Entry And Exit In The Public Charity Sector In The

Entry and Exit in the Public Charity Sector in the U.S. • Features of the Charity Sector • Motivation & Objective • Dunne et. al (September 2009) • Harrison & Laincz (2008) _________________ • (Spike of 2005) • (Objectives of Charities) 1

Two Main Features of a Public Charity*: 1.

Public Charities are usually engaged in activities that advance the public (vs. public) interest, or general welfare of communities – Arts, human services fields, hospitals, educational institutions, churches, youth, sports, environmental, etc.

1.

Charities, while allowed to earn profits and pay reasonable compensation, are forbidden from distributing their retained earnings to people with lots of influence or control over them – Spend: Improve and expand mission-related/program activities (increase quantity/quality, expand capacity, hire more people, raise salaries) – Save: e.g., accumulation of endowment

2

Costs and Benefits of Being a Public Charity: • Benefits: – Federal income tax exemption – Qualify to receive grants and donations that are tax deductible (for donors) • Costs: – Difficult to find investors (no stocks) – Difficult to borrow – Non-distributional constraint (Top management and Boards are not entitled to retained earnings)

3

# of NPs that Received Tax Exempt Status in Year t 2007 2006 2008 2004

0

(count) number_newNP 10000 20000 30000

40000

Motivation:

2003 2002 2001 2000 1999

2005

1998 1997 1996 1995

1994 1993 1992 1991 1990 1985 1984 1986 1987 1983 1988 1989 1982 1980 1981 1979 1978 1975 1976 1977 1974 1972 1973 1971 1966 1967 1968 1969 1970

1970

1980

1990 year

2000

2010

Source: The IRS Master File Data (last updated September 02, 2009) 4

Average Annual Number of NEW 501(c)(3) Public Charities: Time Period

•

Annual Average

% Change

1900-1959

151

-

1960-1969

789

424.3%

1970-1979

1,806

128.9%

1980-1989

2,903

60.8%

1990-1999

13,043

349.2%

2000-2008

31,475*

141.31%*

Overall annual average

8,361

Source: The IRS Master File Data (last updated September 02, 2009)

5

Data on 501(c)(3) Public Charities in the US: 1995

2000

2005

% % change, change, 199519952005 2005 (inflation adjusted )

Public charitie s, 501(c) (3)

572,66 690,32 876,16 0 6 4

Reportin g public charities

187,03 245,74 310,68 66.1%8 9 3

• Source: Nonprofit Almanac, 2008

53%-

6

Goal: • Want to understand what drives the growth of the public charity sector in the US • Study market structure (n) of the charity sector – depends on both entry (E) and exit (X) decisions of individual charity firms, n’ = n + E – X

• Estimate a dynamic, structural model of entry and exit • Dunne, Klimek, Roberts, and Xu (September 2009) • Pakes, Ostrovsky, and Berry (2007) • Harrison and Laincz (2008)

7

What Might Explain the Trend?

8

2007 2006 2008 2004

0

(count) number_newNP 10000 20000 30000

40000

# of NPs that Received Tax Exempt Status in Year t

2003 2002 2001 2000 1999

2005

1998 1997 1996 1995

1994 1993 1992 1991 1990 1985 1984 1986 1987 1983 1988 1989 1982 1980 1981 1979 1978 1975 1976 1977 1974 1972 1973 1971 1966 1967 1968 1969 1970

1970

1980

1990 year

2000

2010

Source: The IRS Master File Data (last updated September 02, 2009) 9

What Might Explain the Trend: • Economic growth, performance of the stock market • Accumulation of wealth (the impact of tax reforms) – Have more to give (donations are an important source of revenue) – Apart from altruistic considerations, donating to a charity is a financial instrument to lower the amount of taxes owed

• • • •

Availability of government contracts and grants Changes in the private foundation sector Changes in demand for nonprofit services (population growth) Changes in the IRS regulations on public charities

• Other things to consider: – Entry costs – Competition faced from other nonprofit and for-profit firms – Exemption from the federal corporate income tax and often state taxes (property, sales taxes)

