13 percentage points. The Indiana Secretary of State lists the following filing acts under which incorporation may be recorded: (1) Indiana Business Corporation Law, (2) Indiana Professional Corporation Act of 1983, (3), Indiana Nonprofit Corporation Act of 1991, (4) Indiana Business Flexibility Act, (5) Revised Uniform Limited Partnership Act, (6) Uniform Partnership Act, (7) Indiana Agricultural Cooperative Act,(8) Indiana Business Trust Act of 1963, (9) The Indiana Financial Institutions Act, (10) Indiana Insurance Law, and (11) Miscellaneous. See http://www.in.gov/sos/business/newcorps/format.html We obtained a list of those recorded under (3). 5 6 TABLE 3 Percent Incorporated by IRS-Registered Nonprofits in Indiana by Sub-Section Code and State Incorporation Status, 2001 and 2005 2005 IRS Exempt Entities % Not % State State IncorIncorAll porated porated (1) (2) (3) 80.7% 171 19.3% 17,631 68.6% 31.4% 2,992 55.7% 44.3% 11.1% 1,507 88.9% 2001 IRS Exempt Entities % Not % State State IncorIncorAll porated porated (4) (5) (6) 46.8% 203 53.2% 48.4% 16,547 51.6% 65.4% 3,951 34.6% 1,902 5.2% 94.8% IRS Sub-Section Code 02 Title-holding corporation 03 Charitable and religious 04 Social welfare/civic/advocacy 05 Labor, agricultural, horticultural 06 Business league, chamber of 61.7% commerce 1,489 57.2% 42.8% 1,753 38.3% 74.9% 07 Pleasure/recreational/social clubs 1,478 44.0% 56.0% 2,242 25.1% 17.0% 08 Fraternal beneficiary orgs 1,674 83.0% 2,269 10.0% 90.0% 1.7% 09 Employee beneficiary association 353 98.3% 511 1.2% 98.8% 12.6% 10 Domestic fraternal society 1,006 87.4% 1,263 6.7% 93.3% 12 Benevolent life insurance, mutual 96.6% 147 3.4% 168 86.9% 13.1% co-operations 50.8% 13 Burial association, cemetery 421 49.2% 433 44.1% 55.9% 61.8% 14 Credit union, mutual corp./assoc. 52 55.8% 44.2% 68 38.2% 7.7% 15 Mutual insurance comp./assoc. 39 92.3% 35 8.6% 91.4% 0.0% 17 Suppl. uemploymt. compensation 7 100.0% 12 0.0% 100.0% 62.5% 19 War veterans associations 987 51.9% 48.1% 1,174 37.5% 100.0% 25 Title-holding comp. for pensions 1 100.0% 0.0% 1 0.0% 26 State-sponsored high risk health 2 0.0% 100.0% ------insurance ------80 Farmers cooperatives 1 100.0% 0.0% ------90 4947 (a)(2) Split Interest Trust 55 32.7% 67.3% ------92 4947 (a)(1) Private Foundation 53 3.8% 96.2% 93 1381 (a)(2) Taxable Farmers ------80.7% 48 2.1% Cooperative All IRS-registered nonprofits 30,114 56.3% 43.7% 32,532 38.3% 61.7% Source: Indiana Nonprofit Database, 2001 and 2005. Note: Numbers in bold are 23 percentage points or more higher in 2005 than in 2001 and those in underlined italics are no more than 13 percentage points higher in 2005 than in 2001. Because registered charities grew markedly, while other subsections declined, the distribution of IRSexempt entities across subsectors changed some between 2001 and 2005 (compare columns 1 and 4 in Table 4) and the index of dissimilarity is 8.3 percent for the two distributions. However, although there were notable differences in the extent to which rates of incorporation increased across subsectors, the rates increased almost across the board. As a result, the distribution of state incorporated nonprofits across IRS subsection codes was quite similar in 2005 to what it had been in 2001 (compare columns 2 and 5 in Table 4), and the index of dissimilarity between these two distributions is only 3 percent, while it is about 6 percent for those not incorporated (compare columns 3 and 6 in Table 4). Tax-Deductible Contribution Status. There are also notable differences among IRS-registered nonprofits in state incorporation rates by whether they are eligible to receive tax-deductible contributions. Those 7 registered under subsection 03 may receive such contributions, but so can those registered under subsections 08, 10, 13, and 19 if the contributions are designated for charitable purposes. We expect those eligible to receive tax deductible contributions to have higher rates of incorporation because of their broader range of contacts with the general public than nonprofits that must depend on sales and investment income. That is the case. Almost 62 percent of those so eligible to receive tax-deductible contributions were incorporated in 2005, compared to only 44 percent of those not eligible (see Panel A of Table 5). TABLE 4 Percent IRS-Registered Nonprofits in Indiana by Sub-Section Code and State Incorporation Status, 2005 and 2001 2005 IRS Exempt Entities % of % of All % of All All IRS State Not Exempt IncorIncorEntities porated porated (1) (2) (3) 0.6% 0.8% 0.3% 58.5% 71.3% 42.1% 9.9% 9.8% 10.1% 5.0% 1.0% 10.2% IRS Sub-Section Code 02 Title-holding corporation 03 Charitable and religious 04 Social welfare/civic/advocacy 05 Labor, agricultural, horticultural 06 Business league, chamber of commerce 4.9% 5.0% 07 Pleasure/recreational/social clubs 4.9% 3.8% 08 Fraternal beneficiary orgs 5.6% 1.7% 09 Employee beneficiary association 1.2% 0.0% 10 Domestic fraternal society 3.3% 0.7% 12 Benevolent life insurance, mutual co-operations 0.5% 0.8% 13 Burial association, cemetery 1.4% 1.3% 14 Credit union, mutual corp./assoc. 