10

Exit From the Charity Sector: • The decision process to shut down a charity must be different from the one in the for-profit sector • Upon dissolution, charity’s assets must go to another charity – Assets cannot be sold to a private individual, whether insider or outsider – No hostile takeovers – Mergers do take place • Harrison and Laincz (The B.E. Journal of Economic Analysis & Policy, 2008) – Gross entry rates for NPs are about half the gross entry rates among for-profits in services – The exit rates are only about one-fourth the for-profit rate – Suggest that low exit rates for NPs partly explain the “growth” of the NP sector 11

Dunne, Klimek, Roberts, and Xu “Entry, Exit, and the Determinants of Market Structure” (NBER, 2009): • Estimate a dynamic, structural model of entry and exit for dentists and chiropractors • Build on the work of Pakes, Ostrovsky, and Berry (2007) • Incumbent Firm: {Continue, Exit (E)} • Potential Entrant Firm: {Enter (E); Stay Out} • n’ = n + E – X, π(n)LR _____________________________________________________ • Continuation Rule for Incumbent: VC(s, θ) > fixed cost λ • Entry Rule for Potential Entrant: VE(s, θ) < entry cost κ • Estimate Probabilities of Entry and Exit

12

Dunne et. al (2009). Express Continuation and Entry Values: • Incumbent: VC = MC[π + δVC – δσ(1- px)] • Entrant: VE = ME[π + δVC – δσ(1- px)] – where MC≡MC(s) gives incumbent’s perceived state transition matrix (from s to s´) – ME≡ME(s) gives potential entrant’s perceived state transition matrix – px is the probability of exit by the incumbent – σ is parameter of the fixed cost distribution, Gλ 13

Dunne et. al (2009): • VC, VE, fixed cost λ, entry cost κ => px & pe

p x ( s; λ ,θ ) = Pr(λi > VC ( s,θ )) = 1 − G λ (VC ( s,θ )) κ

p ( s; κ , λ ,θ ) = Pr(κ i < δVE ( s,θ )) = G (δVE ( s,θ )) e

• To calculate the entry and exit probabilities/flows, px and pe, need to estimate VC(s, θ), VE(s, θ) and parameters of distributions of fixed costs and entry costs, Gλ and Gκ 14

Dunne et. al (2009). Implementation (1): •

Incumbent: VC = MC[π + δVC – δσ(1- px)]

•

Entrant: VE = ME[π + δVC – δσ(1- px)]

1.

•

Estimate parameters of the profit function in market m, year t: πmt = F(nmt ,zmt ) + fm + εmt – –

Measure of profits, πmt ?

–

Alternative objective function? α(output) + β(πmt )

• Lakdawalla and Philipson (2005), Harrison and Laincz (2008) Definition of market m? – Estimate transition matrices MC and ME

15

Dunne et. al (2009). Implementation (2): 3.

Can construct VC(s, θ) and VE(s, θ) as functions of estimated πmt, MC, ME and parameters α and σ of cdfs for the fixed and entry cost distribution, Gλ(σ) and Gκ(α)

4.

Given constructed VC and VE, construct exit and p x ( s;entry λ ,θ ) = probabilities Pr(λ > VC ( s,θ )) = 1 − G λ (VC ( s,θ )) i

p e ( s; κ , λ ,θ ) = Pr(κ i < δVE ( s,θ )) = G κ (δVE ( s,θ )) 16

Dunne et. al (2009). Implementation (3): 5. Construct the probability of observing a market with Xmt exits and Emt entrants: P(Xmt , Emt ; α, σ) = (pe)(E) (1-pe)(A-E) px(X) Pr(1-px)(n-X) = Pr(E enters)(E) Pr(E out)(A-E) Pr(I exits)(X) Pr(I continues)(n-X) 5.