0.2% 0.2% 15 Mutual insurance comp./assoc. 0.1% 0.0% 17 Suppl. uemploymt. compensation 0.0% 0.0% 19 War veterans associations 3.3% 3.0% 25 Title-holding comp. for pensions 0.0% 0.0% 26 State-sponsored high risk health 0.0% 0.0% insurance 80 Farmers cooperatives 0.0% 0.0% 90 4947 (a)(2) Split Interest Trust 0.2% 0.1% 92 4947 (a)(1) Private Foundation 0.2% 0.0% 93 1381 (a)(2) Taxable Farmers 0.2% 0.3% Cooperative All three-database nonprofit 100.0% 100.0% (Number of Entities) (30,114) (16,953) Index of Dissimilarity (2001-2005) 8.3% 3.5% Source: Indiana Nonprofit Database, 2001 and 2005 2001 IRS Exempt Entities % of % of All % of All IRS State Not Exempt IncorIncorEntities porated porated (4) (5) (6) 0.6% 0.9% 0.5% 50.9% 68.5% 39.9% 12.1% 11.0% 12.9% 5.8% 0.8% 9.0% 4.8% 6.3% 10.6% 2.6% 6.7% 5.4% 6.9% 7.0% 1.6% 3.9% 5.4% 4.5% 1.8% 0.0% 0.7% 5.4% 8.4% 10.2% 2.5% 5.9% 0.0% 1.6% 0.2% 0.3% 0.1% 3.6% 0.0% 0.5% 1.3% 0.2% 0.1% 0.0% 3.6% 0.0% 1.2% 1.5% 0.2% ----3.5% 0.0% 0.1% 1.2% 0.2% 0.2% 0.1% 3.7% 0.0% 0.0% 0.0% 0.3% 0.4% ----------- ----------- ----------- 0.0% 100.0% 100.0% 100.0% 100.0% (13,161) (32,532) (12,459) (20,073) 5.8% Because the tax-deductible group includes both charities which grew and several of the sub-sections that declined between 2001 and 2005, the overall change in those eligible to receive tax-deductible contributions was virtually flat (down slightly from 21,478 to 21,408). The percent state incorporated increased by about 17 percentage points for both those eligible to receive tax-deductible contributions and those not so eligible 8 (compare columns 2 and 5 in Panel A of Table 6). As a result, the distribution of state incorporated nonprofits across deductibility category is very similar in 2005 to what it had been in 2001 and the index of dissimilarity is only 2 percent (compare columns 2 and 5 in Panel B in Table 4). There were also only relatively minor changes in the distribution across contribution status between 2001 and 2005 for those not incorporated (compare columns 3 and 6), with an index of dissimilarity between the two distributions of only 4 percent. TABLE 5 Percent Incorporated by IRS Contribution Status (Panel A) and Distribution of Indiana Exempt Entities by Contribution and State Incorporation Status (Panel B), 2005 & 2001 Deductibility of Contributions A. Percent Incorporated Contributions deductible Contributions not deductible Contrib. deductible, foreign treaty All IRS-registered nonprofits 2005 IRS Exempt Entities All % Not Exempt % IncorIncorEntities porated porated (1) (2) (3) 21,408 7,774 3 29,185 % of All IRS Exempt Entities (1) B. Distribution by Deductibility Status Contributions deductible Contributions not deductible Contrib. deductible, foreign treaty All IRS-registered nonprofits Index of Dissimilarity 61.9% 43.8% 66.7% 57.1% % of Those Incorporated (2) 38.1% 56.2% 33.3% 42.9% % of Those Not Incorporated (3) 2001 IRS Exempt Entities All % Not Exempt % IncorIncorEntities porated porated (4) (5) (6) 21,478 10,177 3 31,658 % of All IRS Exempt Entities (4) 55.5% 73.3% 66.7% 61.2% 44.5% 26.7% 33.3% 38.8% % of Those Incorporated (5) % of Those Not Incorporated (6) 73.4% 26.6% 0.0% 100.0% 79.6% 20.4% 0.0% 100.0% 65.1% 34.9% 0.0% 100.0% 67.8% 32.1% 0.0% 100.0% 77.8% 22.1% 0.0% 100.0% 61.5% 38.5% 0.0% 100.0% (29,185) (16,652) (12,530) (31,658) (12,283) (19,375) 5.5% 1.7% 3.6% Source: Indiana Nonprofit Database, 2001 and 2005 Note: For Panel A, numbers in bold are 23 percentage points or more higher in 2005 than in 2001 and those in underlined italics are no more than 13 percentage points higher in 2005 than in 2001. Organizational Types. The application for tax-exempt status (IRS Form 1023 for charities; Form 1024 for all other exempt entities) includes questions about the legal structure of the organization and whether it is affiliated with other organizations. Panel A of Table 6 shows the match between legal structure as reported to the IRS and state incorporation status. As we would expect, and as was the case in database, those listed as corporations in the 2005 IRS file are indeed more likely to be incorporated with the state of Indiana (83 percent vs. 56 percent overall). Presumably, most of the rest are incorporated in other states, operate under other legal structures, or have let their incorporation status lapse since initially registering with the IRS. Similarly, those reported to be associations on the IRS listing are much less likely to be state incorporated (only 23 percent in 2005, 13 percent in 2001, or a difference of only 10 percentage points). However, some are apparently incorporated or have become incorporated after they obtained tax-exempt status. The relatively few nonprofits recorded as trusts also are much more likely not to be incorporated (97 9 percent in 2005, 98 percent in 2001), whereas the even fewer cooperatives are more likely to be incorporated (69 percent in 2005, 58 percent in 2001). TABLE 6 Percent Incorporated by IRS Organization Type (Panel A) and Distribution of Indiana Exempt Entities by Organization Type and State Incorporation Status (Panel B) Type of Organization A. Percent Incorporated by Type Unknown/Omission Error Corporation Trust Co-operative Partnership Association Non-exempt Charitable Trust All IRS-registered nonprofits All Exempt Entities (1) 2005 % State Incorporated (2) 150 17,040 1,108 26 4 11,765 36 30,129 26.0% 82.9% 3.3% 69.2% 75.0% 23.4% 2.8% 56.4% % Not Incorporated (3) All Exempt Entities (4) 2001 % State Incorporated (5) 74.0% 17.1% 96.7% 30.8% 25.0% 76.6% 97.2% 43.6% -16,681 1,549 36 5 14,003 -32,274 -63.0% 2.3% 58.3% 40.0% 13.4% -38.5% % Not Incorporated (6) -37.0% 97.7% 41.7% 60.0% 86.6% -61.5% % of % of % of All % of Those % of All % of Those IRS Those Not IRS Those Not Exempt IncorIncorExempt IncorIncorEntities porated porated Entities porated porated B. Distribution by Type (1) (2) (3) (4) (5) (6) Unknown/Omission Error 0.5% 0.2% 0.8% --Corporation 56.6% 83.1% 22.1% 51.7% 84.6% 31.1% Trust 3.7% 0.2% 8.1% 4.8% 0.3% 7.6% Co-operative 0.1% 0.1% 0.1% 0.1% 0.2% 0.1% Partnership 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Association 39.0% 16.3% 68.5% 43.4% 15.1% 61.1% Non-exempt Charitable Trust 0.1% 0.0% 0.3% 0.0% 0.0% 0.0% All IRS-registered nonprofits 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% (30,129) (16,978) (13,148) (32,274) (19,849) (12,425) Index of Dissimilarity (2001/2005) 5.2% 1.3% 8.5% Source: Indiana Nonprofit Database, 2001 and 2005 Note: For Panel A, numbers in bold are 23 percentage points or more higher in 2005 than in 2001 and those in underlined italics are no more than 13 percentage points higher in 2005 than in 2001. When we compare how IRS-registered nonprofits that are incorporated are distributed across the organizational types in Panel A for 2005 to how they were distributed in 2001 (compare columns 2 and 4 in Panel B in Table 6), we find only very small differences overall – the index of dissimilarity is only 1.3 percent, even thought rates of incorporation did not increase uniformly. By comparison, the 2005 distribution across organizational types for those not incorporated differs by more than 8 percentage points from the 2001 distribution (compare columns 3 and 6 in Panel B in Table 6). Organizational Affiliations. Panel A of Table 7 shows the distribution of Indiana nonprofits by type of organizational affiliation as recorded at the time the organization obtained tax-exempt status. For both 10 years, independent nonprofits are disproportionately incorporated (76 percent in 2005, 57 percent in 2001), as are the relatively few IRS-registered nonprofits that are listed separately but operating as central or intermediary organizations (ranging between 70 percent and 72 percent for 2005 and between 47 percent and 61 percent in 2001). These types of affiliation statuses reflect formal coordinating responsibilities or efforts to establish some form of independence and, as expected, relatively large numbers of them have sought formal incorporation. Moreover, the rates of incorporation for these groups increased by at 15 percentage points or more between 2001 and 2005, and in the case of central, non-church organizations, by close to 23 percent (compare columns 2 and 5 in Panel A in Table 7). The very few central church organizations saw increases of incorporation of only 11 percentage points, up from 61 percent in 2001 to 72 percent in 2005. TABLE 7 Percent Incorporated by Affiliation Status (Panel A) and Distribution of Indiana Exempt Entities by Affiliation Status and State Incorporation Status (Panel B) Affiliation Status A. Percent Incorporated Central, no group exemption Intermediary, no group exemption Independent Central, group exemption, not church Intermediary, group exempt. not church Central, group exemption, church Subordinate, group exemption All IRS-registered nonprofits All Exempt Entities (1) 2005 % State Incorporated (2) % Not Incorporated (3) All Exempt Entities (4) 2001 % State Incorporated (5) 257 38 17,590 113 2 18 12,039 30,057 72.0% 71.1% 76.1% 69.9% 0.0% 72.2% 27.2% 56.4% 28.0% 28.9% 23.9% 30.1% 100.0% 27.8% 72.8% 43.6% 316 42 17,741 115 3 18 14,163 32,398 50.9% 54.8% 56.7% 47.0% 0.0% 61.1% 15.1% 38.4% % Not Incorporated (6) 49.1% 45.2% 43.3% 53.0% 100.0% 38.9% 84.9% 61.6% % of % of % of All % of Those % of All % of Those IRS Those Not IRS Those Not Exempt IncorIncorExempt IncorIncorEntities porated porated Entities porated porated B. Distribution by Affiliation Status (1) (2) (3) (4) (5) (6) Central, no group exemption 0.9% 1.1% 0.5% 1.0% 