Then obtain an estimate of (α, σ):

(α , σ ) = arg max ∑ ∑ P( X mt , Emt ;α , σ ) m

t

17

Dunne et. al (2009). Main Conclusions: • Entry costs faced by potential entrants and fixed costs faced by incumbents matter for πLR and market structure (n) • As the number of firms in the market increases, the value of continuing (VC) and the value of entering (VE) both decline, the probability of exit (px) rises, and the probability of entry declines (pe ) 18

Harrison and Laincz “Entry and Exit in the Nonprofit Sector” (B.E. JEAP, 2008): • Study entry and exit decisions of public charities (1989-2003) • Incorporate Lakdawalla and Philipson’s (2006) model into Jovanovic’s (1982) dynamic model • L&P (2006) develop a model where nonprofit firms can be analyzed as for-profit firms with lower costs (in the context of mixed industries only)

• Nonprofit entry rates are lower than those of FPs in services. But exit rates are extremely low => Net entry rates nearly 3 times larger than that of FPs • Entry patterns are similar (new firms are smaller, grow faster) • Exit patterns are different for NPs and FPs (low exit rates, survival rate of a new nonprofit firm is higher)

19

If Have More Time:

20

Back to Data and the Spike in Year 2005

21

2007 2006 2008 2004

0

(count) number_newNP 10000 20000 30000

40000

# of NPs that Received Tax Exempt Status in Year t

2003 2002 2001 2000 1999

2005

1998 1997 1996 1995

1994 1993 1992 1991 1990 1985 1984 1986 1987 1983 1988 1989 1982 1980 1981 1979 1978 1975 1976 1977 1974 1972 1973 1971 1966 1967 1968 1969 1970

1970

1980

1990 year

2000

2010

Source: The IRS Master File Data (last updated September 02, 2009) 22

Where Does Pressure for Charities Come From?

• Litigation exposure in the nonprofit context is almost nonexistent – No one to bring lawsuits against an NP whose management and/or the board engaged in excessive benefit transactions • No threat of a hostile takeover • The IRS does not have good mechanisms to detect selfdealings among nonprofits • Competition for donations and particularly the media seem to be the only effective mechanisms that discipline charities/nonprofits

• In this light, even small changes in the disclosure requirements by the IRS deserve attention

23

Drop in the Number of New Charities in 2005: • Nonprofit => Form 1023 => IRS => 501(c)(3) Public Charity – Federal tax exemption – Qualify to receive tax deductible donations

• In October 2004 the IRS made several important revisions to Form 1023 – Introduced higher disclosure requirements (on conflicts of interest, CEO compensation, etc.) – Average estimated time to prepare the form went up by more than 50% (from 69.5 hours to 105 hours) – Encourage development of governance policies – Raise awareness

24

Questions: • Higher entry cost? (small vs. large applicants) • Does the October 2004 version of Form 1023 deter entry by poorly run & less responsible organizations? • Does it help improve the average quality of reporting/governance/record-keeping? • Was the 2005 drop caused by the IRS itself? Or was it due to the response by prospective applicants? • What was the impact of this drop in the number of new charities on the charity sector? If any.

25

If Have More Time:

26

Fundamental Questions: 1. Why do charitable organizations exit? • along with for-profit firms and government-run organizations? 1. What is the objective function of a charity? • •

Mimic non-profit firms? How altruistic are they? If charities do not maximize profits, what do they want?

1. Incentives of CEOs and Boards? • •

Not driven by profits (?) No good monitoring/enforcement mechanism 27

Modeling Behavior of NP/Charity Firms (1): • Lakdawalla and Philipson (2006) • Develop a model where nonprofit firms can be analyzed as for-profit firms with lower costs (in the context of mixed industries only) • Glaeser and Shleifer (JPE, 2001) • Entrepreneurs choose not-for-profit status as a means of credibly committing to non-collection of profits at a later date • The risk of diversion of funds is smaller, so donors prefer non-profit firms even if there are no tax advantages • Weisbrod (1998) • NP’s objective function is related to its mission. Since missions differ, so will the objective functions • NPs can be modeled as “individuals” with utilities since they represent strong preferences of managers

28

Modeling Behavior of NP/Charity Firms (2): • Newhouse (AER, 1970) models a nonprofit hospital as maximizing quantity, given a certain level of service quality (quality-quantity tradeoff). • Pauly and Redisch (AER, 1973) develop a model where doctors are the sole decision-makers in a hospital, and each doctor shares equally in the revenue generated by this doctor-controlled hospital. A hospital is modeled not as maximizing the quantity of services delivered, but rather as a collective (or a cooperative) that maximizes each doctor’s share of the net revenue. • One other suggestion made in Newhouse (1970), Pauly and Redisch (1973), and a few other papers is that the nonprofit status acts as a barrier to entry for for-profits firms. 29