1.3% 0.8% Intermediary, no group exemption 0.1% 0.2% 0.1% 0.1% 0.2% 0.1% Independent 58.5% 78.9% 32.0% 54.8% 80.9% 38.5% Central, group exemption, not church 0.4% 0.5% 0.3% 0.4% 0.4% 0.3% Intermediary, group exempt. not church 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Central, group exemption, church 0.1% 0.1% 0.0% 0.1% 0.1% 0.0% Subordinate, group exemption 40.1% 19.3% 66.7% 43.7% 17.2% 60.3% All IRS-registered nonprofits 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% (30,057) (13,093) (16,961) (32,398) (19,957) (12,441) Index of Dissimilarity 3.8% 2.1% 6.7% Source: Indiana Nonprofit Database, 2001 and 2005 Note: For Panel A, numbers in bold are 23 percentage points or more higher in 2005 than in 2001 and those in underlined italics are no more than 13 percentage points higher in 2005 than in 2001. 11 By contrast, very few of the many subordinates registered as parts of a group or family of organizations are state incorporated (27 percent in 2005 and15 percent in 2001). As subsidiaries of other nonprofits, they may not have reached the necessary level of development to manage formal incorporation—or may not have the need for it. Moreover, while rates of incorporation increased for this group, the change was only 12 percentage points and thus notably below the 18 percentage points for all IRS-registered entities. When we compare how IRS-registered nonprofits that are also state incorporated are distributed across the various types of affiliation in 2005 to how they were distributed in 2001 (compare columns 2 and 5 in Panel B in Table 7), we find again relatively small differences – about 2 percentage points (Index of Dissimilarity is 2.1). By comparison, the 2005 distribution across affiliation status for those not incorporated differs by almost 7 percentage points from the 2001 distribution. Size of Assets and Income. IRS-registered nonprofits with revenues of $25,000 or more are required to file financial information on Form 990 with the IRS, unless the organization is part of a group exemption in which case Form 990 for the group includes information for all subsidiaries, regardless of the amount of revenues each may have. A similar arrangement holds for those using another nonprofit as a fiscal agent. For the analysis presented below, we used the actual dollar values reported to the IRS corrected for apparent reporting errors,6 rather than the category variables available in the IRS files. 7 In general, we expect larger organizations to be more likely to be incorporated than smaller ones. For the more than 10,000 IRS-registered nonprofits for which information on assets or income is available, about three-fourths were incorporated in 2005, up from just over half in 2001 (see Table 8). As expected, larger organizations (whether defined in terms of assets or income) are more likely to be incorporated than smaller ones. We had expected all large organizations (e.g., assets or revenues in the millions of dollars) to be incorporated, but 15-20 percent are not included on the Indiana nonprofit corporate registry. We think it likely that that they are indeed incorporated, but not in Indiana, or that they operate under other types of legal structures. As Table 8 also shows, the number of organizations with very low assets or revenues declined notably over the five year period, but the percent incorporated was pronounced for those in the smallest assets or revenue category with increases of 30 percentage points or more. For the very largest organizations, rates of incorporation increased only modestly, by 12 percentage point in the case of assets and only 3 percentage points in the case of revenues. Because of differential growth rates by size and because rates of incorporation increased very differently for the smallest and largest organizations, Table 9 shows that the distributions across categories for state incorporated nonprofits in 2005 differ more distinctly from 2001 than for any of the other comparisons reported above – about 15 percentage points for differences in asset distribution and 13 percent for differences in income distributions (the Index of Dissimilarity is respectively 15.2 percent and 13.3 percent). The 2005 distributions across asset and income size categories for those not incorporated from the 2001 distributions also differ notably, by respectively 11 and 14 percentage points. Year of IRS-Ruling. Finally, we have information on the year in which the IRS recognized the exempt 6 In nine cases we found income and asset values with three digits after the decimal places. Based on the types of organizations involved we determined that the amounts were recorded in thousands of dollars and corrected accordingly. In almost all cases, the corrected values were consistent with IRS codes for size of assets or income. 7 We found a large number of organizations with the code 9 ($50 million or more) in the assets size field, but with no dollar values in the asset field itself. We found a similar, but less pervasive problem for the income fields. 12 status of registered entities – the ruling year. This may be the year in which the organization is founded for those organizations that apply for exempt status within 27 months of being established, since the IRS postdates ruling letters for those organizations to the founding date. However, for those that seek exempt status only when they pass the $5,000 revenue level that requires registration or only later discover the need to seek exempt status, the ruling date may be later than the year of establishment. TABLE 8 Percent Incorporated by IRS Assets and Income Categories, Filers Only, 2001 & 2005 2005 IRS Exempt Entities All % State % Not Exempt IncorIncorEntities porated porated Dimension (1) (2) (3) A. Size of assets on most recent From 990 65.2% $1 to $9,999 846 34.8% $10,000 to $24,999 993 66.1% 33.9% $25,000 to $99,999 2,371 68.5% 31.5% $100,000 to $499,999 3,134 75.2% 24.8% $500,000 to $999,999 1,143 74.6% 25.4% $1,000,000 to $4,999,999 1,504 80.1% 19.9% $5,000,000 to $9,999,999 354 79.9% 20.1% $10,000,000 to $49,999,999 391 83.6% 16.4% 84.0% $50,000,000 or more 162 16.0% All IRS-registered Filers 10,898 73.3% 26.7% B. Size of income on most recent Form 990 $1 to $9,999 380 $10,000 to $24,999 158 $25,000 to $99,999 3,845 $100,000 to $499,999 3,632 $500,000 to $999,999 975 $1,000,000 to $4,999,999 1,257 $5,000,000 to $9,999,999 244 $10,000,000 to $49,999,999 309 $50,000,000 or more 106 48.7% 47.5% 68.3% 78.3% 87.8% 77.7% 82.8% 81.6% 78.3% 51.3% 52.5% 31.7% 21.7% 12.2% 22.3% 17.2% 18.4% 21.7% 2001 IRS Exempt Entities % State % Not All Exempt IncorIncorEntities porated porated (4) (5) (6) 1,238 1,134 2,297 3,100 1,030 1,336 316 361 121 10,933 35.1% 44.8% 52.1% 58.7% 61.5% 65.3% 64.9% 62.9% 71.9% 54.7% 64.9% 55.2% 47.9% 41.3% 38.5% 34.7% 35.1% 37.1% 28.1% 45.3% 523 185 4,097 3,629 936 1,128 229 250 72 17.0% 24.3% 49.8% 59.4% 59.6% 64.9% 71.2% 63.6% 75.0% 83.0% 75.7% 50.2% 40.6% 40.4% 35.1% 28.8% 36.4% 25.0% All IRS-registered Filers 10,906 74.3% 25.7% 11,049 54.3% 45.7% Source: Indiana Nonprofit Database, 2001 and 2005 Note: Numbers in bold are 23 percentage points or more higher in 2005 than in 2001 and those in underlined italics are no more than 13 percentage points higher in 2005 than in 2001. Overall, we find there the patterns of state incorporation for IRS-registered nonprofits with Indiana reporting addresses reported by Grønbjerg and Paarlberg (2002) still hold: By 2005, larger nonprofits, those with coordinating responsibilities, eligible to receive tax-deductible contributions, and to some extend older nonprofits were still more likely to be incorporated than their counterparts (see Table 10). However, rates of incorporation increased almost across the board, regardless of which IRS characteristics we examined. As a result, the distribution of incorporated IRS-registered incorporated entities for most of the IRS characteristics examined – sub-section, deductibility of contributions status, organizational type, and affiliation status – is quite similar for the two years with indexes of dissimilarity that range between 1 and 4 percent. Recall that the index measures the percent of all organizations that would need to change 13 characteristics to match the distribution in the other year. The indexes are larger for IRS-registered that are not incorporated for these characteristics, but still relatively modest (ranging between 4 and 8 percent). TABLE 9 Percent IRS-Registered Nonprofit Filers in Indiana by Asset and Size Categories and State Incorporation Status, 2001 and 2005 2005 IRS Exempt Entities % of All % of All IRS % of All Not Exempt IncorIncorEntities porated porated Dimension (1) (2) (3) A. Size of assets on most recent From 990 $1 to $9,999 7.8% 6.9% 10.1% $10,000 to $24,999 9.1% 8.2% 11.6% $25,000 to $99,999 21.8% 20.3% 25.7% $100,000 to $499,999 28.8% 29.5% 26.8% $500,000 to $999,999 10.5% 10.7% 10.0% $1,000,000 to $4,999,999 13.8% 15.1% 10.3% $5,000,000 to $9,999,999 3.2% 3.5% 2.4% $10,000,000 to $49,999,999 3.6% 4.1% 2.2% $50,000,000 or more 1.5% 1.7% 0.9% All IRS-registered Filers 100.0% 100.0% 100.0% (10,898) Number of Filers (7,991 ) ( 2,907) Index of Dissimilarity (2001/2005) 2001 IRS Exempt Entities % of All IRS % of All % of Not Exempt IncorIncorEntities porated porated (4) (5) (6) 11.3% 10.4% 21.0% 28.4% 9.4% 12.2% 2.9% 3.3% 1.1% 100.0% 16.2% 12.6% 22.2% 25.9% 8.0% 9.4% 2.2% 2.7% 0.7% 100.0% 7.3% 8.5% 20.0% 30.4% 10.6% 14.6% 3.4% 3.8% 1.5% 100.0% (10,933 ) ( 5,980) (4,953) 4.6% 15.2% 11.4% B. Size of income on most recent Form 990 $1 to $9,999 3.5% $10,000 to $24,999 1.4% $25,000 to $99,999 35.3% $100,000 to $499,999 33.3% $500,000 to $999,999 8.9% $1,000,000 to $4,999,999 11.5% $5,000,000 to $9,999,999 2.2% $10,000,000 to $49,999,999 2.8% $50,000,000 or more 1.0% All IRS-registered Filers 100.0% 2.3% 0.9% 32.4% 35.1% 10.6% 12.1% 2.5% 3.1% 1.0% 100.0% 6.9% 3.0% 43.4% 28.1% 4.2% 10.0% 1.5% 2.0% 0.8% 100.0% 4.7% 1.7% 37.1% 32.8% 8.5% 10.2% 2.1% 2.3% 0.7% 100.0% 8.6% 2.8% 40.7% 29.2% 7.5% 7.8% 1.3% 1.8% 0.4% 100.0% 1.5% 0.7% 34.0% 35.9% 9.3% 12.2% 2.7% 2.7% 0.9% 100.0% ( 10,906) (8,099 ) ( 2,807) ( 11,049) (6,000) (5,049 ) 2.7% 13.3% 14.3% Number of Filers Index of Dissimilarity (2001/2005) Source: Indiana Nonprofit Database, 2001 and 2005 However, we found notably larger differences between 2001 and 2005 when we examined distributions by size of assets, size of revenues, or year of IRS ruling. For these dimensions, the indexes of dissimilarity for the 2001 and 2005 distributions ranged between 13 and 22 percent for those incorporated and between 11 and 14 percent for those not incorporated, suggesting that patterns of incorporation related to size and age have become more pronounced over the 2001 to 2005 period. 14 TABLE 10 Percent Incorporated by Year of IRS Ruling (Panel A) and Distribution of IRS-Registered Entities by Year of Ruling and State Incorporation (Panel B): Indiana Nonprofits 2001 & 2005 2005 IRS Exempt Entities % State % Not State All Exempt IncorIncorEntities porated porated Dimension (1) (2) (3) A. Percent Incorporated by Year of IRS Ruling 1900-1929 23 60.9% 39.1% 1930-1949 4,113 40.2% 59.8% 1950-1959 2,786 42.6% 57.4% 1960-1969 3,224 45.8% 54.2% 1970-1979 4,345 43.4% 56.6% 1980-1989 4,112 66.9% 33.1% 1990-1999 6,107 77.2% 22.8% 79.9% 2000-2005 4,825 20.1% All IRS-registered 29,535 60.9% 40.6% 2001 IRS Exempt Entities % State % Not State All Exempt IncorIncorEntities porated porated (4) (5) (6) 26 4,811 3,253 4,603 5,998 5,608 7,917 1 32,217 50.0% 26.0% 28.9% 25.1% 30.0% 46.4% 58.7% 0.0% 38.5% 50.0% 74.0% 71.1% 74.9% 70.0% 53.6% 41.3% 100.0% 61.5% % of All % of State % of Not % of All % of State % of Not Exempt IncorState IncorExempt IncorState IncorEntities porated porated Entities porated porated (1) (3) (2) (4) (6) (5) B. Distribution by Year of IRS Ruling 1900-1929 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 1930-1949 13.9% 9.4% 20.5% 14.9% 10.1% 18.0% 1950-1959 9.4% 6.8% 13.3% 10.1% 7.6% 11.7% 1960-1969 10.9% 8.4% 14.6% 14.3% 9.3% 17.4% 1970-1979 14.7% 10.7% 20.5% 18.6% 14.5% 21.2% 1980-1989 13.9% 15.7% 11.4% 17.4% 21.0% 15.2% 1990-1999 20.7% 26.9% 11.6% 24.6% 37.5% 16.5% 2000-2005 16.3% 22.0% 8.1% 0.0% 0.0% 0.0% All IRS-registered 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Index of Dissimilarity 16.3% 22.0% 12.2% Source: Indiana Nonprofit Database, 2001 and 2005 Note: For Panel A, numbers in bold are 23 percentage points or more higher in 2005 than in 2001 and those in underlined italics are no more than 13 percentage points higher in 2005 than in 2001. State-Incorporated Nonprofits by IRS-Registration The state corporate registration system for Indiana contains much less information than the IRS listings, primarily the year of incorporation. Unfortunately, while the application for nonprofit incorporation in Indiana Secretary of State asks responding organizations to indicate whether they are a public benefit corporations organized for public benefit or charitable purpose, a religious corporation organized primarily or exclusively for religious purpose, or a mutual benefit corporation (all other types), this information is not capture in the electronic listings and therefore is not available for analysis. As a result, we can only examine whether nonprofit incorporations that are also IRS-registered differ in when they were incorporated from those which are not IRS-registered, although we also make an effort to look at the types of nonprofits incorporated, by estimating their primary purpose or mission based on the organization’s name. 15 Year of State-Incorporation. As expected, rates of IRS registration are highest for older nonprofits and lowest for those with more recent incorporations. As Panel A in Table 11 shows, more than two-thirds of nonprofits incorporated before 1900 or between 1930 and 1949 were also registered with the IRS by 2005, compared to only 32 percent of those incorporated after 2000. Rates of IRS-registration were intermediary for the remaining incorporation years. The same pattern held in 2001. TABLE 11 Percent IRS- Registered by Year of IRS Ruling (Panel A) and Distribution of State-Incorporated by Year of Incorporation and IRS-Registration (Panel B): Indiana Nonprofits 2001 & 2005 2005 State Incorporated Entities 2001 State Incorporated Entities All State % IRS % Not IRS All State % IRS % Not IRS Incorporated Registered Registered Incorporated Registered Registered Dimension (1) (2) (3) (4) (5) (6) A. Percent of State Incorporated Before 1900 91 70.3% 29.7% 92 60.9% 39.1% 1900-1929 632 63.6% 36.4% 640 57.3% 42.7% 72.8% 39.5% 1930-1949 1,456 27.2% 1,495 60.5% 1950-1959 1,871 62.1% 37.9% 1,918 52.7% 47.3% 1960-1969 2,813 54.7% 45.3% 2,936 45.0% 55.0% 1970-1979 3,597 58.7% 41.3% 3,814 50.4% 49.6% 1980-1989 5,023 57.0% 43.0% 5,384 49.8% 50.2% 52.8% 1990-1999 8,959 47.2% 11,005 37.6% 62.4% 32.4% 2000-2005 9,344 67.6% 2,058 1.0% 99.0% All State incorporated 33,786 50.2% 49.8% 29,342 42.3% 57.7% % of not % of not % of all state % of IRS IRS % of all state % of IRS IRS incorporated registered registered incorporated registered registered (1) (2) (3) (4) (5) (6) B. Distribution by Year Before 1989 0.3% 0.4% 0.2% 0.3% 0.5% 0.2% 1900-1929 1.9% 2.4% 1.4% 2.2% 3.0% 1.6% 1930-1949 4.3% 6.2% 2.4% 5.1% 7.3% 3.5% 1950-1959 5.5% 6.8% 4.2% 6.5% 8.1% 5.4% 1960-1969 8.3% 9.1% 7.6% 10.0% 10.6% 9.5% 1970-1979 10.6% 12.4% 8.8% 13.0% 15.5% 11.2% 1980-1989 14.9% 16.9% 12.8% 18.3% 21.6% 16.0% 1990-1999 26.5% 27.9% 25.1% 37.5% 33.3% 40.6% 2000-2005 27.7% 17.9% 37.5% 7.0% 0.2% 12.0% All state-incorporated 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% (N) (33,786) (16,967 ) ( 16,819) (29,342 ) ( 12,412) ( 16,930) Index of Dissimilarity 10.6% 17.7% 25.5% Source: Indiana “Three Database” Nonprofit Listing, 2001 and 2005. Note: For Panel A, numbers in bold are 12 percentage points or more higher in 2005 than in 2001. In 2005, half of the 33,900 nonprofits incorporated in Indiana were also registered with the IRS, up from 42 percent in 2001 (see Table 2), or an increase of 8 percentage points. As Panel A in Table 11 shows, rates of IRS-registration increased for all incorporation periods by at least six percentage points, but were especially pronounced for those incorporated after 1990 (up by respectively 15 and 31 percentage points). 16 The latter is not surprising, since few of the organizations incorporated after 2000 had had sufficient time to also secure IRS registration by 2001, but by almost one third of all those incorporated after 2000 were also IRS-registered. However, because of the differential rates of IRS registrations, the distribution of incorporated organizations across years of incorporation in 2005 differs markedly from the corresponding distribution in 2001 (Panel B in Table 11). The index of dissimilarity for those that are also IRS-registered is 18 percent, but 26 percent for those that are not IRS-registered. These patterns confirm our previous argument that nonprofits tend to incorporate with state authorities before seeking tax-exempt status with the IRS and that newly established organizations now move more quickly to secure both state incorporation and IRS registration. Most likely the percent IRS registered is actually higher than the 32 percent reported in Table 11 for 2005, since newly established organizations may delay application for exempt status by up to 27 months after founding and it may take the IRS up to a year to review and process applications. As a result, a number of organizations that were in the process of securing IRS registration may not have appeared on the IRS registration the INS research team obtained in 2005 when work to update the database began. NTEE Purpose/Activity Category. Finally, we find notable major differences in the distribution the types of organizations obtaining state incorporation alone (no IRS registration) compared to organizations in the entire IRS registry (see Table 12). Fully 50 percent of the organizations that do not register with the IRS appear to be either congregations (or similar religious organizations) or local community associations (either L50 homeowner or tenant associations or S22 neighborhood and block associations). By comparison, only 18 percent of organizations in the IRS Business Master File are in these categories and only 6 percent of organizations filing IRS Form 990s. Most notably, less than 3 percent of IRS-registered organizations are local community associations, yet nearly 20 percent of the state-only corporations fit this description. Surprisingly perhaps, there is no indication of a broad base of small community arts or grassroots environmental groups incorporating at the state level, nor is there any evidence of non-IRS registered but state-incorporated sports clubs. The state-only corporations do add substantially to the total number of organizations in these latter categories but they do not make the IRS registry look like just the “tip of the iceberg.” <<Table 12 about here — appears on separate page following conclusion>> CONCLUSION The results of our analysis help advance our understanding of the dimensions of the nonprofit sector and how it is evolving over time. Using Indiana as a case study, we have focused on how nonprofits that are both IRS-registered and state-incorporated (the “bright matter” in the nonprofit universe) compare to those that are either IRS-registered or state-incorporated (the “grey matter” in the nonprofit universe). We draw four major conclusions from this analysis. First, the “bright” matter is not evenly distributed across the dimensions we are able to examine. Thus there are notable differences among IRS-registered nonprofits in whether they are also incorporated for virtually all the IRS-related dimensions available: subsection under which the organization is registered with the IRS, eligibility to receive tax-deductible contributions, organizational type, formal affiliation structures, size (assets and revenues), and ruling year. Similarly, there are notable differences in whether incorporated nonprofits are also IRS-registered depending on when they were incorporated and what appears to be the primary mission. Second, the “bright matter” universe is expanding into some of the formerly “grey matter” universe. We find growing overlap between the list of IRS-registered nonprofits and state incorporated nonprofits, at least 17 in Indiana – up from 23 percent in 2001 to 33 percent in 2005. Third, rates of incorporation among IRSregistered nonprofits increased across the board for virtually all the IRS-related dimension we are able to examine as did rates of IRS-registration among incorporated nonprofits across various years of incorporation. Fourth, however, we also showed that for some dimensions, at least, the patterns of incorporation among IRS-registered nonprofits and the patterns of IRS-registration for incorporated nonprofits that we observe for 2005 are notably different from the corresponding patterns in 2001. This is particularly the case for dimensions related to size (assets and revenues as reported to the IRS) and age (IRS-ruling year and state-incorporation year), where smaller and younger organizations appear to be obtaining dual-status more quickly than larger and older organizations, suggesting that these types of organizations are more sensitive to the advantages that IRS-registration and state incorporation bring. Overall, we hope that the analysis we have presented here will assist researchers in making informed decisions about which types of listings to use when examining the dimensions of the nonprofit sector at a local, state or national level. There are major advantages to using the IRS list of exempt entities, but as we have shown, also some important limitations. The same holds for lists of state incorporated nonprofits. 18 TABLE 12 Distribution of Organizations by IRS Registration Status and NTEE Not IRS Not IRS Registered Registered IRS Registered IRS Filers ('05-06) as % of IRS Number Number Number Description of Orgs Percent of Orgs Percent of Orgs Percent Registered Congregations and other religious organizations 5,366 30.8% 5,183 14.8% 411 3.7% 103.5% Homeowner, tenant, neighborhood associations 3,458 936 285 19.9% 2.7% 2.6% 369.4% Education 1,091 6.3% 5,583 16.0% 1,406 12.7% 19.5% Human services, other 883 5.1% 5,002 14.3% 2,373 21.5% 17.7% Arts 639 3.7% 2,192 6.3% 668 6.1% 29.2% Health care 552 3.2% 1,854 5.3% 935 8.5% 29.8% Sports & recreation: Amateur sports 547 3.1% 1,656 4.7% 612 5.5% 33.0% Sports & recreation: Recreation clubs 482 2.8% 704 2.0% 183 1.7% 68.5% Environment & animals* 447 2.6% 826 2.4% 312 2.8% 54.1% Foundations** 331 1.9% 1,262 3.6% 753 6.8% 26.2% Fire, public safety, disaster preparedness & relief 320 1.8% 631 1.8% 282 2.6% 50.7% Cemeteries 209 1.2% 398 1.1% 107 1.0% 52.5% Sports & recreation: Other 199 1.1% 467 1.3% 183 1.7% 42.6% Veterans organizations 102 0.6% 1,394 4.0% 434 3.9% 7.3% Other 954 5.5% 6,746 19.3% 2,090 18.9% 14.1% Unclassified 1,829 10.5% 96 0.3% 3 0.0% 1905.2% Total 17,409 100.0% 34,930 100.0% 11,037 100.0% 49.8% * Environment and animals: Some may be homeowner/community associations, but we left them coded as environmental if names are ambiguous. For example, “The Cobb Hill Association” or “Cobb Creek Association” could be either a community association for residents of the Cobb Hill Development or an environmental group working to protect Cobb Hill from development. ** “Foundations” in the Non-IRS Registered column includes all organizations with “foundation” in their names. 19 WORK CITED AND OTHER RELEVANT LITERATURE Grønbjerg, K.A. & Paarlberg, L. (2002). "Extent and Nature of Overlap Between Two Nonprofit Databases: IRS Tax-Exempt Registrations and Nonprofit Incorporation in Indiana." Voluntary Sector Quarterly 31 (No. 4, December): 565-94. Grønbjerg, K.A. (2002). "Evaluating Nonprofit Databases" by Kirsten A. American Behavioral Scientist 45 (11, July): 1742-78. Grønbjerg, K.A. & Clerkin, R. (2005). "Examining the Landscape of Indiana's Nonprofit Sector: Does What You Know Depend on Where You Look?" Nonprofit and Voluntary Sector Quarterly 34 (No. 2, June): 232-59. Smith, D.H. (1997). The Rest of the Nonprofit Sector: Grassroots Associations as the Dark Matter Ignored in Prevailing "Flat Earth" Maps of the Sector. Nonprofit and Voluntary Sector Quarterly, Vol. 26, No. 2, 114-131. 20